Remove comment

Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
Co-authored-by: Jamison Lahman <jamison@lahman.dev>

.git-blame-ignore-revs

@@ -0,0 +1,8 @@
+# Exclude these commits from git blame (e.g. mass reformatting).
+# These are ignored by GitHub automatically.
+# To enable this locally, run:
+#
+#    git config blame.ignoreRevsFile .git-blame-ignore-revs
+
+3134e5f840c12c8f32613ce520101a047c89dcc2  # refactor(whitespace): rm temporary react fragments (#7161)
+ed3f72bc75f3e3a9ae9e4d8cd38278f9c97e78b4  # refactor(whitespace): rm react fragment #7190

.github/CODEOWNERS

@@ -1,3 +1,10 @@
 * @onyx-dot-app/onyx-core-team
 # Helm charts Owners
 /helm/ @justin-tahara
+
+# Web standards updates
+/web/STANDARDS.md @raunakab @Weves
+
+# Agent context files
+/CLAUDE.md.template @Weves
+/AGENTS.md.template @Weves

.github/actionlint.yml

@@ -0,0 +1,43 @@
+self-hosted-runner:
+  # Labels of self-hosted runner in array of strings.
+  labels:
+    - extras=ecr-cache
+    - extras=s3-cache
+    - hdd=256
+    - runs-on
+    - runner=1cpu-linux-arm64
+    - runner=1cpu-linux-x64
+    - runner=2cpu-linux-arm64
+    - runner=2cpu-linux-x64
+    - runner=4cpu-linux-arm64
+    - runner=4cpu-linux-x64
+    - runner=8cpu-linux-arm64
+    - runner=8cpu-linux-x64
+    - runner=16cpu-linux-arm64
+    - runner=16cpu-linux-x64
+    - ubuntu-slim # Currently in public preview
+    - volume=40gb
+    - volume=50gb
+
+# Configuration variables in array of strings defined in your repository or
+# organization. `null` means disabling configuration variables check.
+# Empty array means no configuration variable is allowed.
+config-variables: null
+
+# Configuration for file paths. The keys are glob patterns to match to file
+# paths relative to the repository root. The values are the configurations for
+# the file paths. Note that the path separator is always '/'.
+# The following configurations are available.
+#
+# "ignore" is an array of regular expression patterns. Matched error messages
+# are ignored. This is similar to the "-ignore" command line option.
+paths:
+  # Glob pattern relative to the repository root for matching files. The path separator is always '/'.
+  # This example configures any YAML file under the '.github/workflows/' directory.
+  .github/workflows/**/*.{yml,yaml}:
+    # TODO: These are real and should be fixed eventually.
+    ignore:
+      - 'shellcheck reported issue in this script: SC2038:.+'
+      - 'shellcheck reported issue in this script: SC2046:.+'
+      - 'shellcheck reported issue in this script: SC2086:.+'
+      - 'shellcheck reported issue in this script: SC2193:.+'

.github/actions/custom-build-and-push/action.yml

@@ -1,135 +0,0 @@
-name: 'Build and Push Docker Image with Retry'
-description: 'Attempts to build and push a Docker image, with a retry on failure'
-inputs:
-  context:
-    description: 'Build context'
-    required: true
-  file:
-    description: 'Dockerfile location'
-    required: true
-  platforms:
-    description: 'Target platforms'
-    required: true
-  pull:
-    description: 'Always attempt to pull a newer version of the image'
-    required: false
-    default: 'true'
-  push:
-    description: 'Push the image to registry'
-    required: false
-    default: 'true'
-  load:
-    description: 'Load the image into Docker daemon'
-    required: false
-    default: 'true'
-  tags:
-    description: 'Image tags'
-    required: true
-  no-cache:
-    description: 'Read from cache'
-    required: false
-    default: 'false'
-  cache-from:
-    description: 'Cache sources'
-    required: false
-  cache-to:
-    description: 'Cache destinations'
-    required: false
-  outputs:
-    description: 'Output destinations'
-    required: false
-  provenance:
-    description: 'Generate provenance attestation'
-    required: false
-    default: 'false'
-  build-args:
-    description: 'Build arguments'
-    required: false
-  retry-wait-time:
-    description: 'Time to wait before attempt 2 in seconds'
-    required: false
-    default: '60'
-  retry-wait-time-2:
-    description: 'Time to wait before attempt 3 in seconds'
-    required: false
-    default: '120'
-
-runs:
-  using: "composite"
-  steps:
-    - name: Build and push Docker image (Attempt 1 of 3)
-      id: buildx1
-      uses: docker/build-push-action@v6
-      continue-on-error: true
-      with:
-        context: ${{ inputs.context }}
-        file: ${{ inputs.file }}
-        platforms: ${{ inputs.platforms }}
-        pull: ${{ inputs.pull }}
-        push: ${{ inputs.push }}
-        load: ${{ inputs.load }}
-        tags: ${{ inputs.tags }}
-        no-cache: ${{ inputs.no-cache }}
-        cache-from: ${{ inputs.cache-from }}
-        cache-to: ${{ inputs.cache-to }}
-        outputs: ${{ inputs.outputs }}
-        provenance: ${{ inputs.provenance }}
-        build-args: ${{ inputs.build-args }}
-
-    - name: Wait before attempt 2
-      if: steps.buildx1.outcome != 'success'
-      run: |
-        echo "First attempt failed. Waiting ${{ inputs.retry-wait-time }} seconds before retry..."
-        sleep ${{ inputs.retry-wait-time }}
-      shell: bash
-
-    - name: Build and push Docker image (Attempt 2 of 3)
-      id: buildx2
-      if: steps.buildx1.outcome != 'success'
-      uses: docker/build-push-action@v6
-      with:
-        context: ${{ inputs.context }}
-        file: ${{ inputs.file }}
-        platforms: ${{ inputs.platforms }}
-        pull: ${{ inputs.pull }}
-        push: ${{ inputs.push }}
-        load: ${{ inputs.load }}
-        tags: ${{ inputs.tags }}
-        no-cache: ${{ inputs.no-cache }}
-        cache-from: ${{ inputs.cache-from }}
-        cache-to: ${{ inputs.cache-to }}
-        outputs: ${{ inputs.outputs }}
-        provenance: ${{ inputs.provenance }}
-        build-args: ${{ inputs.build-args }}
-
-    - name: Wait before attempt 3
-      if: steps.buildx1.outcome != 'success' && steps.buildx2.outcome != 'success'
-      run: |
-        echo "Second attempt failed. Waiting ${{ inputs.retry-wait-time-2 }} seconds before retry..."
-        sleep ${{ inputs.retry-wait-time-2 }}
-      shell: bash
-
-    - name: Build and push Docker image (Attempt 3 of 3)
-      id: buildx3
-      if: steps.buildx1.outcome != 'success' && steps.buildx2.outcome != 'success'
-      uses: docker/build-push-action@v6
-      with:
-        context: ${{ inputs.context }}
-        file: ${{ inputs.file }}
-        platforms: ${{ inputs.platforms }}
-        pull: ${{ inputs.pull }}
-        push: ${{ inputs.push }}
-        load: ${{ inputs.load }}
-        tags: ${{ inputs.tags }}
-        no-cache: ${{ inputs.no-cache }}
-        cache-from: ${{ inputs.cache-from }}
-        cache-to: ${{ inputs.cache-to }}
-        outputs: ${{ inputs.outputs }}
-        provenance: ${{ inputs.provenance }}
-        build-args: ${{ inputs.build-args }}
-
-    - name: Report failure
-      if: steps.buildx1.outcome != 'success' && steps.buildx2.outcome != 'success' && steps.buildx3.outcome != 'success'
-      run: |
-        echo "All attempts failed. Possible transient infrastucture issues? Try again later or inspect logs for details."
-      shell: bash

.github/actions/setup-playwright/action.yml

@@ -0,0 +1,17 @@
+name: "Setup Playwright"
+description: "Sets up Playwright and system deps (assumes Python and Playwright are installed)"
+runs:
+  using: "composite"
+  steps:
+    - name: Cache playwright cache
+      uses: runs-on/cache@50350ad4242587b6c8c2baa2e740b1bc11285ff4 # ratchet:runs-on/cache@v4
+      with:
+        path: ~/.cache/ms-playwright
+        key: ${{ runner.os }}-${{ runner.arch }}-playwright-${{ hashFiles('backend/requirements/default.txt') }}
+        restore-keys: |
+          ${{ runner.os }}-${{ runner.arch }}-playwright-
+
+    - name: Install playwright
+      shell: bash
+      run: |
+        playwright install chromium --with-deps

.github/actions/setup-python-and-install-dependencies/action.yml

@@ -0,0 +1,76 @@
+name: "Setup Python and Install Dependencies"
+description: "Sets up Python with uv and installs deps"
+inputs:
+  requirements:
+    description: "Newline-separated list of requirement files to install (relative to repo root)"
+    required: true
+runs:
+  using: "composite"
+  steps:
+    - name: Compute requirements hash
+      id: req-hash
+      shell: bash
+      env:
+        REQUIREMENTS: ${{ inputs.requirements }}
+      run: |
+        # Hash the contents of the specified requirement files
+        hash=""
+        while IFS= read -r req; do
+          if [ -n "$req" ] && [ -f "$req" ]; then
+            hash="$hash$(sha256sum "$req")"
+          fi
+        done <<< "$REQUIREMENTS"
+        echo "hash=$(echo "$hash" | sha256sum | cut -d' ' -f1)" >> "$GITHUB_OUTPUT"
+
+    # NOTE: This comes before Setup uv since clean-ups run in reverse chronological order
+    # such that Setup uv's prune-cache is able to prune the cache before we upload.
+    - name: Cache uv cache directory
+      uses: runs-on/cache@50350ad4242587b6c8c2baa2e740b1bc11285ff4 # ratchet:runs-on/cache@v4
+      with:
+        path: ~/.cache/uv
+        key: ${{ runner.os }}-uv-${{ steps.req-hash.outputs.hash }}
+        restore-keys: |
+          ${{ runner.os }}-uv-
+
+    - name: Setup uv
+      uses: astral-sh/setup-uv@ed21f2f24f8dd64503750218de024bcf64c7250a # ratchet:astral-sh/setup-uv@v7
+      with:
+        version: "0.9.9"
+      # TODO: Enable caching once there is a uv.lock file checked in.
+      # with:
+      #   enable-cache: true
+
+    - name: Setup Python
+      uses: actions/setup-python@a26af69be951a213d495a4c3e4e4022e16d87065 # ratchet:actions/setup-python@v5
+      with:
+        python-version: "3.11"
+
+    - name: Create virtual environment
+      shell: bash
+      env:
+        VENV_DIR: ${{ runner.temp }}/venv
+      run: | # zizmor: ignore[github-env]
+        uv venv "$VENV_DIR"
+        # Validate path before adding to GITHUB_PATH to prevent code injection
+        if [ -d "$VENV_DIR/bin" ]; then
+          realpath "$VENV_DIR/bin" >> "$GITHUB_PATH"
+        else
+          echo "Error: $VENV_DIR/bin does not exist"
+          exit 1
+        fi
+
+    - name: Install Python dependencies with uv
+      shell: bash
+      env:
+        REQUIREMENTS: ${{ inputs.requirements }}
+      run: |
+        # Build the uv pip install command with each requirement file as array elements
+        cmd=("uv" "pip" "install")
+        while IFS= read -r req; do
+          # Skip empty lines
+          if [ -n "$req" ]; then
+            cmd+=("-r" "$req")
+          fi
+        done <<< "$REQUIREMENTS"
+        echo "Running: ${cmd[*]}"
+        "${cmd[@]}"

.github/actions/slack-notify/action.yml

@@ -0,0 +1,102 @@
+name: "Slack Notify on Failure"
+description: "Sends a Slack notification when a workflow fails"
+inputs:
+  webhook-url:
+    description: "Slack webhook URL (can also use SLACK_WEBHOOK_URL env var)"
+    required: false
+  failed-jobs:
+    description: "List of failed job names (newline-separated)"
+    required: false
+  title:
+    description: "Title for the notification"
+    required: false
+    default: "🚨 Workflow Failed"
+  ref-name:
+    description: "Git ref name (tag/branch)"
+    required: false
+runs:
+  using: "composite"
+  steps:
+    - name: Send Slack notification
+      shell: bash
+      env:
+        SLACK_WEBHOOK_URL: ${{ inputs.webhook-url }}
+        FAILED_JOBS: ${{ inputs.failed-jobs }}
+        TITLE: ${{ inputs.title }}
+        REF_NAME: ${{ inputs.ref-name }}
+        REPO: ${{ github.repository }}
+        WORKFLOW: ${{ github.workflow }}
+        RUN_NUMBER: ${{ github.run_number }}
+        RUN_ID: ${{ github.run_id }}
+        SERVER_URL: ${{ github.server_url }}
+        GITHUB_REF_NAME: ${{ github.ref_name }}
+      run: |
+        if [ -z "$SLACK_WEBHOOK_URL" ]; then
+          echo "webhook-url input or SLACK_WEBHOOK_URL env var is not set, skipping notification"
+          exit 0
+        fi
+
+        # Build workflow URL
+        WORKFLOW_URL="${SERVER_URL}/${REPO}/actions/runs/${RUN_ID}"
+
+        # Use ref_name from input or fall back to github.ref_name
+        if [ -z "$REF_NAME" ]; then
+          REF_NAME="$GITHUB_REF_NAME"
+        fi
+
+        # Escape JSON special characters
+        escape_json() {
+          local input="$1"
+          # Escape backslashes first (but preserve \n sequences)
+          # Protect \n sequences temporarily
+          input=$(printf '%s' "$input" | sed 's/\\n/\x01NL\x01/g')
+          # Escape remaining backslashes
+          input=$(printf '%s' "$input" | sed 's/\\/\\\\/g')
+          # Restore \n sequences (single backslash, will be correct in JSON)
+          input=$(printf '%s' "$input" | sed 's/\x01NL\x01/\\n/g')
+          # Escape quotes
+          printf '%s' "$input" | sed 's/"/\\"/g'
+        }
+
+        REF_NAME_ESC=$(escape_json "$REF_NAME")
+        FAILED_JOBS_ESC=$(escape_json "$FAILED_JOBS")
+        WORKFLOW_URL_ESC=$(escape_json "$WORKFLOW_URL")
+        TITLE_ESC=$(escape_json "$TITLE")
+
+        # Build JSON payload piece by piece
+        # Note: FAILED_JOBS_ESC already contains \n sequences that should remain as \n in JSON
+        PAYLOAD="{"
+        PAYLOAD="${PAYLOAD}\"text\":\"${TITLE_ESC}\","
+        PAYLOAD="${PAYLOAD}\"blocks\":[{"
+        PAYLOAD="${PAYLOAD}\"type\":\"header\","
+        PAYLOAD="${PAYLOAD}\"text\":{\"type\":\"plain_text\",\"text\":\"${TITLE_ESC}\"}"
+        PAYLOAD="${PAYLOAD}},{"
+        PAYLOAD="${PAYLOAD}\"type\":\"section\","
+        PAYLOAD="${PAYLOAD}\"fields\":["
+        if [ -n "$REF_NAME" ]; then
+          PAYLOAD="${PAYLOAD}{\"type\":\"mrkdwn\",\"text\":\"*Ref:*\\n${REF_NAME_ESC}\"},"
+        fi
+        PAYLOAD="${PAYLOAD}{\"type\":\"mrkdwn\",\"text\":\"*Run ID:*\\n#${RUN_NUMBER}\"}"
+        PAYLOAD="${PAYLOAD}]"
+        PAYLOAD="${PAYLOAD}}"
+        if [ -n "$FAILED_JOBS" ]; then
+          PAYLOAD="${PAYLOAD},{"
+          PAYLOAD="${PAYLOAD}\"type\":\"section\","
+          PAYLOAD="${PAYLOAD}\"text\":{\"type\":\"mrkdwn\",\"text\":\"*Failed Jobs:*\\n${FAILED_JOBS_ESC}\"}"
+          PAYLOAD="${PAYLOAD}}"
+        fi
+        PAYLOAD="${PAYLOAD},{"
+        PAYLOAD="${PAYLOAD}\"type\":\"actions\","
+        PAYLOAD="${PAYLOAD}\"elements\":[{"
+        PAYLOAD="${PAYLOAD}\"type\":\"button\","
+        PAYLOAD="${PAYLOAD}\"text\":{\"type\":\"plain_text\",\"text\":\"View Workflow Run\"},"
+        PAYLOAD="${PAYLOAD}\"url\":\"${WORKFLOW_URL_ESC}\""
+        PAYLOAD="${PAYLOAD}}]"
+        PAYLOAD="${PAYLOAD}}"
+        PAYLOAD="${PAYLOAD}]"
+        PAYLOAD="${PAYLOAD}}"
+
+        curl -X POST -H 'Content-type: application/json' \
+          --data "$PAYLOAD" \
+          "$SLACK_WEBHOOK_URL"
+

.github/dependabot.yml

@@ -0,0 +1,24 @@
+version: 2
+updates:
+  - package-ecosystem: "github-actions"
+    directory: "/"
+    schedule:
+      interval: "weekly"
+    cooldown:
+      default-days: 7
+    open-pull-requests-limit: 3
+    assignees:
+      - "jmelahman"
+    labels:
+      - "dependabot:actions"
+  - package-ecosystem: "pip"
+    directory: "/backend"
+    schedule:
+      interval: "weekly"
+    cooldown:
+      default-days: 7
+    open-pull-requests-limit: 3
+    assignees:
+      - "jmelahman"
+    labels:
+      - "dependabot:python"

.github/pull_request_template.md

@@ -1,10 +1,10 @@
 ## Description
 
-[Provide a brief description of the changes in this PR]
+<!--- Provide a brief description of the changes in this PR --->
 
 ## How Has This Been Tested?
 
-[Describe the tests you ran to verify your changes]
+<!--- Describe the tests you ran to verify your changes --->
 
 ## Additional Options
 

.github/runs-on.yml

@@ -0,0 +1 @@
+_extend: .github-private

.github/workflows/check-lazy-imports.yml

@@ -1,24 +0,0 @@
-name: Check Lazy Imports
-
-on:
-  merge_group:
-  pull_request:
-    branches:
-      - main
-      - 'release/**'
-
-jobs:
-  check-lazy-imports:
-    runs-on: ubuntu-latest
-
-    steps:
-    - name: Checkout code
-      uses: actions/checkout@v4
-
-    - name: Set up Python
-      uses: actions/setup-python@v4
-      with:
-        python-version: '3.11'
-
-    - name: Check lazy imports
-      run: python3 backend/scripts/check_lazy_imports.py

.github/workflows/docker-build-push-backend-container-on-tag.yml

@@ -1,166 +0,0 @@
-name: Build and Push Backend Image on Tag
-
-on:
-  push:
-    tags:
-      - "*"
-
-env:
-  REGISTRY_IMAGE: ${{ contains(github.ref_name, 'cloud') && 'onyxdotapp/onyx-backend-cloud' || 'onyxdotapp/onyx-backend' }}
-  DEPLOYMENT: ${{ contains(github.ref_name, 'cloud') && 'cloud' || 'standalone' }}
-  
-  # don't tag cloud images with "latest"
-  LATEST_TAG: ${{ contains(github.ref_name, 'latest') && !contains(github.ref_name, 'cloud') }}
-
-jobs:
-  build-and-push:
-    # TODO: investigate a matrix build like the web container
-    # See https://runs-on.com/runners/linux/
-    runs-on:
-      - runs-on
-      - runner=${{ matrix.platform == 'linux/amd64' && '8cpu-linux-x64' || '8cpu-linux-arm64' }}
-      - run-id=${{ github.run_id }}
-      - tag=platform-${{ matrix.platform }}
-    strategy:
-      fail-fast: false
-      matrix:
-        platform:
-          - linux/amd64
-          - linux/arm64
-          
-    steps:
-      - name: Prepare
-        run: |
-          platform=${{ matrix.platform }}
-          echo "PLATFORM_PAIR=${platform//\//-}" >> $GITHUB_ENV
-          
-      - name: Checkout code
-        uses: actions/checkout@v4
-
-      - name: Docker meta
-        id: meta
-        uses: docker/metadata-action@v5
-        with:
-          images: ${{ env.REGISTRY_IMAGE }}
-          flavor: |
-            latest=false
-          tags: |
-            type=raw,value=${{ github.ref_name }}
-            type=raw,value=${{ env.LATEST_TAG == 'true' && 'latest' || '' }}
-            
-      - name: Set up Docker Buildx
-        uses: docker/setup-buildx-action@v3
-
-      - name: Login to Docker Hub
-        uses: docker/login-action@v3
-        with:
-          username: ${{ secrets.DOCKER_USERNAME }}
-          password: ${{ secrets.DOCKER_TOKEN }}
-
-      - name: Install build-essential
-        run: |
-          sudo apt-get update
-          sudo apt-get install -y build-essential
-
-      - name: Backend Image Docker Build and Push
-        id: build
-        uses: docker/build-push-action@v6
-        with:
-          context: ./backend
-          file: ./backend/Dockerfile
-          platforms: ${{ matrix.platform }}
-          push: true
-          build-args: |
-            ONYX_VERSION=${{ github.ref_name }}
-          labels: ${{ steps.meta.outputs.labels }}
-          outputs: type=image,name=${{ env.REGISTRY_IMAGE }},push-by-digest=true,name-canonical=true,push=true
-          cache-from: type=s3,prefix=cache/${{ github.repository }}/${{ env.DEPLOYMENT }}/backend-${{ env.PLATFORM_PAIR }}/,region=${{ env.RUNS_ON_AWS_REGION }},bucket=${{ env.RUNS_ON_S3_BUCKET_CACHE }}
-          cache-to: type=s3,prefix=cache/${{ github.repository }}/${{ env.DEPLOYMENT }}/backend-${{ env.PLATFORM_PAIR }}/,region=${{ env.RUNS_ON_AWS_REGION }},bucket=${{ env.RUNS_ON_S3_BUCKET_CACHE }},mode=max
-
-      - name: Export digest      
-        run: |
-          mkdir -p /tmp/digests
-          digest="${{ steps.build.outputs.digest }}"
-          touch "/tmp/digests/${digest#sha256:}"
-
-      - name: Upload digest
-        uses: actions/upload-artifact@v4
-        with:
-          name: backend-digests-${{ env.PLATFORM_PAIR }}-${{ github.run_id }}
-          path: /tmp/digests/*
-          if-no-files-found: error
-          retention-days: 1
-          
-  merge:
-    runs-on: ubuntu-latest
-    needs:
-      - build-and-push
-    steps:
-      # Needed for trivyignore
-      - name: Checkout
-        uses: actions/checkout@v4
-        
-      - name: Download digests
-        uses: actions/download-artifact@v4
-        with:
-          path: /tmp/digests
-          pattern: backend-digests-*-${{ github.run_id }}
-          merge-multiple: true
-
-      - name: Set up Docker Buildx
-        uses: docker/setup-buildx-action@v3
-
-      - name: Docker meta
-        id: meta
-        uses: docker/metadata-action@v5
-        with:
-          images: ${{ env.REGISTRY_IMAGE }}
-          flavor: |
-            latest=false
-          tags: |
-            type=raw,value=${{ github.ref_name }}
-            type=raw,value=${{ env.LATEST_TAG == 'true' && 'latest' || '' }}
-
-      - name: Login to Docker Hub
-        uses: docker/login-action@v3
-        with:
-          username: ${{ secrets.DOCKER_USERNAME }}
-          password: ${{ secrets.DOCKER_TOKEN }}
-
-      - name: Create manifest list and push
-        working-directory: /tmp/digests
-        run: |
-          docker buildx imagetools create $(jq -cr '.tags | map("-t " + .) | join(" ")' <<< "$DOCKER_METADATA_OUTPUT_JSON") \
-            $(printf '${{ env.REGISTRY_IMAGE }}@sha256:%s ' *)
-
-      - name: Inspect image
-        run: |
-          docker buildx imagetools inspect ${{ env.REGISTRY_IMAGE }}:${{ steps.meta.outputs.version }}
-          
-      # trivy has their own rate limiting issues causing this action to flake
-      # we worked around it by hardcoding to different db repos in env
-      # can re-enable when they figure it out
-      # https://github.com/aquasecurity/trivy/discussions/7538
-      # https://github.com/aquasecurity/trivy-action/issues/389
-      # Security: Using pinned digest (0.65.0@sha256:a22415a38938a56c379387a8163fcb0ce38b10ace73e593475d3658d578b2436)
-      # Security: No Docker socket mount needed for remote registry scanning
-      - name: Run Trivy vulnerability scanner
-        uses: nick-fields/retry@v3
-        with:
-          timeout_minutes: 30
-          max_attempts: 3
-          retry_wait_seconds: 10
-          command: |
-            docker run --rm -v $HOME/.cache/trivy:/root/.cache/trivy \
-              -v ${{ github.workspace }}/backend/.trivyignore:/tmp/.trivyignore:ro \
-              -e TRIVY_DB_REPOSITORY="public.ecr.aws/aquasecurity/trivy-db:2" \
-              -e TRIVY_JAVA_DB_REPOSITORY="public.ecr.aws/aquasecurity/trivy-java-db:1" \
-              -e TRIVY_USERNAME="${{ secrets.DOCKER_USERNAME }}" \
-              -e TRIVY_PASSWORD="${{ secrets.DOCKER_TOKEN }}" \
-              aquasec/trivy@sha256:a22415a38938a56c379387a8163fcb0ce38b10ace73e593475d3658d578b2436 \
-              image \
-              --skip-version-check \
-              --timeout 20m \
-              --severity CRITICAL,HIGH \
-              --ignorefile /tmp/.trivyignore \
-              docker.io/${{ env.REGISTRY_IMAGE }}:${{ github.ref_name }}

.github/workflows/docker-build-push-cloud-web-container-on-tag.yml

@@ -1,158 +0,0 @@
-name: Build and Push Cloud Web Image on Tag
-# Identical to the web container build, but with correct image tag and build args
-
-on:
-  push:
-    tags:
-      - "*cloud*"
-
-env:
-  REGISTRY_IMAGE: onyxdotapp/onyx-web-server-cloud
-  DEPLOYMENT: cloud
-  
-jobs:
-  build:
-    runs-on:
-      - runs-on
-      - runner=${{ matrix.platform == 'linux/amd64' && '8cpu-linux-x64' || '8cpu-linux-arm64' }}
-      - run-id=${{ github.run_id }}
-      - tag=platform-${{ matrix.platform }}
-    strategy:
-      fail-fast: false
-      matrix:
-        platform:
-          - linux/amd64
-          - linux/arm64
-
-    steps:
-      - name: Prepare
-        run: |
-          platform=${{ matrix.platform }}
-          echo "PLATFORM_PAIR=${platform//\//-}" >> $GITHUB_ENV
-
-      - name: Checkout
-        uses: actions/checkout@v4
-
-      - name: Docker meta
-        id: meta
-        uses: docker/metadata-action@v5
-        with:
-          images: ${{ env.REGISTRY_IMAGE }}
-          flavor: |
-            latest=false
-          tags: |
-            type=raw,value=${{ github.ref_name }}
-
-      - name: Set up Docker Buildx
-        uses: docker/setup-buildx-action@v3
-
-      - name: Login to Docker Hub
-        uses: docker/login-action@v3
-        with:
-          username: ${{ secrets.DOCKER_USERNAME }}
-          password: ${{ secrets.DOCKER_TOKEN }}
-
-      - name: Build and push by digest
-        id: build
-        uses: docker/build-push-action@v6
-        with:
-          context: ./web
-          file: ./web/Dockerfile
-          platforms: ${{ matrix.platform }}
-          push: true
-          build-args: |
-            ONYX_VERSION=${{ github.ref_name }}
-            NEXT_PUBLIC_CLOUD_ENABLED=true
-            NEXT_PUBLIC_POSTHOG_KEY=${{ secrets.POSTHOG_KEY }}
-            NEXT_PUBLIC_POSTHOG_HOST=${{ secrets.POSTHOG_HOST }}
-            NEXT_PUBLIC_SENTRY_DSN=${{ secrets.SENTRY_DSN }}
-            NEXT_PUBLIC_STRIPE_PUBLISHABLE_KEY=${{ secrets.STRIPE_PUBLISHABLE_KEY }}
-            NEXT_PUBLIC_GTM_ENABLED=true
-            NEXT_PUBLIC_FORGOT_PASSWORD_ENABLED=true
-            NEXT_PUBLIC_INCLUDE_ERROR_POPUP_SUPPORT_LINK=true
-            NODE_OPTIONS=--max-old-space-size=8192
-          labels: ${{ steps.meta.outputs.labels }}
-          outputs: type=image,name=${{ env.REGISTRY_IMAGE }},push-by-digest=true,name-canonical=true,push=true
-          cache-from: type=s3,prefix=cache/${{ github.repository }}/${{ env.DEPLOYMENT }}/cloudweb-${{ env.PLATFORM_PAIR }}/,region=${{ env.RUNS_ON_AWS_REGION }},bucket=${{ env.RUNS_ON_S3_BUCKET_CACHE }}
-          cache-to: type=s3,prefix=cache/${{ github.repository }}/${{ env.DEPLOYMENT }}/cloudweb-${{ env.PLATFORM_PAIR }}/,region=${{ env.RUNS_ON_AWS_REGION }},bucket=${{ env.RUNS_ON_S3_BUCKET_CACHE }},mode=max
-          # no-cache needed due to weird interactions with the builds for different platforms
-          # NOTE(rkuo): this may not be true any more with the proper cache prefixing by architecture - currently testing with it off
-
-      - name: Export digest
-        run: |
-          mkdir -p /tmp/digests
-          digest="${{ steps.build.outputs.digest }}"
-          touch "/tmp/digests/${digest#sha256:}"
-
-      - name: Upload digest
-        uses: actions/upload-artifact@v4
-        with:
-          name: cloudweb-digests-${{ env.PLATFORM_PAIR }}-${{ github.run_id }}
-          path: /tmp/digests/*
-          if-no-files-found: error
-          retention-days: 1
-
-  merge:
-    runs-on: ubuntu-latest
-    needs:
-      - build
-    steps:
-      - name: Download digests
-        uses: actions/download-artifact@v4
-        with:
-          path: /tmp/digests
-          pattern: cloudweb-digests-*-${{ github.run_id }}
-          merge-multiple: true
-
-      - name: Set up Docker Buildx
-        uses: docker/setup-buildx-action@v3
-
-      - name: Docker meta
-        id: meta
-        uses: docker/metadata-action@v5
-        with:
-          images: ${{ env.REGISTRY_IMAGE }}
-          flavor: |
-            latest=false
-          tags: |
-            type=raw,value=${{ github.ref_name }}
-
-      - name: Login to Docker Hub
-        uses: docker/login-action@v3
-        with:
-          username: ${{ secrets.DOCKER_USERNAME }}
-          password: ${{ secrets.DOCKER_TOKEN }}
-
-      - name: Create manifest list and push
-        working-directory: /tmp/digests
-        run: |
-          docker buildx imagetools create $(jq -cr '.tags | map("-t " + .) | join(" ")' <<< "$DOCKER_METADATA_OUTPUT_JSON") \
-            $(printf '${{ env.REGISTRY_IMAGE }}@sha256:%s ' *)
-
-      - name: Inspect image
-        run: |
-          docker buildx imagetools inspect ${{ env.REGISTRY_IMAGE }}:${{ steps.meta.outputs.version }}
-
-      # trivy has their own rate limiting issues causing this action to flake
-      # we worked around it by hardcoding to different db repos in env
-      # can re-enable when they figure it out
-      # https://github.com/aquasecurity/trivy/discussions/7538
-      # https://github.com/aquasecurity/trivy-action/issues/389
-      - name: Run Trivy vulnerability scanner
-        uses: nick-fields/retry@v3
-        with:
-          timeout_minutes: 30
-          max_attempts: 3
-          retry_wait_seconds: 10
-          command: |
-            docker run --rm -v $HOME/.cache/trivy:/root/.cache/trivy \
-              -e TRIVY_DB_REPOSITORY="public.ecr.aws/aquasecurity/trivy-db:2" \
-              -e TRIVY_JAVA_DB_REPOSITORY="public.ecr.aws/aquasecurity/trivy-java-db:1" \
-              -e TRIVY_USERNAME="${{ secrets.DOCKER_USERNAME }}" \
-              -e TRIVY_PASSWORD="${{ secrets.DOCKER_TOKEN }}" \
-              aquasec/trivy@sha256:a22415a38938a56c379387a8163fcb0ce38b10ace73e593475d3658d578b2436 \
-              image \
-              --skip-version-check \
-              --timeout 20m \
-              --severity CRITICAL,HIGH \
-              docker.io/${{ env.REGISTRY_IMAGE }}:${{ github.ref_name }}

.github/workflows/docker-build-push-model-server-container-on-tag.yml

@@ -1,183 +0,0 @@
-name: Build and Push Model Server Image on Tag
-
-on:
-  push:
-    tags:
-      - "*"
-
-env:
-  REGISTRY_IMAGE: ${{ contains(github.ref_name, 'cloud') && 'onyxdotapp/onyx-model-server-cloud' || 'onyxdotapp/onyx-model-server' }}
-  DOCKER_BUILDKIT: 1
-  BUILDKIT_PROGRESS: plain
-  DEPLOYMENT: ${{ contains(github.ref_name, 'cloud') && 'cloud' || 'standalone' }}
-
-  # don't tag cloud images with "latest"
-  LATEST_TAG: ${{ contains(github.ref_name, 'latest') && !contains(github.ref_name, 'cloud') }}
-  
-jobs:
-
-#   Bypassing this for now as the idea of not building is glitching
-#   releases and builds that depends on everything being tagged in docker
-#   1) Preliminary job to check if the changed files are relevant
-#   check_model_server_changes:
-#     runs-on: ubuntu-latest
-#     outputs:
-#       changed: ${{ steps.check.outputs.changed }}
-#     steps:
-#       - name: Checkout code
-#         uses: actions/checkout@v4
-# 
-#       - name: Check if relevant files changed
-#         id: check
-#         run: |
-#           # Default to "false"
-#           echo "changed=false" >> $GITHUB_OUTPUT
-# 
-#           # Compare the previous commit (github.event.before) to the current one (github.sha)
-#           # If any file in backend/model_server/** or backend/Dockerfile.model_server is changed,
-#           # set changed=true
-#           if git diff --name-only ${{ github.event.before }} ${{ github.sha }} \
-#              | grep -E '^backend/model_server/|^backend/Dockerfile.model_server'; then
-#             echo "changed=true" >> $GITHUB_OUTPUT
-#           fi
-
-  check_model_server_changes:
-    runs-on: ubuntu-latest
-    outputs:
-      changed: "true"
-    steps:
-      - name: Bypass check and set output
-        run: echo "changed=true" >> $GITHUB_OUTPUT
-        
-  build-amd64:
-    needs: [check_model_server_changes]
-    if: needs.check_model_server_changes.outputs.changed == 'true'
-    runs-on:
-      [runs-on, runner=8cpu-linux-x64, "run-id=${{ github.run_id }}-amd64"]
-    env:
-      PLATFORM_PAIR: linux-amd64
-    steps:
-      - name: Checkout code
-        uses: actions/checkout@v4
-
-      - name: System Info
-        run: |
-          df -h
-          free -h
-          docker system prune -af --volumes
-
-      - name: Set up Docker Buildx
-        uses: docker/setup-buildx-action@v3
-        with:
-          driver-opts: |
-            image=moby/buildkit:latest
-            network=host
-
-      - name: Login to Docker Hub
-        uses: docker/login-action@v3
-        with:
-          username: ${{ secrets.DOCKER_USERNAME }}
-          password: ${{ secrets.DOCKER_TOKEN }}
-
-      - name: Build and Push AMD64
-        uses: docker/build-push-action@v6
-        with:
-          context: ./backend
-          file: ./backend/Dockerfile.model_server
-          platforms: linux/amd64
-          push: true
-          tags: ${{ env.REGISTRY_IMAGE }}:${{ github.ref_name }}-amd64
-          build-args: |
-            DANSWER_VERSION=${{ github.ref_name }}
-          outputs: type=registry
-          provenance: false
-          cache-from: type=s3,prefix=cache/${{ github.repository }}/${{ env.DEPLOYMENT }}/model-server-${{ env.PLATFORM_PAIR }}/,region=${{ env.RUNS_ON_AWS_REGION }},bucket=${{ env.RUNS_ON_S3_BUCKET_CACHE }}
-          cache-to: type=s3,prefix=cache/${{ github.repository }}/${{ env.DEPLOYMENT }}/model-server-${{ env.PLATFORM_PAIR }}/,region=${{ env.RUNS_ON_AWS_REGION }},bucket=${{ env.RUNS_ON_S3_BUCKET_CACHE }},mode=max
-#           no-cache: true
-
-  build-arm64:
-    needs: [check_model_server_changes]
-    if: needs.check_model_server_changes.outputs.changed == 'true'
-    runs-on:
-      [runs-on, runner=8cpu-linux-arm64, "run-id=${{ github.run_id }}-arm64"]
-    env:
-      PLATFORM_PAIR: linux-arm64
-    steps:
-      - name: Checkout code
-        uses: actions/checkout@v4
-
-      - name: System Info
-        run: |
-          df -h
-          free -h
-          docker system prune -af --volumes
-
-      - name: Set up Docker Buildx
-        uses: docker/setup-buildx-action@v3
-        with:
-          driver-opts: |
-            image=moby/buildkit:latest
-            network=host
-
-      - name: Login to Docker Hub
-        uses: docker/login-action@v3
-        with:
-          username: ${{ secrets.DOCKER_USERNAME }}
-          password: ${{ secrets.DOCKER_TOKEN }}
-
-      - name: Build and Push ARM64
-        uses: docker/build-push-action@v6
-        with:
-          context: ./backend
-          file: ./backend/Dockerfile.model_server
-          platforms: linux/arm64
-          push: true
-          tags: ${{ env.REGISTRY_IMAGE }}:${{ github.ref_name }}-arm64
-          build-args: |
-            DANSWER_VERSION=${{ github.ref_name }}
-          outputs: type=registry
-          provenance: false
-          cache-from: type=s3,prefix=cache/${{ github.repository }}/${{ env.DEPLOYMENT }}/model-server-${{ env.PLATFORM_PAIR }}/,region=${{ env.RUNS_ON_AWS_REGION }},bucket=${{ env.RUNS_ON_S3_BUCKET_CACHE }}
-          cache-to: type=s3,prefix=cache/${{ github.repository }}/${{ env.DEPLOYMENT }}/model-server-${{ env.PLATFORM_PAIR }}/,region=${{ env.RUNS_ON_AWS_REGION }},bucket=${{ env.RUNS_ON_S3_BUCKET_CACHE }},mode=max
-
-  merge-and-scan:
-    needs: [build-amd64, build-arm64, check_model_server_changes]
-    if: needs.check_model_server_changes.outputs.changed == 'true'
-    runs-on: ubuntu-latest
-    steps:
-      - name: Login to Docker Hub
-        uses: docker/login-action@v3
-        with:
-          username: ${{ secrets.DOCKER_USERNAME }}
-          password: ${{ secrets.DOCKER_TOKEN }}
-
-      - name: Create and Push Multi-arch Manifest
-        run: |
-          docker buildx create --use
-          docker buildx imagetools create -t ${{ env.REGISTRY_IMAGE }}:${{ github.ref_name }} \
-            ${{ env.REGISTRY_IMAGE }}:${{ github.ref_name }}-amd64 \
-            ${{ env.REGISTRY_IMAGE }}:${{ github.ref_name }}-arm64
-          if [[ "${{ env.LATEST_TAG }}" == "true" ]]; then
-            docker buildx imagetools create -t ${{ env.REGISTRY_IMAGE }}:latest \
-              ${{ env.REGISTRY_IMAGE }}:${{ github.ref_name }}-amd64 \
-              ${{ env.REGISTRY_IMAGE }}:${{ github.ref_name }}-arm64
-          fi
-
-      - name: Run Trivy vulnerability scanner
-        uses: nick-fields/retry@v3
-        with:
-          timeout_minutes: 30
-          max_attempts: 3
-          retry_wait_seconds: 10
-          command: |
-            docker run --rm -v $HOME/.cache/trivy:/root/.cache/trivy \
-              -e TRIVY_DB_REPOSITORY="public.ecr.aws/aquasecurity/trivy-db:2" \
-              -e TRIVY_JAVA_DB_REPOSITORY="public.ecr.aws/aquasecurity/trivy-java-db:1" \
-              -e TRIVY_USERNAME="${{ secrets.DOCKER_USERNAME }}" \
-              -e TRIVY_PASSWORD="${{ secrets.DOCKER_TOKEN }}" \
-              aquasec/trivy@sha256:a22415a38938a56c379387a8163fcb0ce38b10ace73e593475d3658d578b2436 \
-              image \
-              --skip-version-check \
-              --timeout 20m \
-              --severity CRITICAL,HIGH \
-              docker.io/${{ env.REGISTRY_IMAGE }}:${{ github.ref_name }}

.github/workflows/docker-build-push-web-container-on-tag.yml

@@ -1,169 +0,0 @@
-name: Build and Push Web Image on Tag
-
-on:
-  push:
-    tags:
-      - "*"
-
-env:
-  REGISTRY_IMAGE: onyxdotapp/onyx-web-server
-  LATEST_TAG: ${{ contains(github.ref_name, 'latest') }}
-  DEPLOYMENT: standalone
-
-jobs:
-  precheck:
-    runs-on: [runs-on, runner=2cpu-linux-x64, "run-id=${{ github.run_id }}"]
-    outputs:
-      should-run: ${{ steps.set-output.outputs.should-run }}
-    steps:
-      - name: Check if tag contains "cloud"
-        id: set-output
-        run: |
-          if [[ "${{ github.ref_name }}" == *cloud* ]]; then
-            echo "should-run=false" >> "$GITHUB_OUTPUT"
-          else
-            echo "should-run=true" >> "$GITHUB_OUTPUT"
-          fi
-  build:
-    needs: precheck
-    if: needs.precheck.outputs.should-run == 'true'
-    runs-on:
-      - runs-on
-      - runner=${{ matrix.platform == 'linux/amd64' && '8cpu-linux-x64' || '8cpu-linux-arm64' }}
-      - run-id=${{ github.run_id }}
-      - tag=platform-${{ matrix.platform }}
-    strategy:
-      fail-fast: false
-      matrix:
-        platform:
-          - linux/amd64
-          - linux/arm64
-
-    steps:
-      - name: Prepare
-        run: |
-          platform=${{ matrix.platform }}
-          echo "PLATFORM_PAIR=${platform//\//-}" >> $GITHUB_ENV
-
-      - name: Checkout
-        uses: actions/checkout@v4
-
-      - name: Docker meta
-        id: meta
-        uses: docker/metadata-action@v5
-        with:
-          images: ${{ env.REGISTRY_IMAGE }}
-          flavor: |
-            latest=false
-          tags: |
-            type=raw,value=${{ github.ref_name }}
-            type=raw,value=${{ env.LATEST_TAG == 'true' && 'latest' || '' }}
-
-      - name: Set up Docker Buildx
-        uses: docker/setup-buildx-action@v3
-
-      - name: Login to Docker Hub
-        uses: docker/login-action@v3
-        with:
-          username: ${{ secrets.DOCKER_USERNAME }}
-          password: ${{ secrets.DOCKER_TOKEN }}
-
-      - name: Build and push by digest
-        id: build
-        uses: docker/build-push-action@v6
-        with:
-          context: ./web
-          file: ./web/Dockerfile
-          platforms: ${{ matrix.platform }}
-          push: true
-          build-args: |
-            ONYX_VERSION=${{ github.ref_name }}
-            NODE_OPTIONS=--max-old-space-size=8192
-
-          labels: ${{ steps.meta.outputs.labels }}
-          outputs: type=image,name=${{ env.REGISTRY_IMAGE }},push-by-digest=true,name-canonical=true,push=true
-          cache-from: type=s3,prefix=cache/${{ github.repository }}/${{ env.DEPLOYMENT }}/web-${{ env.PLATFORM_PAIR }}/,region=${{ env.RUNS_ON_AWS_REGION }},bucket=${{ env.RUNS_ON_S3_BUCKET_CACHE }}
-          cache-to: type=s3,prefix=cache/${{ github.repository }}/${{ env.DEPLOYMENT }}/web-${{ env.PLATFORM_PAIR }}/,region=${{ env.RUNS_ON_AWS_REGION }},bucket=${{ env.RUNS_ON_S3_BUCKET_CACHE }},mode=max
-          # no-cache needed due to weird interactions with the builds for different platforms
-          # NOTE(rkuo): this may not be true any more with the proper cache prefixing by architecture - currently testing with it off
-          
-      - name: Export digest
-        run: |
-          mkdir -p /tmp/digests
-          digest="${{ steps.build.outputs.digest }}"
-          touch "/tmp/digests/${digest#sha256:}"
-
-      - name: Upload digest
-        uses: actions/upload-artifact@v4
-        with:
-          name: web-digests-${{ env.PLATFORM_PAIR }}-${{ github.run_id }}
-          path: /tmp/digests/*
-          if-no-files-found: error
-          retention-days: 1
-
-  merge:
-    needs:
-      - build
-    if: needs.precheck.outputs.should-run == 'true'
-    runs-on: ubuntu-latest
-    steps:
-      - name: Download digests
-        uses: actions/download-artifact@v4
-        with:
-          path: /tmp/digests
-          pattern: web-digests-*-${{ github.run_id }}
-          merge-multiple: true
-
-      - name: Set up Docker Buildx
-        uses: docker/setup-buildx-action@v3
-
-      - name: Docker meta
-        id: meta
-        uses: docker/metadata-action@v5
-        with:
-          images: ${{ env.REGISTRY_IMAGE }}
-          flavor: |
-            latest=false
-          tags: |
-            type=raw,value=${{ github.ref_name }}
-            type=raw,value=${{ env.LATEST_TAG == 'true' && 'latest' || '' }}
-
-      - name: Login to Docker Hub
-        uses: docker/login-action@v3
-        with:
-          username: ${{ secrets.DOCKER_USERNAME }}
-          password: ${{ secrets.DOCKER_TOKEN }}
-
-      - name: Create manifest list and push
-        working-directory: /tmp/digests
-        run: |
-          docker buildx imagetools create $(jq -cr '.tags | map("-t " + .) | join(" ")' <<< "$DOCKER_METADATA_OUTPUT_JSON") \
-            $(printf '${{ env.REGISTRY_IMAGE }}@sha256:%s ' *)
-
-      - name: Inspect image
-        run: |
-          docker buildx imagetools inspect ${{ env.REGISTRY_IMAGE }}:${{ steps.meta.outputs.version }}
-
-      # trivy has their own rate limiting issues causing this action to flake
-      # we worked around it by hardcoding to different db repos in env
-      # can re-enable when they figure it out
-      # https://github.com/aquasecurity/trivy/discussions/7538
-      # https://github.com/aquasecurity/trivy-action/issues/389
-      - name: Run Trivy vulnerability scanner
-        uses: nick-fields/retry@v3
-        with:
-          timeout_minutes: 30
-          max_attempts: 3
-          retry_wait_seconds: 10
-          command: |
-            docker run --rm -v $HOME/.cache/trivy:/root/.cache/trivy \
-              -e TRIVY_DB_REPOSITORY="public.ecr.aws/aquasecurity/trivy-db:2" \
-              -e TRIVY_JAVA_DB_REPOSITORY="public.ecr.aws/aquasecurity/trivy-java-db:1" \
-              -e TRIVY_USERNAME="${{ secrets.DOCKER_USERNAME }}" \
-              -e TRIVY_PASSWORD="${{ secrets.DOCKER_TOKEN }}" \
-              aquasec/trivy@sha256:a22415a38938a56c379387a8163fcb0ce38b10ace73e593475d3658d578b2436 \
-              image \
-              --skip-version-check \
-              --timeout 20m \
-              --severity CRITICAL,HIGH \
-              docker.io/${{ env.REGISTRY_IMAGE }}:${{ github.ref_name }}

.github/workflows/docker-tag-beta.yml

@@ -0,0 +1,51 @@
+# This workflow is set up to be manually triggered via the GitHub Action tab.
+# Given a version, it will tag those backend and webserver images as "beta".
+
+name: Tag Beta Version
+
+on:
+  workflow_dispatch:
+    inputs:
+      version:
+        description: "The version (ie v1.0.0-beta.0) to tag as beta"
+        required: true
+
+permissions:
+  contents: read
+
+jobs:
+  tag:
+    # See https://runs-on.com/runners/linux/
+    # use a lower powered instance since this just does i/o to docker hub
+    runs-on: [runs-on, runner=2cpu-linux-x64, "run-id=${{ github.run_id }}-tag"]
+    timeout-minutes: 45
+    steps:
+      - name: Set up Docker Buildx
+        uses: docker/setup-buildx-action@e468171a9de216ec08956ac3ada2f0791b6bd435 # ratchet:docker/setup-buildx-action@v3
+
+      - name: Login to Docker Hub
+        uses: docker/login-action@5e57cd118135c172c3672efd75eb46360885c0ef # ratchet:docker/login-action@v3
+        with:
+          username: ${{ secrets.DOCKER_USERNAME }}
+          password: ${{ secrets.DOCKER_TOKEN }}
+
+      - name: Enable Docker CLI experimental features
+        run: echo "DOCKER_CLI_EXPERIMENTAL=enabled" >> $GITHUB_ENV
+
+      - name: Pull, Tag and Push Web Server Image
+        env:
+          VERSION: ${{ github.event.inputs.version }}
+        run: |
+          docker buildx imagetools create -t onyxdotapp/onyx-web-server:beta onyxdotapp/onyx-web-server:${VERSION}
+
+      - name: Pull, Tag and Push API Server Image
+        env:
+          VERSION: ${{ github.event.inputs.version }}
+        run: |
+          docker buildx imagetools create -t onyxdotapp/onyx-backend:beta onyxdotapp/onyx-backend:${VERSION}
+
+      - name: Pull, Tag and Push Model Server Image
+        env:
+          VERSION: ${{ github.event.inputs.version }}
+        run: |
+          docker buildx imagetools create -t onyxdotapp/onyx-model-server:beta onyxdotapp/onyx-model-server:${VERSION}

.github/workflows/docker-tag-latest.yml

@@ -10,17 +10,21 @@ on:
         description: "The version (ie v0.0.1) to tag as latest"
         required: true
 
+permissions:
+  contents: read
+
 jobs:
   tag:
     # See https://runs-on.com/runners/linux/
     # use a lower powered instance since this just does i/o to docker hub
-    runs-on: [runs-on, runner=2cpu-linux-x64, "run-id=${{ github.run_id }}"]
+    runs-on: [runs-on, runner=2cpu-linux-x64, "run-id=${{ github.run_id }}-tag"]
+    timeout-minutes: 45
     steps:
       - name: Set up Docker Buildx
-        uses: docker/setup-buildx-action@v1
+        uses: docker/setup-buildx-action@e468171a9de216ec08956ac3ada2f0791b6bd435 # ratchet:docker/setup-buildx-action@v3
 
       - name: Login to Docker Hub
-        uses: docker/login-action@v1
+        uses: docker/login-action@5e57cd118135c172c3672efd75eb46360885c0ef # ratchet:docker/login-action@v3
         with:
           username: ${{ secrets.DOCKER_USERNAME }}
           password: ${{ secrets.DOCKER_TOKEN }}
@@ -29,9 +33,19 @@ jobs:
         run: echo "DOCKER_CLI_EXPERIMENTAL=enabled" >> $GITHUB_ENV
 
       - name: Pull, Tag and Push Web Server Image
+        env:
+          VERSION: ${{ github.event.inputs.version }}
         run: |
-          docker buildx imagetools create -t onyxdotapp/onyx-web-server:latest onyxdotapp/onyx-web-server:${{ github.event.inputs.version }}
+          docker buildx imagetools create -t onyxdotapp/onyx-web-server:latest onyxdotapp/onyx-web-server:${VERSION}
 
       - name: Pull, Tag and Push API Server Image
+        env:
+          VERSION: ${{ github.event.inputs.version }}
         run: |
-          docker buildx imagetools create -t onyxdotapp/onyx-backend:latest onyxdotapp/onyx-backend:${{ github.event.inputs.version }}
+          docker buildx imagetools create -t onyxdotapp/onyx-backend:latest onyxdotapp/onyx-backend:${VERSION}
+
+      - name: Pull, Tag and Push Model Server Image
+        env:
+          VERSION: ${{ github.event.inputs.version }}
+        run: |
+          docker buildx imagetools create -t onyxdotapp/onyx-model-server:latest onyxdotapp/onyx-model-server:${VERSION}

.github/workflows/helm-chart-releases.yml

@@ -12,22 +12,27 @@ jobs:
     permissions:
       contents: write
     runs-on: ubuntu-latest
+    timeout-minutes: 45
     steps:
       - name: Checkout
-        uses: actions/checkout@v4
+        uses: actions/checkout@8e8c483db84b4bee98b60c0593521ed34d9990e8 # ratchet:actions/checkout@v6
         with:
           fetch-depth: 0
+          persist-credentials: false
 
       - name: Install Helm CLI
-        uses: azure/setup-helm@v4
+        uses: azure/setup-helm@1a275c3b69536ee54be43f2070a358922e12c8d4 # ratchet:azure/setup-helm@v4
         with:
           version: v3.12.1
 
       - name: Add required Helm repositories
         run: |
-          helm repo add bitnami https://charts.bitnami.com/bitnami
+          helm repo add ingress-nginx https://kubernetes.github.io/ingress-nginx
           helm repo add onyx-vespa https://onyx-dot-app.github.io/vespa-helm-charts
-          helm repo add keda https://kedacore.github.io/charts
+          helm repo add cloudnative-pg https://cloudnative-pg.github.io/charts
+          helm repo add ot-container-kit https://ot-container-kit.github.io/helm-charts
+          helm repo add minio https://charts.min.io/
+          helm repo add code-interpreter https://onyx-dot-app.github.io/code-interpreter/
           helm repo update
 
       - name: Build chart dependencies
@@ -41,7 +46,7 @@ jobs:
           done
 
       - name: Publish Helm charts to gh-pages
-        uses: stefanprodan/helm-gh-pages@v1.7.0
+        uses: stefanprodan/helm-gh-pages@0ad2bb377311d61ac04ad9eb6f252fb68e207260 # ratchet:stefanprodan/helm-gh-pages@v1.7.0
         with:
           token: ${{ secrets.GITHUB_TOKEN }}
           charts_dir: deployment/helm/charts

.github/workflows/hotfix-release-branches.yml

@@ -1,171 +0,0 @@
-# This workflow is intended to be manually triggered via the GitHub Action tab.
-# Given a hotfix branch, it will attempt to open a PR to all release branches and
-# by default auto merge them
-
-name: Hotfix release branches
-
-on:
-  workflow_dispatch:
-    inputs:
-      hotfix_commit:
-        description: "Hotfix commit hash"
-        required: true
-      hotfix_suffix:
-        description: "Hotfix branch suffix (e.g. hotfix/v0.8-{suffix})"
-        required: true
-      release_branch_pattern:
-        description: "Release branch pattern (regex)"
-        required: true
-        default: "release/.*"
-      auto_merge:
-        description: "Automatically merge the hotfix PRs"
-        required: true
-        type: choice
-        default: "true"
-        options:
-          - true
-          - false
-
-jobs:
-  hotfix_release_branches:
-    permissions: write-all
-    # See https://runs-on.com/runners/linux/
-    # use a lower powered instance since this just does i/o to docker hub
-    runs-on: [runs-on, runner=2cpu-linux-x64, "run-id=${{ github.run_id }}"]
-    steps:
-      # needs RKUO_DEPLOY_KEY for write access to merge PR's
-      - name: Checkout Repository
-        uses: actions/checkout@v4
-        with:
-          ssh-key: "${{ secrets.RKUO_DEPLOY_KEY }}"
-          fetch-depth: 0
-
-      - name: Set up Git user
-        run: |
-          git config user.name "Richard Kuo [bot]"
-          git config user.email "rkuo[bot]@onyx.app"
-
-      - name: Fetch All Branches
-        run: |
-          git fetch --all --prune
-
-      - name: Verify Hotfix Commit Exists
-        run: |
-          git rev-parse --verify "${{ github.event.inputs.hotfix_commit }}" || { echo "Commit not found: ${{ github.event.inputs.hotfix_commit }}"; exit 1; }
-
-      - name: Get Release Branches
-        id: get_release_branches
-        run: |
-          BRANCHES=$(git branch -r | grep -E "${{ github.event.inputs.release_branch_pattern }}" | sed 's|origin/||' | tr -d ' ')
-          if [ -z "$BRANCHES" ]; then
-            echo "No release branches found matching pattern '${{ github.event.inputs.release_branch_pattern }}'."
-            exit 1
-          fi
-
-          echo "Found release branches:"
-          echo "$BRANCHES"
-
-          # Join the branches into a single line separated by commas
-          BRANCHES_JOINED=$(echo "$BRANCHES" | tr '\n' ',' | sed 's/,$//')
-
-          # Set the branches as an output
-          echo "branches=$BRANCHES_JOINED" >> $GITHUB_OUTPUT
-
-      # notes on all the vagaries of wiring up automated PR's
-      # https://github.com/peter-evans/create-pull-request/blob/main/docs/concepts-guidelines.md#triggering-further-workflow-runs
-      # we must use a custom token for GH_TOKEN to trigger the subsequent PR checks
-      - name: Create and Merge Pull Requests to Matching Release Branches
-        env:
-          HOTFIX_COMMIT: ${{ github.event.inputs.hotfix_commit }}
-          HOTFIX_SUFFIX: ${{ github.event.inputs.hotfix_suffix }}
-          AUTO_MERGE: ${{ github.event.inputs.auto_merge }}
-          GH_TOKEN: ${{ secrets.RKUO_PERSONAL_ACCESS_TOKEN }}
-        run: |
-          # Get the branches from the previous step
-          BRANCHES="${{ steps.get_release_branches.outputs.branches }}"
-
-          # Convert BRANCHES to an array
-          IFS=$',' read -ra BRANCH_ARRAY <<< "$BRANCHES"
-
-          # Loop through each release branch and create and merge a PR
-          for RELEASE_BRANCH in "${BRANCH_ARRAY[@]}"; do
-            echo "Processing $RELEASE_BRANCH..."
-            
-            # Parse out the release version by removing "release/" from the branch name
-            RELEASE_VERSION=${RELEASE_BRANCH#release/}
-            echo "Release version parsed: $RELEASE_VERSION"
-            
-            HOTFIX_BRANCH="hotfix/${RELEASE_VERSION}-${HOTFIX_SUFFIX}"
-            echo "Creating PR from $HOTFIX_BRANCH to $RELEASE_BRANCH"
-
-            # Checkout the release branch
-            echo "Checking out $RELEASE_BRANCH"
-            git checkout "$RELEASE_BRANCH"
-
-            # Create the new hotfix branch
-            if git rev-parse --verify "$HOTFIX_BRANCH" >/dev/null 2>&1; then
-              echo "Hotfix branch $HOTFIX_BRANCH already exists. Skipping branch creation."
-            else
-              echo "Branching $RELEASE_BRANCH to $HOTFIX_BRANCH"
-              git checkout -b "$HOTFIX_BRANCH"
-            fi
-            
-            # Check if the hotfix commit is a merge commit
-            if git rev-list --merges -n 1 "$HOTFIX_COMMIT" >/dev/null 2>&1; then
-              # -m 1 uses the target branch as the base (which is what we want)
-              echo "Hotfix commit $HOTFIX_COMMIT is a merge commit, using -m 1 for cherry-pick"
-              CHERRY_PICK_CMD="git cherry-pick -m 1 $HOTFIX_COMMIT"
-            else
-              CHERRY_PICK_CMD="git cherry-pick $HOTFIX_COMMIT"
-            fi
-
-            # Perform the cherry-pick
-            echo "Executing: $CHERRY_PICK_CMD"
-            eval "$CHERRY_PICK_CMD"
-
-            if [ $? -ne 0 ]; then
-              echo "Cherry-pick failed for $HOTFIX_COMMIT on $HOTFIX_BRANCH. Aborting..."
-              git cherry-pick --abort
-              continue
-            fi
-
-            # Push the hotfix branch to the remote
-            echo "Pushing $HOTFIX_BRANCH..."
-            git push origin "$HOTFIX_BRANCH"
-            echo "Hotfix branch $HOTFIX_BRANCH created and pushed."
-            
-            # Check if PR already exists
-            EXISTING_PR=$(gh pr list --head "$HOTFIX_BRANCH" --base "$RELEASE_BRANCH" --state open --json number --jq '.[0].number')
-
-            if [ -n "$EXISTING_PR" ]; then
-              echo "An open PR already exists: #$EXISTING_PR. Skipping..."
-              continue
-            fi
-            
-            # Create a new PR and capture the output
-            PR_OUTPUT=$(gh pr create --title "Merge $HOTFIX_BRANCH into $RELEASE_BRANCH" \
-              --body "Automated PR to merge \`$HOTFIX_BRANCH\` into \`$RELEASE_BRANCH\`." \
-              --head "$HOTFIX_BRANCH" --base "$RELEASE_BRANCH")
-
-            # Extract the URL from the output
-            PR_URL=$(echo "$PR_OUTPUT" | grep -Eo 'https://github.com/[^ ]+')
-            echo "Pull request created: $PR_URL"
-
-            # Extract PR number from URL
-            PR_NUMBER=$(basename "$PR_URL")
-            echo "Pull request created: $PR_NUMBER"
-
-            if [ "$AUTO_MERGE" == "true" ]; then
-              echo "Attempting to merge pull request #$PR_NUMBER"
-
-              # Attempt to merge the PR
-              gh pr merge "$PR_NUMBER" --merge --auto --delete-branch
-
-              if [ $? -eq 0 ]; then
-                echo "Pull request #$PR_NUMBER merged successfully."
-              else
-                # Optionally, handle the error or continue
-                echo "Failed to merge pull request #$PR_NUMBER."
-              fi
-            fi
-          done

.github/workflows/merge-group.yml

@@ -0,0 +1,31 @@
+name: Merge Group-Specific
+
+on:
+  merge_group:
+
+permissions:
+  contents: read
+
+jobs:
+  # This job immediately succeeds to satisfy branch protection rules on merge_group events.
+  # There is a similarly named "required" job in pr-integration-tests.yml which runs the actual
+  # integration tests. That job runs on both pull_request and merge_group events, and this job
+  # exists solely to provide a fast-passing check with the same name for branch protection.
+  # The actual tests remain enforced on presubmit (pull_request events).
+  required:
+    runs-on: ubuntu-latest
+    timeout-minutes: 45
+    steps:
+      - name: Success
+        run: echo "Success"
+  # This job immediately succeeds to satisfy branch protection rules on merge_group events.
+  # There is a similarly named "playwright-required" job in pr-playwright-tests.yml which runs
+  # the actual playwright tests. That job runs on both pull_request and merge_group events, and
+  # this job exists solely to provide a fast-passing check with the same name for branch protection.
+  # The actual tests remain enforced on presubmit (pull_request events).
+  playwright-required:
+    runs-on: ubuntu-latest
+    timeout-minutes: 45
+    steps:
+      - name: Success
+        run: echo "Success"

.github/workflows/nightly-close-stale-issues.yml

@@ -7,12 +7,13 @@ permissions:
   # contents: write # only for delete-branch option
   issues: write
   pull-requests: write
-  
+
 jobs:
   stale:
     runs-on: ubuntu-latest
+    timeout-minutes: 45
     steps:
-      - uses: actions/stale@v9
+      - uses: actions/stale@5f858e3efba33a5ca4407a664cc011ad407f2008 # ratchet:actions/stale@v10
         with:
           stale-issue-message: 'This issue is stale because it has been open 75 days with no activity. Remove stale label or comment or this will be closed in 15 days.'
           stale-pr-message: 'This PR is stale because it has been open 75 days with no activity. Remove stale label or comment or this will be closed in 15 days.'
@@ -20,4 +21,3 @@ jobs:
           close-pr-message: 'This PR was closed because it has been stalled for 90 days with no activity.'
           days-before-stale: 75
 #           days-before-close: 90  # uncomment after we test stale behavior
-          

.github/workflows/nightly-scan-licenses.yml

@@ -15,19 +15,25 @@ on:
 permissions:
   actions: read
   contents: read
-  security-events: write
-  
+
 jobs:
   scan-licenses:
     # See https://runs-on.com/runners/linux/
-    runs-on: [runs-on,runner=2cpu-linux-x64,"run-id=${{ github.run_id }}"]
+    runs-on: [runs-on,runner=2cpu-linux-x64,"run-id=${{ github.run_id }}-scan-licenses"]
+    timeout-minutes: 45
+    permissions:
+      actions: read
+      contents: read
+      security-events: write
 
     steps:
       - name: Checkout code
-        uses: actions/checkout@v4
-        
+        uses: actions/checkout@8e8c483db84b4bee98b60c0593521ed34d9990e8 # ratchet:actions/checkout@v6
+        with:
+          persist-credentials: false
+
       - name: Set up Python
-        uses: actions/setup-python@v5
+        uses: actions/setup-python@83679a892e2d95755f2dac6acb0bfd1e9ac5d548 # ratchet:actions/setup-python@v6
         with:
           python-version: '3.11'
           cache: 'pip'
@@ -35,7 +41,7 @@ jobs:
             backend/requirements/default.txt
             backend/requirements/dev.txt
             backend/requirements/model_server.txt
-      
+
       - name: Get explicit and transitive dependencies
         run: |
           python -m pip install --upgrade pip
@@ -43,28 +49,30 @@ jobs:
           pip install --retries 5 --timeout 30 -r backend/requirements/dev.txt
           pip install --retries 5 --timeout 30 -r backend/requirements/model_server.txt
           pip freeze > requirements-all.txt
-                    
+
       - name: Check python
         id: license_check_report
-        uses: pilosus/action-pip-license-checker@v2
+        uses: pilosus/action-pip-license-checker@e909b0226ff49d3235c99c4585bc617f49fff16a # ratchet:pilosus/action-pip-license-checker@v3
         with:
           requirements: 'requirements-all.txt'
           fail: 'Copyleft'
           exclude: '(?i)^(pylint|aio[-_]*).*'
-          
+
       - name: Print report
         if: always()
-        run: echo "${{ steps.license_check_report.outputs.report }}"
-      
+        env:
+          REPORT: ${{ steps.license_check_report.outputs.report }}
+        run: echo "$REPORT"
+
       - name: Install npm dependencies
         working-directory: ./web
         run: npm ci
 
         # be careful enabling the sarif and upload as it may spam the security tab
-        # with a huge amount of items. Work out the issues before enabling upload.       
+        # with a huge amount of items. Work out the issues before enabling upload.
 #       - name: Run Trivy vulnerability scanner in repo mode
 #         if: always()
-#         uses: aquasecurity/trivy-action@0.29.0
+#         uses: aquasecurity/trivy-action@b6643a29fecd7f34b3597bc6acb0a98b03d33ff8 # ratchet:aquasecurity/trivy-action@0.33.1
 #         with:
 #           scan-type: fs
 #           scan-ref: .
@@ -73,7 +81,7 @@ jobs:
 #           severity: HIGH,CRITICAL
 # #           format: sarif
 # #           output: trivy-results.sarif
-# 
+#
 # #       - name: Upload Trivy scan results to GitHub Security tab
 # #         uses: github/codeql-action/upload-sarif@v3
 # #         with:
@@ -81,14 +89,15 @@ jobs:
 
   scan-trivy:
     # See https://runs-on.com/runners/linux/
-    runs-on: [runs-on,runner=2cpu-linux-x64,"run-id=${{ github.run_id }}"]
-      
+    runs-on: [runs-on,runner=2cpu-linux-x64,"run-id=${{ github.run_id }}-scan-trivy"]
+    timeout-minutes: 45
+
     steps:
     - name: Set up Docker Buildx
-      uses: docker/setup-buildx-action@v3
+      uses: docker/setup-buildx-action@e468171a9de216ec08956ac3ada2f0791b6bd435 # ratchet:docker/setup-buildx-action@v3
 
     - name: Login to Docker Hub
-      uses: docker/login-action@v3
+      uses: docker/login-action@5e57cd118135c172c3672efd75eb46360885c0ef # ratchet:docker/login-action@v3
       with:
         username: ${{ secrets.DOCKER_USERNAME }}
         password: ${{ secrets.DOCKER_TOKEN }}
@@ -98,7 +107,7 @@ jobs:
       run: docker pull onyxdotapp/onyx-backend:latest
 
     - name: Run Trivy vulnerability scanner on backend
-      uses: aquasecurity/trivy-action@0.29.0
+      uses: aquasecurity/trivy-action@b6643a29fecd7f34b3597bc6acb0a98b03d33ff8 # ratchet:aquasecurity/trivy-action@0.33.1
       env:
         TRIVY_DB_REPOSITORY: 'public.ecr.aws/aquasecurity/trivy-db:2'
         TRIVY_JAVA_DB_REPOSITORY: 'public.ecr.aws/aquasecurity/trivy-java-db:1'
@@ -112,9 +121,9 @@ jobs:
     # Web server
     - name: Pull web server docker image
       run: docker pull onyxdotapp/onyx-web-server:latest
-          
+
     - name: Run Trivy vulnerability scanner on web server
-      uses: aquasecurity/trivy-action@0.29.0
+      uses: aquasecurity/trivy-action@b6643a29fecd7f34b3597bc6acb0a98b03d33ff8 # ratchet:aquasecurity/trivy-action@0.33.1
       env:
         TRIVY_DB_REPOSITORY: 'public.ecr.aws/aquasecurity/trivy-db:2'
         TRIVY_JAVA_DB_REPOSITORY: 'public.ecr.aws/aquasecurity/trivy-java-db:1'
@@ -130,7 +139,7 @@ jobs:
       run: docker pull onyxdotapp/onyx-model-server:latest
 
     - name: Run Trivy vulnerability scanner
-      uses: aquasecurity/trivy-action@0.29.0
+      uses: aquasecurity/trivy-action@b6643a29fecd7f34b3597bc6acb0a98b03d33ff8 # ratchet:aquasecurity/trivy-action@0.33.1
       env:
         TRIVY_DB_REPOSITORY: 'public.ecr.aws/aquasecurity/trivy-db:2'
         TRIVY_JAVA_DB_REPOSITORY: 'public.ecr.aws/aquasecurity/trivy-java-db:1'
@@ -139,4 +148,4 @@ jobs:
         scanners: license
         severity: HIGH,CRITICAL
         vuln-type: library
-        exit-code: 0
+        exit-code: 0

.github/workflows/pr-database-tests.yml

@@ -0,0 +1,62 @@
+name: Database Tests
+concurrency:
+  group: Database-Tests-${{ github.workflow }}-${{ github.head_ref || github.event.workflow_run.head_branch || github.run_id }}
+  cancel-in-progress: true
+
+on:
+  merge_group:
+  pull_request:
+    branches:
+      - main
+      - "release/**"
+  push:
+    tags:
+      - "v*.*.*"
+
+permissions:
+  contents: read
+
+jobs:
+  database-tests:
+    runs-on:
+      - runs-on
+      - runner=2cpu-linux-arm64
+      - "run-id=${{ github.run_id }}-database-tests"
+    timeout-minutes: 45
+    steps:
+      - uses: runs-on/action@cd2b598b0515d39d78c38a02d529db87d2196d1e # ratchet:runs-on/action@v2
+
+      - name: Checkout code
+        uses: actions/checkout@8e8c483db84b4bee98b60c0593521ed34d9990e8 # ratchet:actions/checkout@v6
+        with:
+          persist-credentials: false
+
+      - name: Setup Python and Install Dependencies
+        uses: ./.github/actions/setup-python-and-install-dependencies
+        with:
+          requirements: |
+            backend/requirements/default.txt
+            backend/requirements/dev.txt
+
+      - name: Generate OpenAPI schema and Python client
+        shell: bash
+        run: |
+          ods openapi all
+
+      # needed for pulling external images otherwise, we hit the "Unauthenticated users" limit
+      # https://docs.docker.com/docker-hub/usage/
+      - name: Login to Docker Hub
+        uses: docker/login-action@5e57cd118135c172c3672efd75eb46360885c0ef # ratchet:docker/login-action@v3
+        with:
+          username: ${{ secrets.DOCKER_USERNAME }}
+          password: ${{ secrets.DOCKER_TOKEN }}
+
+      - name: Start Docker containers
+        working-directory: ./deployment/docker_compose
+        run: |
+          docker compose -f docker-compose.yml -f docker-compose.dev.yml up -d \
+            relational_db
+
+      - name: Run Database Tests
+        working-directory: ./backend
+        run: pytest -m alembic tests/integration/tests/migrations/

.github/workflows/pr-external-dependency-unit-tests.yml

@@ -1,39 +1,63 @@
 name: External Dependency Unit Tests
+concurrency:
+  group: External-Dependency-Unit-Tests-${{ github.workflow }}-${{ github.head_ref || github.event.workflow_run.head_branch || github.run_id }}
+  cancel-in-progress: true
 
 on:
   merge_group:
   pull_request:
     branches: [main]
+  push:
+    tags:
+      - "v*.*.*"
+
+permissions:
+  contents: read
 
 env:
-  # AWS
-  S3_AWS_ACCESS_KEY_ID: ${{ secrets.S3_AWS_ACCESS_KEY_ID }}
-  S3_AWS_SECRET_ACCESS_KEY: ${{ secrets.S3_AWS_SECRET_ACCESS_KEY }}
+  # AWS credentials for S3-specific test
+  S3_AWS_ACCESS_KEY_ID_FOR_TEST: ${{ secrets.S3_AWS_ACCESS_KEY_ID }}
+  S3_AWS_SECRET_ACCESS_KEY_FOR_TEST: ${{ secrets.S3_AWS_SECRET_ACCESS_KEY }}
 
   # MinIO
   S3_ENDPOINT_URL: "http://localhost:9004"
+  S3_AWS_ACCESS_KEY_ID: "minioadmin"
+  S3_AWS_SECRET_ACCESS_KEY: "minioadmin"
 
   # Confluence
-  CONFLUENCE_TEST_SPACE_URL: ${{ secrets.CONFLUENCE_TEST_SPACE_URL }}
-  CONFLUENCE_TEST_SPACE: ${{ secrets.CONFLUENCE_TEST_SPACE }}
+  CONFLUENCE_TEST_SPACE_URL: ${{ vars.CONFLUENCE_TEST_SPACE_URL }}
+  CONFLUENCE_TEST_SPACE: ${{ vars.CONFLUENCE_TEST_SPACE }}
   CONFLUENCE_TEST_PAGE_ID: ${{ secrets.CONFLUENCE_TEST_PAGE_ID }}
-  CONFLUENCE_IS_CLOUD: ${{ secrets.CONFLUENCE_IS_CLOUD }}
-  CONFLUENCE_USER_NAME: ${{ secrets.CONFLUENCE_USER_NAME }}
+  CONFLUENCE_USER_NAME: ${{ vars.CONFLUENCE_USER_NAME }}
   CONFLUENCE_ACCESS_TOKEN: ${{ secrets.CONFLUENCE_ACCESS_TOKEN }}
+  CONFLUENCE_ACCESS_TOKEN_SCOPED: ${{ secrets.CONFLUENCE_ACCESS_TOKEN_SCOPED }}
+
+  # Jira
+  JIRA_ADMIN_API_TOKEN: ${{ secrets.JIRA_ADMIN_API_TOKEN }}
 
   # LLMs
   OPENAI_API_KEY: ${{ secrets.OPENAI_API_KEY }}
   ANTHROPIC_API_KEY: ${{ secrets.ANTHROPIC_API_KEY }}
+  VERTEX_CREDENTIALS: ${{ secrets.VERTEX_CREDENTIALS }}
+  VERTEX_LOCATION: ${{ vars.VERTEX_LOCATION }}
+
+  # Code Interpreter
+  # TODO: debug why this is failing and enable
+  CODE_INTERPRETER_BASE_URL: http://localhost:8000
 
 jobs:
   discover-test-dirs:
-    runs-on: ubuntu-latest
+    # NOTE: Github-hosted runners have about 20s faster queue times and are preferred here.
+    runs-on: ubuntu-slim
+    timeout-minutes: 45
     outputs:
       test-dirs: ${{ steps.set-matrix.outputs.test-dirs }}
     steps:
       - name: Checkout code
-        uses: actions/checkout@v4
-      
+        uses: actions/checkout@8e8c483db84b4bee98b60c0593521ed34d9990e8 # ratchet:actions/checkout@v6
+        with:
+          persist-credentials: false
+
       - name: Discover test directories
         id: set-matrix
         run: |
@@ -44,8 +68,12 @@ jobs:
   external-dependency-unit-tests:
     needs: discover-test-dirs
     # Use larger runner with more resources for Vespa
-    runs-on: [runs-on, runner=16cpu-linux-x64, "run-id=${{ github.run_id }}"]
-    
+    runs-on:
+      - runs-on
+      - runner=2cpu-linux-arm64
+      - ${{ format('run-id={0}-external-dependency-unit-tests-job-{1}', github.run_id, strategy['job-index']) }}
+      - extras=s3-cache
+    timeout-minutes: 45
     strategy:
       fail-fast: false
       matrix:
@@ -54,43 +82,55 @@ jobs:
     env:
       PYTHONPATH: ./backend
       MODEL_SERVER_HOST: "disabled"
+      DISABLE_TELEMETRY: "true"
 
     steps:
-      - name: Checkout code
-        uses: actions/checkout@v4
+      - uses: runs-on/action@cd2b598b0515d39d78c38a02d529db87d2196d1e # ratchet:runs-on/action@v2
 
-      - name: Set up Python
-        uses: actions/setup-python@v5
+      - name: Checkout code
+        uses: actions/checkout@8e8c483db84b4bee98b60c0593521ed34d9990e8 # ratchet:actions/checkout@v6
         with:
-          python-version: "3.11"
-          cache: "pip"
-          cache-dependency-path: |
+          persist-credentials: false
+
+      - name: Setup Python and Install Dependencies
+        uses: ./.github/actions/setup-python-and-install-dependencies
+        with:
+          requirements: |
             backend/requirements/default.txt
             backend/requirements/dev.txt
+            backend/requirements/ee.txt
 
-      - name: Install Dependencies
+      - name: Setup Playwright
+        uses: ./.github/actions/setup-playwright
+
+      # needed for pulling Vespa, Redis, Postgres, and Minio images
+      # otherwise, we hit the "Unauthenticated users" limit
+      # https://docs.docker.com/docker-hub/usage/
+      - name: Login to Docker Hub
+        uses: docker/login-action@5e57cd118135c172c3672efd75eb46360885c0ef # ratchet:docker/login-action@v3
+        with:
+          username: ${{ secrets.DOCKER_USERNAME }}
+          password: ${{ secrets.DOCKER_TOKEN }}
+
+      - name: Create .env file for Docker Compose
         run: |
-          python -m pip install --upgrade pip
-          pip install --retries 5 --timeout 30 -r backend/requirements/default.txt
-          pip install --retries 5 --timeout 30 -r backend/requirements/dev.txt
-          playwright install chromium
-          playwright install-deps chromium
+          cat <<EOF > deployment/docker_compose/.env
+          CODE_INTERPRETER_BETA_ENABLED=true
+          DISABLE_TELEMETRY=true
+          EOF
 
       - name: Set up Standard Dependencies
         run: |
           cd deployment/docker_compose
-          docker compose up -d minio relational_db cache index
-
-      - name: Wait for services
-        run: |
-          echo "Waiting for services to be ready..."
-          sleep 30
-          
-          # Wait for Vespa specifically
-          echo "Waiting for Vespa to be ready..."
-          timeout 300 bash -c 'until curl -f -s http://localhost:8081/ApplicationStatus > /dev/null 2>&1; do echo "Vespa not ready, waiting..."; sleep 10; done' || echo "Vespa timeout - continuing anyway"
-          
-          echo "Services should be ready now"
+          docker compose \
+            -f docker-compose.yml \
+            -f docker-compose.dev.yml \
+            up -d \
+            minio \
+            relational_db \
+            cache \
+            index \
+            code-interpreter
 
       - name: Run migrations
         run: |
@@ -101,10 +141,39 @@ jobs:
 
       - name: Run Tests for ${{ matrix.test-dir }}
         shell: script -q -e -c "bash --noprofile --norc -eo pipefail {0}"
+        env:
+          TEST_DIR: ${{ matrix.test-dir }}
         run: |
           py.test \
             --durations=8 \
             -o junit_family=xunit2 \
             -xv \
             --ff \
-            backend/tests/external_dependency_unit/${{ matrix.test-dir }}
+            backend/tests/external_dependency_unit/${TEST_DIR}
+
+      - name: Collect Docker logs on failure
+        if: failure()
+        run: |
+          mkdir -p docker-logs
+          cd deployment/docker_compose
+
+          # Get list of running containers
+          containers=$(docker compose -f docker-compose.yml -f docker-compose.dev.yml ps -q)
+
+          # Collect logs from each container
+          for container in $containers; do
+            container_name=$(docker inspect --format='{{.Name}}' $container | sed 's/^\///')
+            echo "Collecting logs from $container_name..."
+            docker logs $container > ../../docker-logs/${container_name}.log 2>&1
+          done
+
+          cd ../..
+          echo "Docker logs collected in docker-logs directory"
+
+      - name: Upload Docker logs
+        if: failure()
+        uses: actions/upload-artifact@330a01c490aca151604b8cf639adc76d48f6c5d4 # ratchet:actions/upload-artifact@v5
+        with:
+          name: docker-logs-${{ matrix.test-dir }}
+          path: docker-logs/
+          retention-days: 7

.github/workflows/pr-helm-chart-testing.yml

@@ -1,206 +1,250 @@
 name: Helm - Lint and Test Charts
+concurrency:
+  group: Helm-Lint-and-Test-Charts-${{ github.workflow }}-${{ github.head_ref || github.event.workflow_run.head_branch || github.run_id }}
+  cancel-in-progress: true
 
 on:
   merge_group:
   pull_request:
-    branches: [ main ]
-  workflow_dispatch:  # Allows manual triggering
-  
+    branches: [main]
+  push:
+    tags:
+      - "v*.*.*"
+  workflow_dispatch: # Allows manual triggering
+
+permissions:
+  contents: read
+
 jobs:
   helm-chart-check:
     # See https://runs-on.com/runners/linux/
-    runs-on: [runs-on,runner=8cpu-linux-x64,hdd=256,"run-id=${{ github.run_id }}"]
+    runs-on:
+      [
+        runs-on,
+        runner=8cpu-linux-x64,
+        hdd=256,
+        "run-id=${{ github.run_id }}-helm-chart-check",
+      ]
+    timeout-minutes: 45
 
     # fetch-depth 0 is required for helm/chart-testing-action
     steps:
-    - name: Checkout code
-      uses: actions/checkout@v4
-      with:
-        fetch-depth: 0
-        
-    - name: Set up Helm
-      uses: azure/setup-helm@v4.2.0
-      with:
-        version: v3.17.0
-      
-    - name: Set up chart-testing
-      uses: helm/chart-testing-action@v2.7.0
+      - name: Checkout code
+        uses: actions/checkout@8e8c483db84b4bee98b60c0593521ed34d9990e8 # ratchet:actions/checkout@v6
+        with:
+          fetch-depth: 0
+          persist-credentials: false
 
-    # even though we specify chart-dirs in ct.yaml, it isn't used by ct for the list-changed command...
-    - name: Run chart-testing (list-changed)
-      id: list-changed
-      run: |
-        echo "default_branch: ${{ github.event.repository.default_branch }}"
-        changed=$(ct list-changed --remote origin --target-branch ${{ github.event.repository.default_branch }} --chart-dirs deployment/helm/charts)
-        echo "list-changed output: $changed"
-        if [[ -n "$changed" ]]; then
-          echo "changed=true" >> "$GITHUB_OUTPUT"
-        fi
+      - name: Set up Helm
+        uses: azure/setup-helm@1a275c3b69536ee54be43f2070a358922e12c8d4 # ratchet:azure/setup-helm@v4.3.1
+        with:
+          version: v3.19.0
 
-    # uncomment to force run chart-testing
-#     - name: Force run chart-testing (list-changed)
-#       id: list-changed
-#       run: echo "changed=true" >> $GITHUB_OUTPUT
-        
-    # lint all charts if any changes were detected
-    - name: Run chart-testing (lint)
-      if: steps.list-changed.outputs.changed == 'true'
-      run: ct lint --config ct.yaml --all
-      # the following would lint only changed charts, but linting isn't expensive
-      # run: ct lint --config ct.yaml --target-branch ${{ github.event.repository.default_branch }}
+      - name: Set up chart-testing
+        # NOTE: This is Jamison's patch from https://github.com/helm/chart-testing-action/pull/194
+        uses: helm/chart-testing-action@8958a6ac472cbd8ee9a8fbb6f1acbc1b0e966e44 # zizmor: ignore[impostor-commit]
+        with:
+          uv_version: "0.9.9"
 
-    - name: Create kind cluster
-      if: steps.list-changed.outputs.changed == 'true'
-      uses: helm/kind-action@v1.12.0
+      # even though we specify chart-dirs in ct.yaml, it isn't used by ct for the list-changed command...
+      - name: Run chart-testing (list-changed)
+        id: list-changed
+        env:
+          DEFAULT_BRANCH: ${{ github.event.repository.default_branch }}
+        run: |
+          echo "default_branch: ${DEFAULT_BRANCH}"
+          changed=$(ct list-changed --remote origin --target-branch ${DEFAULT_BRANCH} --chart-dirs deployment/helm/charts)
+          echo "list-changed output: $changed"
+          if [[ -n "$changed" ]]; then
+            echo "changed=true" >> "$GITHUB_OUTPUT"
+          fi
 
-    - name: Pre-install cluster status check
-      if: steps.list-changed.outputs.changed == 'true'
-      run: |
-        echo "=== Pre-install Cluster Status ==="
-        kubectl get nodes -o wide
-        kubectl get pods --all-namespaces
-        kubectl get storageclass
+      # uncomment to force run chart-testing
+      #     - name: Force run chart-testing (list-changed)
+      #       id: list-changed
+      #       run: echo "changed=true" >> $GITHUB_OUTPUT
+      # lint all charts if any changes were detected
+      - name: Run chart-testing (lint)
+        if: steps.list-changed.outputs.changed == 'true'
+        run: ct lint --config ct.yaml --all
+        # the following would lint only changed charts, but linting isn't expensive
+        # run: ct lint --config ct.yaml --target-branch ${{ github.event.repository.default_branch }}
 
-    - name: Add Helm repositories and update
-      if: steps.list-changed.outputs.changed == 'true'
-      run: |
-        echo "=== Adding Helm repositories ==="
-        helm repo add bitnami https://charts.bitnami.com/bitnami
-        helm repo add vespa https://onyx-dot-app.github.io/vespa-helm-charts
-        helm repo update
+      - name: Create kind cluster
+        if: steps.list-changed.outputs.changed == 'true'
+        uses: helm/kind-action@92086f6be054225fa813e0a4b13787fc9088faab # ratchet:helm/kind-action@v1.13.0
 
-    - name: Pre-pull critical images
-      if: steps.list-changed.outputs.changed == 'true'
-      run: |
-        echo "=== Pre-pulling critical images to avoid timeout ==="
-        # Get kind cluster name
-        KIND_CLUSTER=$(kubectl config current-context | sed 's/kind-//')
-        echo "Kind cluster: $KIND_CLUSTER"
-        
-        # Pre-pull images that are likely to be used
-        echo "Pre-pulling PostgreSQL image..."
-        docker pull postgres:15-alpine || echo "Failed to pull postgres:15-alpine"
-        kind load docker-image postgres:15-alpine --name $KIND_CLUSTER || echo "Failed to load postgres image"
-        
-        echo "Pre-pulling Redis image..."
-        docker pull redis:7-alpine || echo "Failed to pull redis:7-alpine"
-        kind load docker-image redis:7-alpine --name $KIND_CLUSTER || echo "Failed to load redis image"
-        
-        echo "Pre-pulling Onyx images..."
-        docker pull docker.io/onyxdotapp/onyx-web-server:latest || echo "Failed to pull onyx web server"
-        docker pull docker.io/onyxdotapp/onyx-backend:latest || echo "Failed to pull onyx backend"
-        kind load docker-image docker.io/onyxdotapp/onyx-web-server:latest --name $KIND_CLUSTER || echo "Failed to load onyx web server"
-        kind load docker-image docker.io/onyxdotapp/onyx-backend:latest --name $KIND_CLUSTER || echo "Failed to load onyx backend"
-        
-        echo "=== Images loaded into Kind cluster ==="
-        docker exec $KIND_CLUSTER-control-plane crictl images | grep -E "(postgres|redis|onyx)" || echo "Some images may still be loading..."
+      - name: Pre-install cluster status check
+        if: steps.list-changed.outputs.changed == 'true'
+        run: |
+          echo "=== Pre-install Cluster Status ==="
+          kubectl get nodes -o wide
+          kubectl get pods --all-namespaces
+          kubectl get storageclass
 
-    - name: Validate chart dependencies
-      if: steps.list-changed.outputs.changed == 'true'
-      run: |
-        echo "=== Validating chart dependencies ==="
-        cd deployment/helm/charts/onyx
-        helm dependency update
-        helm lint .
+      - name: Add Helm repositories and update
+        if: steps.list-changed.outputs.changed == 'true'
+        run: |
+          echo "=== Adding Helm repositories ==="
+          helm repo add ingress-nginx https://kubernetes.github.io/ingress-nginx
+          helm repo add vespa https://onyx-dot-app.github.io/vespa-helm-charts
+          helm repo add cloudnative-pg https://cloudnative-pg.github.io/charts
+          helm repo add ot-container-kit https://ot-container-kit.github.io/helm-charts
+          helm repo add minio https://charts.min.io/
+          helm repo add code-interpreter https://onyx-dot-app.github.io/code-interpreter/
+          helm repo update
 
-    - name: Run chart-testing (install) with enhanced monitoring
-      timeout-minutes: 25
-      if: steps.list-changed.outputs.changed == 'true'
-      run: |
-        echo "=== Starting chart installation with monitoring ==="
-        
-        # Function to monitor cluster state
-        monitor_cluster() {
-          while true; do
-            echo "=== Cluster Status Check at $(date) ==="
-            # Only show non-running pods to reduce noise
-            NON_RUNNING_PODS=$(kubectl get pods --all-namespaces --field-selector=status.phase!=Running,status.phase!=Succeeded --no-headers 2>/dev/null | wc -l)
-            if [ "$NON_RUNNING_PODS" -gt 0 ]; then
-              echo "Non-running pods:"
-              kubectl get pods --all-namespaces --field-selector=status.phase!=Running,status.phase!=Succeeded
+      - name: Install Redis operator
+        if: steps.list-changed.outputs.changed == 'true'
+        shell: bash
+        run: |
+          echo "=== Installing redis-operator CRDs ==="
+          helm upgrade --install redis-operator ot-container-kit/redis-operator \
+            --namespace redis-operator --create-namespace --wait --timeout 300s
+
+      - name: Pre-pull required images
+        if: steps.list-changed.outputs.changed == 'true'
+        run: |
+          echo "=== Pre-pulling required images to avoid timeout ==="
+          KIND_CLUSTER=$(kubectl config current-context | sed 's/kind-//')
+          echo "Kind cluster: $KIND_CLUSTER"
+
+          IMAGES=(
+            "ghcr.io/cloudnative-pg/cloudnative-pg:1.27.0"
+            "quay.io/opstree/redis:v7.0.15"
+            "docker.io/onyxdotapp/onyx-web-server:latest"
+          )
+
+          for image in "${IMAGES[@]}"; do
+            echo "Pre-pulling $image"
+            if docker pull "$image"; then
+              kind load docker-image "$image" --name "$KIND_CLUSTER" || echo "Failed to load $image into kind"
             else
-              echo "All pods running successfully"
+              echo "Failed to pull $image"
             fi
-            # Only show recent events if there are issues
-            RECENT_EVENTS=$(kubectl get events --sort-by=.lastTimestamp --all-namespaces --field-selector=type!=Normal 2>/dev/null | tail -5)
-            if [ -n "$RECENT_EVENTS" ]; then
-              echo "Recent warnings/errors:"
-              echo "$RECENT_EVENTS"
-            fi
-            sleep 60
           done
-        }
-        
-        # Start monitoring in background
-        monitor_cluster &
-        MONITOR_PID=$!
-        
-        # Set up cleanup
-        cleanup() {
-          echo "=== Cleaning up monitoring process ==="
-          kill $MONITOR_PID 2>/dev/null || true
+
+          echo "=== Images loaded into Kind cluster ==="
+          docker exec "$KIND_CLUSTER"-control-plane crictl images | grep -E "(cloudnative-pg|redis|onyx)" || echo "Some images may still be loading..."
+
+      - name: Validate chart dependencies
+        if: steps.list-changed.outputs.changed == 'true'
+        run: |
+          echo "=== Validating chart dependencies ==="
+          cd deployment/helm/charts/onyx
+          helm dependency update
+          helm lint .
+
+      - name: Run chart-testing (install) with enhanced monitoring
+        timeout-minutes: 25
+        if: steps.list-changed.outputs.changed == 'true'
+        run: |
+          echo "=== Starting chart installation with monitoring ==="
+
+          # Function to monitor cluster state
+          monitor_cluster() {
+            while true; do
+              echo "=== Cluster Status Check at $(date) ==="
+              # Only show non-running pods to reduce noise
+              NON_RUNNING_PODS=$(kubectl get pods --all-namespaces --field-selector=status.phase!=Running,status.phase!=Succeeded --no-headers 2>/dev/null | wc -l)
+              if [ "$NON_RUNNING_PODS" -gt 0 ]; then
+                echo "Non-running pods:"
+                kubectl get pods --all-namespaces --field-selector=status.phase!=Running,status.phase!=Succeeded
+              else
+                echo "All pods running successfully"
+              fi
+              # Only show recent events if there are issues
+              RECENT_EVENTS=$(kubectl get events --sort-by=.lastTimestamp --all-namespaces --field-selector=type!=Normal 2>/dev/null | tail -5)
+              if [ -n "$RECENT_EVENTS" ]; then
+                echo "Recent warnings/errors:"
+                echo "$RECENT_EVENTS"
+              fi
+              sleep 60
+            done
+          }
+
+          # Start monitoring in background
+          monitor_cluster &
+          MONITOR_PID=$!
+
+          # Set up cleanup
+          cleanup() {
+            echo "=== Cleaning up monitoring process ==="
+            kill $MONITOR_PID 2>/dev/null || true
+            echo "=== Final cluster state ==="
+            kubectl get pods --all-namespaces
+            kubectl get events --all-namespaces --sort-by=.lastTimestamp | tail -20
+          }
+
+          # Trap cleanup on exit
+          trap cleanup EXIT
+
+          # Run the actual installation with detailed logging
+          echo "=== Starting ct install ==="
+          set +e
+          ct install --all \
+            --helm-extra-set-args="\
+              --set=nginx.enabled=false \
+              --set=minio.enabled=false \
+              --set=vespa.enabled=false \
+              --set=slackbot.enabled=false \
+              --set=postgresql.enabled=true \
+              --set=postgresql.nameOverride=cloudnative-pg \
+              --set=postgresql.cluster.storage.storageClass=standard \
+              --set=redis.enabled=true \
+              --set=redis.storageSpec.volumeClaimTemplate.spec.storageClassName=standard \
+              --set=webserver.replicaCount=1 \
+              --set=api.replicaCount=0 \
+              --set=inferenceCapability.replicaCount=0 \
+              --set=indexCapability.replicaCount=0 \
+              --set=celery_beat.replicaCount=0 \
+              --set=celery_worker_heavy.replicaCount=0 \
+              --set=celery_worker_docfetching.replicaCount=0 \
+              --set=celery_worker_docprocessing.replicaCount=0 \
+              --set=celery_worker_light.replicaCount=0 \
+              --set=celery_worker_monitoring.replicaCount=0 \
+              --set=celery_worker_primary.replicaCount=0 \
+              --set=celery_worker_user_file_processing.replicaCount=0 \
+              --set=celery_worker_user_files_indexing.replicaCount=0" \
+            --helm-extra-args="--timeout 900s --debug" \
+            --debug --config ct.yaml
+          CT_EXIT=$?
+          set -e
+
+          if [[ $CT_EXIT -ne 0 ]]; then
+            echo "ct install failed with exit code $CT_EXIT"
+            exit $CT_EXIT
+          else
+            echo "=== Installation completed successfully ==="
+          fi
+
+          kubectl get pods --all-namespaces
+
+      - name: Post-install verification
+        if: steps.list-changed.outputs.changed == 'true'
+        run: |
+          echo "=== Post-install verification ==="
+          kubectl get pods --all-namespaces
+          kubectl get services --all-namespaces
+          # Only show issues if they exist
+          kubectl describe pods --all-namespaces | grep -A 5 -B 2 "Failed\|Error\|Warning" || echo "No pod issues found"
+
+      - name: Cleanup on failure
+        if: failure() && steps.list-changed.outputs.changed == 'true'
+        run: |
+          echo "=== Cleanup on failure ==="
           echo "=== Final cluster state ==="
           kubectl get pods --all-namespaces
-          kubectl get events --all-namespaces --sort-by=.lastTimestamp | tail -20
-        }
-        
-        # Trap cleanup on exit
-        trap cleanup EXIT
-        
-        # Run the actual installation with detailed logging
-        echo "=== Starting ct install ==="
-        ct install --all \
-          --helm-extra-set-args="\
-            --set=nginx.enabled=false \
-            --set=minio.enabled=false \
-            --set=vespa.enabled=false \
-            --set=slackbot.enabled=false \
-            --set=postgresql.enabled=true \
-            --set=postgresql.primary.persistence.enabled=false \
-            --set=redis.enabled=true \
-            --set=webserver.replicaCount=1 \
-            --set=api.replicaCount=0 \
-            --set=inferenceCapability.replicaCount=0 \
-            --set=indexCapability.replicaCount=0 \
-            --set=celery_beat.replicaCount=0 \
-            --set=celery_worker_heavy.replicaCount=0 \
-            --set=celery_worker_docfetching.replicaCount=0 \
-            --set=celery_worker_docprocessing.replicaCount=0 \
-            --set=celery_worker_light.replicaCount=0 \
-            --set=celery_worker_monitoring.replicaCount=0 \
-            --set=celery_worker_primary.replicaCount=0 \
-            --set=celery_worker_user_files_indexing.replicaCount=0" \
-          --helm-extra-args="--timeout 900s --debug" \
-          --debug --config ct.yaml
-        
-        echo "=== Installation completed successfully ==="
-        kubectl get pods --all-namespaces
+          kubectl get events --all-namespaces --sort-by=.lastTimestamp | tail -10
 
-    - name: Post-install verification
-      if: steps.list-changed.outputs.changed == 'true'
-      run: |
-        echo "=== Post-install verification ==="
-        kubectl get pods --all-namespaces
-        kubectl get services --all-namespaces
-        # Only show issues if they exist
-        kubectl describe pods --all-namespaces | grep -A 5 -B 2 "Failed\|Error\|Warning" || echo "No pod issues found"
+          echo "=== Pod descriptions for debugging ==="
+          kubectl describe pods --all-namespaces | grep -A 10 -B 3 "Failed\|Error\|Warning\|Pending" || echo "No problematic pods found"
 
-    - name: Cleanup on failure
-      if: failure() && steps.list-changed.outputs.changed == 'true'
-      run: |
-        echo "=== Cleanup on failure ==="
-        echo "=== Final cluster state ==="
-        kubectl get pods --all-namespaces
-        kubectl get events --all-namespaces --sort-by=.lastTimestamp | tail -10
-        
-        echo "=== Pod descriptions for debugging ==="
-        kubectl describe pods --all-namespaces | grep -A 10 -B 3 "Failed\|Error\|Warning\|Pending" || echo "No problematic pods found"
-        
-        echo "=== Recent logs for debugging ==="
-        kubectl logs --all-namespaces --tail=50 | grep -i "error\|timeout\|failed\|pull" || echo "No error logs found"
-        
-        echo "=== Helm releases ==="
-        helm list --all-namespaces
-      # the following would install only changed charts, but we only have one chart so 
-      # don't worry about that for now
-      # run: ct install --target-branch ${{ github.event.repository.default_branch }}
+          echo "=== Recent logs for debugging ==="
+          kubectl logs --all-namespaces --tail=50 | grep -i "error\|timeout\|failed\|pull" || echo "No error logs found"
+
+          echo "=== Helm releases ==="
+          helm list --all-namespaces
+        # the following would install only changed charts, but we only have one chart so
+        # don't worry about that for now
+        # run: ct install --target-branch ${{ github.event.repository.default_branch }}

.github/workflows/pr-integration-tests.yml

@@ -9,41 +9,54 @@ on:
     branches:
       - main
       - "release/**"
+  push:
+    tags:
+      - "v*.*.*"
+
+permissions:
+  contents: read
 
 env:
-  # Private Registry Configuration
-  PRIVATE_REGISTRY: experimental-registry.blacksmith.sh:5000
-  PRIVATE_REGISTRY_USERNAME: ${{ secrets.PRIVATE_REGISTRY_USERNAME }}
-  PRIVATE_REGISTRY_PASSWORD: ${{ secrets.PRIVATE_REGISTRY_PASSWORD }}
-
   # Test Environment Variables
   OPENAI_API_KEY: ${{ secrets.OPENAI_API_KEY }}
   SLACK_BOT_TOKEN: ${{ secrets.SLACK_BOT_TOKEN }}
-  CONFLUENCE_TEST_SPACE_URL: ${{ secrets.CONFLUENCE_TEST_SPACE_URL }}
-  CONFLUENCE_USER_NAME: ${{ secrets.CONFLUENCE_USER_NAME }}
+  CONFLUENCE_TEST_SPACE_URL: ${{ vars.CONFLUENCE_TEST_SPACE_URL }}
+  CONFLUENCE_USER_NAME: ${{ vars.CONFLUENCE_USER_NAME }}
   CONFLUENCE_ACCESS_TOKEN: ${{ secrets.CONFLUENCE_ACCESS_TOKEN }}
+  CONFLUENCE_ACCESS_TOKEN_SCOPED: ${{ secrets.CONFLUENCE_ACCESS_TOKEN_SCOPED }}
   JIRA_BASE_URL: ${{ secrets.JIRA_BASE_URL }}
   JIRA_USER_EMAIL: ${{ secrets.JIRA_USER_EMAIL }}
   JIRA_API_TOKEN: ${{ secrets.JIRA_API_TOKEN }}
+  JIRA_API_TOKEN_SCOPED: ${{ secrets.JIRA_API_TOKEN_SCOPED }}
   PERM_SYNC_SHAREPOINT_CLIENT_ID: ${{ secrets.PERM_SYNC_SHAREPOINT_CLIENT_ID }}
   PERM_SYNC_SHAREPOINT_PRIVATE_KEY: ${{ secrets.PERM_SYNC_SHAREPOINT_PRIVATE_KEY }}
   PERM_SYNC_SHAREPOINT_CERTIFICATE_PASSWORD: ${{ secrets.PERM_SYNC_SHAREPOINT_CERTIFICATE_PASSWORD }}
   PERM_SYNC_SHAREPOINT_DIRECTORY_ID: ${{ secrets.PERM_SYNC_SHAREPOINT_DIRECTORY_ID }}
+  EXA_API_KEY: ${{ secrets.EXA_API_KEY }}
+  GITHUB_PERMISSION_SYNC_TEST_ACCESS_TOKEN: ${{ secrets.ONYX_GITHUB_PERMISSION_SYNC_TEST_ACCESS_TOKEN }}
+  GITHUB_PERMISSION_SYNC_TEST_ACCESS_TOKEN_CLASSIC: ${{ secrets.ONYX_GITHUB_PERMISSION_SYNC_TEST_ACCESS_TOKEN_CLASSIC }}
+  GITHUB_ADMIN_EMAIL: ${{ secrets.ONYX_GITHUB_ADMIN_EMAIL }}
+  GITHUB_TEST_USER_1_EMAIL: ${{ secrets.ONYX_GITHUB_TEST_USER_1_EMAIL }}
+  GITHUB_TEST_USER_2_EMAIL: ${{ secrets.ONYX_GITHUB_TEST_USER_2_EMAIL }}
 
 jobs:
   discover-test-dirs:
-    runs-on: blacksmith-2vcpu-ubuntu-2404-arm
+    # NOTE: Github-hosted runners have about 20s faster queue times and are preferred here.
+    runs-on: ubuntu-slim
+    timeout-minutes: 45
     outputs:
       test-dirs: ${{ steps.set-matrix.outputs.test-dirs }}
     steps:
       - name: Checkout code
-        uses: actions/checkout@v4
+        uses: actions/checkout@8e8c483db84b4bee98b60c0593521ed34d9990e8 # ratchet:actions/checkout@v6
+        with:
+          persist-credentials: false
 
       - name: Discover test directories
         id: set-matrix
         run: |
           # Find all leaf-level directories in both test directories
-          tests_dirs=$(find backend/tests/integration/tests -mindepth 1 -maxdepth 1 -type d ! -name "__pycache__" -exec basename {} \; | sort)
+          tests_dirs=$(find backend/tests/integration/tests -mindepth 1 -maxdepth 1 -type d ! -name "__pycache__" ! -name "mcp" -exec basename {} \; | sort)
           connector_dirs=$(find backend/tests/integration/connector_job_tests -mindepth 1 -maxdepth 1 -type d ! -name "__pycache__" -exec basename {} \; | sort)
 
           # Create JSON array with directory info
@@ -59,141 +72,189 @@ jobs:
           all_dirs="[${all_dirs%,}]"
           echo "test-dirs=$all_dirs" >> $GITHUB_OUTPUT
 
-  prepare-build:
-    runs-on: blacksmith-2vcpu-ubuntu-2404-arm
-    steps:
-      - name: Checkout code
-        uses: actions/checkout@v4
-
-      - name: Setup Python
-        uses: actions/setup-python@v5
-        with:
-          python-version: "3.11"
-          cache: "pip"
-          cache-dependency-path: |
-            backend/requirements/default.txt
-            backend/requirements/dev.txt
-
-      - name: Install Python dependencies
-        run: |
-          python -m pip install --upgrade pip
-          pip install --retries 5 --timeout 30 -r backend/requirements/default.txt
-          pip install --retries 5 --timeout 30 -r backend/requirements/dev.txt
-
-      - name: Generate OpenAPI schema
-        working-directory: ./backend
-        env:
-          PYTHONPATH: "."
-        run: |
-          python scripts/onyx_openapi_schema.py --filename generated/openapi.json
-
-      - name: Generate OpenAPI Python client
-        working-directory: ./backend
-        run: |
-          docker run --rm \
-            -v "${{ github.workspace }}/backend/generated:/local" \
-            openapitools/openapi-generator-cli generate \
-            -i /local/openapi.json \
-            -g python \
-            -o /local/onyx_openapi_client \
-            --package-name onyx_openapi_client \
-            --skip-validate-spec \
-            --openapi-normalizer "SIMPLIFY_ONEOF_ANYOF=true,SET_OAS3_NULLABLE=true"
-
-      - name: Upload OpenAPI artifacts
-        uses: actions/upload-artifact@v4
-        with:
-          name: openapi-artifacts
-          path: backend/generated/
-
   build-backend-image:
-    runs-on: blacksmith-16vcpu-ubuntu-2404-arm
+    runs-on:
+      [
+        runs-on,
+        runner=1cpu-linux-arm64,
+        "run-id=${{ github.run_id }}-build-backend-image",
+        "extras=ecr-cache",
+      ]
+    timeout-minutes: 45
     steps:
+      - uses: runs-on/action@cd2b598b0515d39d78c38a02d529db87d2196d1e # ratchet:runs-on/action@v2
       - name: Checkout code
-        uses: actions/checkout@v4
-
-      - name: Login to Private Registry
-        uses: docker/login-action@v3
+        uses: actions/checkout@8e8c483db84b4bee98b60c0593521ed34d9990e8 # ratchet:actions/checkout@v6
         with:
-          registry: ${{ env.PRIVATE_REGISTRY }}
-          username: ${{ env.PRIVATE_REGISTRY_USERNAME }}
-          password: ${{ env.PRIVATE_REGISTRY_PASSWORD }}
+          persist-credentials: false
+
+      - name: Format branch name for cache
+        id: format-branch
+        env:
+          PR_NUMBER: ${{ github.event.pull_request.number }}
+          REF_NAME: ${{ github.ref_name }}
+        run: |
+          if [ -n "${PR_NUMBER}" ]; then
+            CACHE_SUFFIX="${PR_NUMBER}"
+          else
+            # shellcheck disable=SC2001
+            CACHE_SUFFIX=$(echo "${REF_NAME}" | sed 's/[^A-Za-z0-9._-]/-/g')
+          fi
+          echo "cache-suffix=${CACHE_SUFFIX}" >> $GITHUB_OUTPUT
 
       - name: Set up Docker Buildx
-        uses: useblacksmith/setup-docker-builder@v1
+        uses: docker/setup-buildx-action@e468171a9de216ec08956ac3ada2f0791b6bd435 # ratchet:docker/setup-buildx-action@v3
+
+      # needed for pulling Vespa, Redis, Postgres, and Minio images
+      # otherwise, we hit the "Unauthenticated users" limit
+      # https://docs.docker.com/docker-hub/usage/
+      - name: Login to Docker Hub
+        uses: docker/login-action@5e57cd118135c172c3672efd75eb46360885c0ef # ratchet:docker/login-action@v3
+        with:
+          username: ${{ secrets.DOCKER_USERNAME }}
+          password: ${{ secrets.DOCKER_TOKEN }}
 
       - name: Build and push Backend Docker image
-        uses: useblacksmith/build-push-action@v2
+        uses: docker/build-push-action@263435318d21b8e681c14492fe198d362a7d2c83 # ratchet:docker/build-push-action@v6
         with:
           context: ./backend
           file: ./backend/Dockerfile
-          platforms: linux/arm64
-          tags: ${{ env.PRIVATE_REGISTRY }}/integration-test-onyx-backend:test-${{ github.run_id }}
           push: true
-          outputs: type=registry
-          no-cache: true
+          tags: ${{ env.RUNS_ON_ECR_CACHE }}:integration-test-backend-test-${{ github.run_id }}
+          cache-from: |
+            type=registry,ref=${{ env.RUNS_ON_ECR_CACHE }}:backend-cache-${{ github.event.pull_request.head.sha || github.sha }}
+            type=registry,ref=${{ env.RUNS_ON_ECR_CACHE }}:backend-cache-${{ steps.format-branch.outputs.cache-suffix }}
+            type=registry,ref=${{ env.RUNS_ON_ECR_CACHE }}:backend-cache
+            type=registry,ref=onyxdotapp/onyx-backend:latest
+          cache-to: |
+            type=registry,ref=${{ env.RUNS_ON_ECR_CACHE }}:backend-cache-${{ github.event.pull_request.head.sha || github.sha }},mode=max
+            type=registry,ref=${{ env.RUNS_ON_ECR_CACHE }}:backend-cache-${{ steps.format-branch.outputs.cache-suffix }},mode=max
+            type=registry,ref=${{ env.RUNS_ON_ECR_CACHE }}:backend-cache,mode=max
+          no-cache: ${{ vars.DOCKER_NO_CACHE == 'true' }}
 
   build-model-server-image:
-    runs-on: blacksmith-16vcpu-ubuntu-2404-arm
+    runs-on:
+      [
+        runs-on,
+        runner=1cpu-linux-arm64,
+        "run-id=${{ github.run_id }}-build-model-server-image",
+        "extras=ecr-cache",
+      ]
+    timeout-minutes: 45
     steps:
+      - uses: runs-on/action@cd2b598b0515d39d78c38a02d529db87d2196d1e # ratchet:runs-on/action@v2
       - name: Checkout code
-        uses: actions/checkout@v4
-
-      - name: Login to Private Registry
-        uses: docker/login-action@v3
+        uses: actions/checkout@8e8c483db84b4bee98b60c0593521ed34d9990e8 # ratchet:actions/checkout@v6
         with:
-          registry: ${{ env.PRIVATE_REGISTRY }}
-          username: ${{ env.PRIVATE_REGISTRY_USERNAME }}
-          password: ${{ env.PRIVATE_REGISTRY_PASSWORD }}
+          persist-credentials: false
+
+      - name: Format branch name for cache
+        id: format-branch
+        env:
+          PR_NUMBER: ${{ github.event.pull_request.number }}
+          REF_NAME: ${{ github.ref_name }}
+        run: |
+          if [ -n "${PR_NUMBER}" ]; then
+            CACHE_SUFFIX="${PR_NUMBER}"
+          else
+            # shellcheck disable=SC2001
+            CACHE_SUFFIX=$(echo "${REF_NAME}" | sed 's/[^A-Za-z0-9._-]/-/g')
+          fi
+          echo "cache-suffix=${CACHE_SUFFIX}" >> $GITHUB_OUTPUT
 
       - name: Set up Docker Buildx
-        uses: useblacksmith/setup-docker-builder@v1
+        uses: docker/setup-buildx-action@e468171a9de216ec08956ac3ada2f0791b6bd435 # ratchet:docker/setup-buildx-action@v3
+
+      # needed for pulling Vespa, Redis, Postgres, and Minio images
+      # otherwise, we hit the "Unauthenticated users" limit
+      # https://docs.docker.com/docker-hub/usage/
+      - name: Login to Docker Hub
+        uses: docker/login-action@5e57cd118135c172c3672efd75eb46360885c0ef # ratchet:docker/login-action@v3
+        with:
+          username: ${{ secrets.DOCKER_USERNAME }}
+          password: ${{ secrets.DOCKER_TOKEN }}
 
       - name: Build and push Model Server Docker image
-        uses: useblacksmith/build-push-action@v2
+        uses: docker/build-push-action@263435318d21b8e681c14492fe198d362a7d2c83 # ratchet:docker/build-push-action@v6
         with:
           context: ./backend
           file: ./backend/Dockerfile.model_server
-          platforms: linux/arm64
-          tags: ${{ env.PRIVATE_REGISTRY }}/integration-test-onyx-model-server:test-${{ github.run_id }}
           push: true
-          outputs: type=registry
-          provenance: false
-          no-cache: true
+          tags: ${{ env.RUNS_ON_ECR_CACHE }}:integration-test-model-server-test-${{ github.run_id }}
+          cache-from: |
+            type=registry,ref=${{ env.RUNS_ON_ECR_CACHE }}:model-server-cache-${{ github.event.pull_request.head.sha || github.sha }}
+            type=registry,ref=${{ env.RUNS_ON_ECR_CACHE }}:model-server-cache-${{ steps.format-branch.outputs.cache-suffix }}
+            type=registry,ref=${{ env.RUNS_ON_ECR_CACHE }}:model-server-cache
+            type=registry,ref=onyxdotapp/onyx-model-server:latest
+          cache-to: |
+            type=registry,ref=${{ env.RUNS_ON_ECR_CACHE }}:model-server-cache-${{ github.event.pull_request.head.sha || github.sha }},mode=max
+            type=registry,ref=${{ env.RUNS_ON_ECR_CACHE }}:model-server-cache-${{ steps.format-branch.outputs.cache-suffix }},mode=max
+            type=registry,ref=${{ env.RUNS_ON_ECR_CACHE }}:model-server-cache,mode=max
 
   build-integration-image:
-    needs: prepare-build
-    runs-on: blacksmith-16vcpu-ubuntu-2404-arm
+    runs-on:
+      [
+        runs-on,
+        runner=2cpu-linux-arm64,
+        "run-id=${{ github.run_id }}-build-integration-image",
+        "extras=ecr-cache",
+      ]
+    timeout-minutes: 45
     steps:
+      - uses: runs-on/action@cd2b598b0515d39d78c38a02d529db87d2196d1e # ratchet:runs-on/action@v2
       - name: Checkout code
-        uses: actions/checkout@v4
-
-      - name: Login to Private Registry
-        uses: docker/login-action@v3
+        uses: actions/checkout@8e8c483db84b4bee98b60c0593521ed34d9990e8 # ratchet:actions/checkout@v6
         with:
-          registry: ${{ env.PRIVATE_REGISTRY }}
-          username: ${{ env.PRIVATE_REGISTRY_USERNAME }}
-          password: ${{ env.PRIVATE_REGISTRY_PASSWORD }}
-
-      - name: Download OpenAPI artifacts
-        uses: actions/download-artifact@v4
-        with:
-          name: openapi-artifacts
-          path: backend/generated/
+          persist-credentials: false
 
       - name: Set up Docker Buildx
-        uses: useblacksmith/setup-docker-builder@v1
+        uses: docker/setup-buildx-action@e468171a9de216ec08956ac3ada2f0791b6bd435 # ratchet:docker/setup-buildx-action@v3
 
-      - name: Build and push integration test Docker image
-        uses: useblacksmith/build-push-action@v2
+      # needed for pulling openapitools/openapi-generator-cli
+      # otherwise, we hit the "Unauthenticated users" limit
+      # https://docs.docker.com/docker-hub/usage/
+      - name: Login to Docker Hub
+        uses: docker/login-action@5e57cd118135c172c3672efd75eb46360885c0ef # ratchet:docker/login-action@v3
         with:
-          context: ./backend
-          file: ./backend/tests/integration/Dockerfile
-          platforms: linux/arm64
-          tags: ${{ env.PRIVATE_REGISTRY }}/integration-test-onyx-integration:test-${{ github.run_id }}
-          push: true
-          outputs: type=registry
-          no-cache: true
+          username: ${{ secrets.DOCKER_USERNAME }}
+          password: ${{ secrets.DOCKER_TOKEN }}
+
+      - name: Format branch name for cache
+        id: format-branch
+        env:
+          PR_NUMBER: ${{ github.event.pull_request.number }}
+          REF_NAME: ${{ github.ref_name }}
+        run: |
+          if [ -n "${PR_NUMBER}" ]; then
+            CACHE_SUFFIX="${PR_NUMBER}"
+          else
+            # shellcheck disable=SC2001
+            CACHE_SUFFIX=$(echo "${REF_NAME}" | sed 's/[^A-Za-z0-9._-]/-/g')
+          fi
+          echo "cache-suffix=${CACHE_SUFFIX}" >> $GITHUB_OUTPUT
+
+      - name: Build and push integration test image with Docker Bake
+        env:
+          INTEGRATION_REPOSITORY: ${{ env.RUNS_ON_ECR_CACHE }}
+          TAG: integration-test-${{ github.run_id }}
+          CACHE_SUFFIX: ${{ steps.format-branch.outputs.cache-suffix }}
+          HEAD_SHA: ${{ github.event.pull_request.head.sha || github.sha }}
+        run: |
+          docker buildx bake --push \
+            --set backend.cache-from=type=registry,ref=${RUNS_ON_ECR_CACHE}:backend-cache-${HEAD_SHA} \
+            --set backend.cache-from=type=registry,ref=${RUNS_ON_ECR_CACHE}:backend-cache-${CACHE_SUFFIX} \
+            --set backend.cache-from=type=registry,ref=${RUNS_ON_ECR_CACHE}:backend-cache \
+            --set backend.cache-from=type=registry,ref=onyxdotapp/onyx-backend:latest \
+            --set backend.cache-to=type=registry,ref=${RUNS_ON_ECR_CACHE}:backend-cache-${HEAD_SHA},mode=max \
+            --set backend.cache-to=type=registry,ref=${RUNS_ON_ECR_CACHE}:backend-cache-${CACHE_SUFFIX},mode=max \
+            --set backend.cache-to=type=registry,ref=${RUNS_ON_ECR_CACHE}:backend-cache,mode=max \
+            --set integration.cache-from=type=registry,ref=${RUNS_ON_ECR_CACHE}:integration-cache-${HEAD_SHA} \
+            --set integration.cache-from=type=registry,ref=${RUNS_ON_ECR_CACHE}:integration-cache-${CACHE_SUFFIX} \
+            --set integration.cache-from=type=registry,ref=${RUNS_ON_ECR_CACHE}:integration-cache \
+            --set integration.cache-to=type=registry,ref=${RUNS_ON_ECR_CACHE}:integration-cache-${HEAD_SHA},mode=max \
+            --set integration.cache-to=type=registry,ref=${RUNS_ON_ECR_CACHE}:integration-cache-${CACHE_SUFFIX},mode=max \
+            --set integration.cache-to=type=registry,ref=${RUNS_ON_ECR_CACHE}:integration-cache,mode=max \
+            integration
 
   integration-tests:
     needs:
@@ -203,7 +264,12 @@ jobs:
         build-model-server-image,
         build-integration-image,
       ]
-    runs-on: blacksmith-8vcpu-ubuntu-2404-arm
+    runs-on:
+      - runs-on
+      - runner=4cpu-linux-arm64
+      - ${{ format('run-id={0}-integration-tests-job-{1}', github.run_id, strategy['job-index']) }}
+      - extras=ecr-cache
+    timeout-minutes: 45
 
     strategy:
       fail-fast: false
@@ -211,58 +277,47 @@ jobs:
         test-dir: ${{ fromJson(needs.discover-test-dirs.outputs.test-dirs) }}
 
     steps:
+      - uses: runs-on/action@cd2b598b0515d39d78c38a02d529db87d2196d1e # ratchet:runs-on/action@v2
       - name: Checkout code
-        uses: actions/checkout@v4
-
-      - name: Login to Private Registry
-        uses: docker/login-action@v3
+        uses: actions/checkout@8e8c483db84b4bee98b60c0593521ed34d9990e8 # ratchet:actions/checkout@v6
         with:
-          registry: ${{ env.PRIVATE_REGISTRY }}
-          username: ${{ env.PRIVATE_REGISTRY_USERNAME }}
-          password: ${{ env.PRIVATE_REGISTRY_PASSWORD }}
+          persist-credentials: false
 
       # needed for pulling Vespa, Redis, Postgres, and Minio images
       # otherwise, we hit the "Unauthenticated users" limit
       # https://docs.docker.com/docker-hub/usage/
       - name: Login to Docker Hub
-        uses: docker/login-action@v3
+        uses: docker/login-action@5e57cd118135c172c3672efd75eb46360885c0ef # ratchet:docker/login-action@v3
         with:
           username: ${{ secrets.DOCKER_USERNAME }}
           password: ${{ secrets.DOCKER_TOKEN }}
 
-      - name: Pull Docker images
-        run: |
-          # Pull all images from registry in parallel
-          echo "Pulling Docker images in parallel..."
-          # Pull images from private registry
-          (docker pull --platform linux/arm64 ${{ env.PRIVATE_REGISTRY }}/integration-test-onyx-backend:test-${{ github.run_id }}) &
-          (docker pull --platform linux/arm64 ${{ env.PRIVATE_REGISTRY }}/integration-test-onyx-model-server:test-${{ github.run_id }}) &
-          (docker pull --platform linux/arm64 ${{ env.PRIVATE_REGISTRY }}/integration-test-onyx-integration:test-${{ github.run_id }}) &
-
-          # Wait for all background jobs to complete
-          wait
-          echo "All Docker images pulled successfully"
-
-          # Re-tag to remove registry prefix for docker-compose
-          docker tag ${{ env.PRIVATE_REGISTRY }}/integration-test-onyx-backend:test-${{ github.run_id }} onyxdotapp/onyx-backend:test
-          docker tag ${{ env.PRIVATE_REGISTRY }}/integration-test-onyx-model-server:test-${{ github.run_id }} onyxdotapp/onyx-model-server:test
-          docker tag ${{ env.PRIVATE_REGISTRY }}/integration-test-onyx-integration:test-${{ github.run_id }} onyxdotapp/onyx-integration:test
-
       # NOTE: Use pre-ping/null pool to reduce flakiness due to dropped connections
       # NOTE: don't need web server for integration tests
+      - name: Create .env file for Docker Compose
+        env:
+          ECR_CACHE: ${{ env.RUNS_ON_ECR_CACHE }}
+          RUN_ID: ${{ github.run_id }}
+        run: |
+          cat <<EOF > deployment/docker_compose/.env
+          ENABLE_PAID_ENTERPRISE_EDITION_FEATURES=true
+          AUTH_TYPE=basic
+          POSTGRES_POOL_PRE_PING=true
+          POSTGRES_USE_NULL_POOL=true
+          REQUIRE_EMAIL_VERIFICATION=false
+          DISABLE_TELEMETRY=true
+          ONYX_BACKEND_IMAGE=${ECR_CACHE}:integration-test-backend-test-${RUN_ID}
+          ONYX_MODEL_SERVER_IMAGE=${ECR_CACHE}:integration-test-model-server-test-${RUN_ID}
+          INTEGRATION_TESTS_MODE=true
+          CHECK_TTL_MANAGEMENT_TASK_FREQUENCY_IN_HOURS=0.001
+          AUTO_LLM_UPDATE_INTERVAL_SECONDS=1
+          MCP_SERVER_ENABLED=true
+          EOF
+
       - name: Start Docker containers
         run: |
           cd deployment/docker_compose
-          ENABLE_PAID_ENTERPRISE_EDITION_FEATURES=true \
-          AUTH_TYPE=basic \
-          POSTGRES_POOL_PRE_PING=true \
-          POSTGRES_USE_NULL_POOL=true \
-          REQUIRE_EMAIL_VERIFICATION=false \
-          DISABLE_TELEMETRY=true \
-          IMAGE_TAG=test \
-          INTEGRATION_TESTS_MODE=true \
-          CHECK_TTL_MANAGEMENT_TASK_FREQUENCY_IN_HOURS=0.001 \
-          docker compose up \
+          docker compose -f docker-compose.yml -f docker-compose.dev.yml up \
             relational_db \
             index \
             cache \
@@ -274,39 +329,45 @@ jobs:
             -d
         id: start_docker
 
-      - name: Wait for service to be ready
+      - name: Wait for services to be ready
         run: |
           echo "Starting wait-for-service script..."
 
-          docker logs -f onyx-api_server-1 &
+          wait_for_service() {
+            local url=$1
+            local label=$2
+            local timeout=${3:-300}  # default 5 minutes
+            local start_time
+            start_time=$(date +%s)
 
-          start_time=$(date +%s)
-          timeout=300  # 5 minutes in seconds
+            while true; do
+              local current_time
+              current_time=$(date +%s)
+              local elapsed_time=$((current_time - start_time))
 
-          while true; do
-            current_time=$(date +%s)
-            elapsed_time=$((current_time - start_time))
-            
-            if [ $elapsed_time -ge $timeout ]; then
-              echo "Timeout reached. Service did not become ready in 5 minutes."
-              exit 1
-            fi
-            
-            # Use curl with error handling to ignore specific exit code 56
-            response=$(curl -s -o /dev/null -w "%{http_code}" http://localhost:8080/health || echo "curl_error")
-            
-            if [ "$response" = "200" ]; then
-              echo "Service is ready!"
-              break
-            elif [ "$response" = "curl_error" ]; then
-              echo "Curl encountered an error, possibly exit code 56. Continuing to retry..."
-            else
-              echo "Service not ready yet (HTTP status $response). Retrying in 5 seconds..."
-            fi
-            
-            sleep 5
-          done
-          echo "Finished waiting for service."
+              if [ $elapsed_time -ge $timeout ]; then
+                echo "Timeout reached. ${label} did not become ready in $timeout seconds."
+                exit 1
+              fi
+
+              local response
+              response=$(curl -s -o /dev/null -w "%{http_code}" "$url" || echo "curl_error")
+
+              if [ "$response" = "200" ]; then
+                echo "${label} is ready!"
+                break
+              elif [ "$response" = "curl_error" ]; then
+                echo "Curl encountered an error while checking ${label}. Retrying in 5 seconds..."
+              else
+                echo "${label} not ready yet (HTTP status $response). Retrying in 5 seconds..."
+              fi
+
+              sleep 5
+            done
+          }
+
+          wait_for_service "http://localhost:8080/health" "API server"
+          echo "Finished waiting for services."
 
       - name: Start Mock Services
         run: |
@@ -315,7 +376,7 @@ jobs:
             -p mock-it-services-stack up -d
 
       - name: Run Integration Tests for ${{ matrix.test-dir.name }}
-        uses: nick-fields/retry@v3
+        uses: nick-fields/retry@ce71cc2ab81d554ebbe88c79ab5975992d79ba08 # ratchet:nick-fields/retry@v3
         with:
           timeout_minutes: 20
           max_attempts: 3
@@ -336,21 +397,29 @@ jobs:
               -e REDIS_HOST=cache \
               -e API_SERVER_HOST=api_server \
               -e OPENAI_API_KEY=${OPENAI_API_KEY} \
+              -e EXA_API_KEY=${EXA_API_KEY} \
               -e SLACK_BOT_TOKEN=${SLACK_BOT_TOKEN} \
               -e CONFLUENCE_TEST_SPACE_URL=${CONFLUENCE_TEST_SPACE_URL} \
               -e CONFLUENCE_USER_NAME=${CONFLUENCE_USER_NAME} \
               -e CONFLUENCE_ACCESS_TOKEN=${CONFLUENCE_ACCESS_TOKEN} \
+              -e CONFLUENCE_ACCESS_TOKEN_SCOPED=${CONFLUENCE_ACCESS_TOKEN_SCOPED} \
               -e JIRA_BASE_URL=${JIRA_BASE_URL} \
               -e JIRA_USER_EMAIL=${JIRA_USER_EMAIL} \
               -e JIRA_API_TOKEN=${JIRA_API_TOKEN} \
+              -e JIRA_API_TOKEN_SCOPED=${JIRA_API_TOKEN_SCOPED} \
               -e PERM_SYNC_SHAREPOINT_CLIENT_ID=${PERM_SYNC_SHAREPOINT_CLIENT_ID} \
               -e PERM_SYNC_SHAREPOINT_PRIVATE_KEY="${PERM_SYNC_SHAREPOINT_PRIVATE_KEY}" \
               -e PERM_SYNC_SHAREPOINT_CERTIFICATE_PASSWORD=${PERM_SYNC_SHAREPOINT_CERTIFICATE_PASSWORD} \
               -e PERM_SYNC_SHAREPOINT_DIRECTORY_ID=${PERM_SYNC_SHAREPOINT_DIRECTORY_ID} \
+              -e GITHUB_PERMISSION_SYNC_TEST_ACCESS_TOKEN=${GITHUB_PERMISSION_SYNC_TEST_ACCESS_TOKEN} \
+              -e GITHUB_PERMISSION_SYNC_TEST_ACCESS_TOKEN_CLASSIC=${GITHUB_PERMISSION_SYNC_TEST_ACCESS_TOKEN_CLASSIC} \
+              -e GITHUB_ADMIN_EMAIL=${GITHUB_ADMIN_EMAIL} \
+              -e GITHUB_TEST_USER_1_EMAIL=${GITHUB_TEST_USER_1_EMAIL} \
+              -e GITHUB_TEST_USER_2_EMAIL=${GITHUB_TEST_USER_2_EMAIL} \
               -e TEST_WEB_HOSTNAME=test-runner \
               -e MOCK_CONNECTOR_SERVER_HOST=mock_connector_server \
               -e MOCK_CONNECTOR_SERVER_PORT=8001 \
-              onyxdotapp/onyx-integration:test \
+              ${{ env.RUNS_ON_ECR_CACHE }}:integration-test-${{ github.run_id }} \
               /app/tests/integration/${{ matrix.test-dir.path }}
 
       # ------------------------------------------------------------
@@ -369,56 +438,41 @@ jobs:
 
       - name: Upload logs
         if: always()
-        uses: actions/upload-artifact@v4
+        uses: actions/upload-artifact@330a01c490aca151604b8cf639adc76d48f6c5d4 # ratchet:actions/upload-artifact@v4
         with:
           name: docker-all-logs-${{ matrix.test-dir.name }}
           path: ${{ github.workspace }}/docker-compose.log
       # ------------------------------------------------------------
 
-      - name: Stop Docker containers
-        if: always()
-        run: |
-          cd deployment/docker_compose
-          docker compose down -v
-
-
   multitenant-tests:
     needs:
+      [build-backend-image, build-model-server-image, build-integration-image]
+    runs-on:
       [
-        build-backend-image,
-        build-model-server-image,
-        build-integration-image,
+        runs-on,
+        runner=8cpu-linux-arm64,
+        "run-id=${{ github.run_id }}-multitenant-tests",
+        "extras=ecr-cache",
       ]
-    runs-on: blacksmith-8vcpu-ubuntu-2404-arm
+    timeout-minutes: 45
 
     steps:
+      - uses: runs-on/action@cd2b598b0515d39d78c38a02d529db87d2196d1e # ratchet:runs-on/action@v2
       - name: Checkout code
-        uses: actions/checkout@v4
-
-      - name: Login to Private Registry
-        uses: docker/login-action@v3
+        uses: actions/checkout@8e8c483db84b4bee98b60c0593521ed34d9990e8 # ratchet:actions/checkout@v6
         with:
-          registry: ${{ env.PRIVATE_REGISTRY }}
-          username: ${{ env.PRIVATE_REGISTRY_USERNAME }}
-          password: ${{ env.PRIVATE_REGISTRY_PASSWORD }}
+          persist-credentials: false
 
       - name: Login to Docker Hub
-        uses: docker/login-action@v3
+        uses: docker/login-action@5e57cd118135c172c3672efd75eb46360885c0ef # ratchet:docker/login-action@v3
         with:
           username: ${{ secrets.DOCKER_USERNAME }}
           password: ${{ secrets.DOCKER_TOKEN }}
 
-      - name: Pull Docker images
-        run: |
-          (docker pull --platform linux/arm64 ${{ env.PRIVATE_REGISTRY }}/integration-test-onyx-backend:test-${{ github.run_id }}) &
-          (docker pull --platform linux/arm64 ${{ env.PRIVATE_REGISTRY }}/integration-test-onyx-model-server:test-${{ github.run_id }}) &
-          (docker pull --platform linux/arm64 ${{ env.PRIVATE_REGISTRY }}/integration-test-onyx-integration:test-${{ github.run_id }}) &
-          wait
-          docker tag ${{ env.PRIVATE_REGISTRY }}/integration-test-onyx-backend:test-${{ github.run_id }} onyxdotapp/onyx-backend:test
-          docker tag ${{ env.PRIVATE_REGISTRY }}/integration-test-onyx-model-server:test-${{ github.run_id }} onyxdotapp/onyx-model-server:test
-          docker tag ${{ env.PRIVATE_REGISTRY }}/integration-test-onyx-integration:test-${{ github.run_id }} onyxdotapp/onyx-integration:test
-
       - name: Start Docker containers for multi-tenant tests
+        env:
+          ECR_CACHE: ${{ env.RUNS_ON_ECR_CACHE }}
+          RUN_ID: ${{ github.run_id }}
         run: |
           cd deployment/docker_compose
           ENABLE_PAID_ENTERPRISE_EDITION_FEATURES=true \
@@ -426,7 +480,9 @@ jobs:
           AUTH_TYPE=cloud \
           REQUIRE_EMAIL_VERIFICATION=false \
           DISABLE_TELEMETRY=true \
-          IMAGE_TAG=test \
+          OPENAI_DEFAULT_API_KEY=${OPENAI_API_KEY} \
+          ONYX_BACKEND_IMAGE=${ECR_CACHE}:integration-test-backend-test-${RUN_ID} \
+          ONYX_MODEL_SERVER_IMAGE=${ECR_CACHE}:integration-test-model-server-test-${RUN_ID} \
           DEV_MODE=true \
           docker compose -f docker-compose.multitenant-dev.yml up \
             relational_db \
@@ -467,6 +523,9 @@ jobs:
           echo "Finished waiting for service."
 
       - name: Run Multi-Tenant Integration Tests
+        env:
+          ECR_CACHE: ${{ env.RUNS_ON_ECR_CACHE }}
+          RUN_ID: ${{ github.run_id }}
         run: |
           echo "Running multi-tenant integration tests..."
           docker run --rm --network onyx_default \
@@ -482,6 +541,7 @@ jobs:
             -e REDIS_HOST=cache \
             -e API_SERVER_HOST=api_server \
             -e OPENAI_API_KEY=${OPENAI_API_KEY} \
+            -e EXA_API_KEY=${EXA_API_KEY} \
             -e SLACK_BOT_TOKEN=${SLACK_BOT_TOKEN} \
             -e TEST_WEB_HOSTNAME=test-runner \
             -e AUTH_TYPE=cloud \
@@ -489,9 +549,8 @@ jobs:
             -e SKIP_RESET=true \
             -e REQUIRE_EMAIL_VERIFICATION=false \
             -e DISABLE_TELEMETRY=true \
-            -e IMAGE_TAG=test \
             -e DEV_MODE=true \
-            onyxdotapp/onyx-integration:test \
+            ${ECR_CACHE}:integration-test-${RUN_ID} \
             /app/tests/integration/multitenant_tests
 
       - name: Dump API server logs (multi-tenant)
@@ -508,7 +567,7 @@ jobs:
 
       - name: Upload logs (multi-tenant)
         if: always()
-        uses: actions/upload-artifact@v4
+        uses: actions/upload-artifact@330a01c490aca151604b8cf639adc76d48f6c5d4 # ratchet:actions/upload-artifact@v4
         with:
           name: docker-all-logs-multitenant
           path: ${{ github.workspace }}/docker-compose-multitenant.log
@@ -519,18 +578,13 @@ jobs:
           cd deployment/docker_compose
           docker compose -f docker-compose.multitenant-dev.yml down -v
 
-  required: 
-    runs-on: blacksmith-2vcpu-ubuntu-2404-arm
+  required:
+    # NOTE: Github-hosted runners have about 20s faster queue times and are preferred here.
+    runs-on: ubuntu-slim
+    timeout-minutes: 45
     needs: [integration-tests, multitenant-tests]
     if: ${{ always() }}
     steps:
-      - uses: actions/github-script@v7
-        with:
-          script: |
-            const needs = ${{ toJSON(needs) }};
-            const failed = Object.values(needs).some(n => n.result !== 'success');
-            if (failed) {
-              core.setFailed('One or more upstream jobs failed or were cancelled.');
-            } else {
-              core.notice('All required jobs succeeded.');
-            }
+      - name: Check job status
+        if: ${{ contains(needs.*.result, 'failure') || contains(needs.*.result, 'cancelled') || contains(needs.*.result, 'skipped') }}
+        run: exit 1

.github/workflows/pr-jest-tests.yml

@@ -0,0 +1,51 @@
+name: Run Jest Tests
+concurrency:
+  group: Run-Jest-Tests-${{ github.workflow }}-${{ github.head_ref || github.event.workflow_run.head_branch || github.run_id }}
+  cancel-in-progress: true
+
+on:
+  merge_group:
+  pull_request:
+    branches:
+      - main
+      - "release/**"
+  push:
+    tags:
+      - "v*.*.*"
+
+permissions:
+  contents: read
+
+jobs:
+  jest-tests:
+    name: Jest Tests
+    runs-on: ubuntu-latest
+    timeout-minutes: 45
+    steps:
+      - name: Checkout code
+        uses: actions/checkout@8e8c483db84b4bee98b60c0593521ed34d9990e8 # ratchet:actions/checkout@v6
+        with:
+          persist-credentials: false
+
+      - name: Setup node
+        uses: actions/setup-node@395ad3262231945c25e8478fd5baf05154b1d79f # ratchet:actions/setup-node@v4
+        with:
+          node-version: 22
+          cache: "npm"
+          cache-dependency-path: ./web/package-lock.json
+
+      - name: Install node dependencies
+        working-directory: ./web
+        run: npm ci
+
+      - name: Run Jest tests
+        working-directory: ./web
+        run: npm test -- --ci --coverage --maxWorkers=50%
+
+      - name: Upload coverage reports
+        if: always()
+        uses: actions/upload-artifact@330a01c490aca151604b8cf639adc76d48f6c5d4 # ratchet:actions/upload-artifact@v4
+        with:
+          name: jest-coverage-${{ github.run_id }}
+          path: ./web/coverage
+          retention-days: 7

.github/workflows/pr-labeler.yml

@@ -1,7 +1,7 @@
 name: PR Labeler
 
 on:
-  pull_request_target:
+  pull_request:
     branches:
       - main
     types:
@@ -12,11 +12,11 @@ on:
 
 permissions:
   contents: read
-  pull-requests: write
 
 jobs:
   validate_pr_title:
     runs-on: ubuntu-latest
+    timeout-minutes: 45
     steps:
       - name: Check PR title for Conventional Commits
         env:
@@ -27,7 +27,7 @@ jobs:
             echo "::error::❌ Your PR title does not follow the Conventional Commits format.
               This check ensures that all pull requests use clear, consistent titles that help automate changelogs and improve project history.
 
-              Please update your PR title to follow the Conventional Commits style.  
+              Please update your PR title to follow the Conventional Commits style.
               Here is a link to a blog explaining the reason why we've included the Conventional Commits style into our PR titles: https://xfuture-blog.com/working-with-conventional-commits
 
               **Here are some examples of valid PR titles:**

.github/workflows/pr-linear-check.yml

@@ -1,12 +1,19 @@
 name: Ensure PR references Linear
+concurrency:
+  group: Ensure-PR-references-Linear-${{ github.workflow }}-${{ github.head_ref || github.event.workflow_run.head_branch || github.run_id }}
+  cancel-in-progress: true
 
 on:
   pull_request:
     types: [opened, edited, reopened, synchronize]
 
+permissions:
+  contents: read
+
 jobs:
   linear-check:
     runs-on: ubuntu-latest
+    timeout-minutes: 45
     steps:
       - name: Check PR body for Linear link or override
         env:

.github/workflows/pr-mit-integration-tests.yml

@@ -6,22 +6,26 @@ concurrency:
 on:
   merge_group:
     types: [checks_requested]
+  push:
+    tags:
+      - "v*.*.*"
+
+permissions:
+  contents: read
 
 env:
-  # Private Registry Configuration
-  PRIVATE_REGISTRY: experimental-registry.blacksmith.sh:5000
-  PRIVATE_REGISTRY_USERNAME: ${{ secrets.PRIVATE_REGISTRY_USERNAME }}
-  PRIVATE_REGISTRY_PASSWORD: ${{ secrets.PRIVATE_REGISTRY_PASSWORD }}
-
   # Test Environment Variables
   OPENAI_API_KEY: ${{ secrets.OPENAI_API_KEY }}
   SLACK_BOT_TOKEN: ${{ secrets.SLACK_BOT_TOKEN }}
-  CONFLUENCE_TEST_SPACE_URL: ${{ secrets.CONFLUENCE_TEST_SPACE_URL }}
-  CONFLUENCE_USER_NAME: ${{ secrets.CONFLUENCE_USER_NAME }}
+  EXA_API_KEY: ${{ secrets.EXA_API_KEY }}
+  CONFLUENCE_TEST_SPACE_URL: ${{ vars.CONFLUENCE_TEST_SPACE_URL }}
+  CONFLUENCE_USER_NAME: ${{ vars.CONFLUENCE_USER_NAME }}
   CONFLUENCE_ACCESS_TOKEN: ${{ secrets.CONFLUENCE_ACCESS_TOKEN }}
+  CONFLUENCE_ACCESS_TOKEN_SCOPED: ${{ secrets.CONFLUENCE_ACCESS_TOKEN_SCOPED }}
   JIRA_BASE_URL: ${{ secrets.JIRA_BASE_URL }}
   JIRA_USER_EMAIL: ${{ secrets.JIRA_USER_EMAIL }}
   JIRA_API_TOKEN: ${{ secrets.JIRA_API_TOKEN }}
+  JIRA_API_TOKEN_SCOPED: ${{ secrets.JIRA_API_TOKEN_SCOPED }}
   PERM_SYNC_SHAREPOINT_CLIENT_ID: ${{ secrets.PERM_SYNC_SHAREPOINT_CLIENT_ID }}
   PERM_SYNC_SHAREPOINT_PRIVATE_KEY: ${{ secrets.PERM_SYNC_SHAREPOINT_PRIVATE_KEY }}
   PERM_SYNC_SHAREPOINT_CERTIFICATE_PASSWORD: ${{ secrets.PERM_SYNC_SHAREPOINT_CERTIFICATE_PASSWORD }}
@@ -29,18 +33,22 @@ env:
 
 jobs:
   discover-test-dirs:
-    runs-on: blacksmith-2vcpu-ubuntu-2404-arm
+    # NOTE: Github-hosted runners have about 20s faster queue times and are preferred here.
+    runs-on: ubuntu-slim
+    timeout-minutes: 45
     outputs:
       test-dirs: ${{ steps.set-matrix.outputs.test-dirs }}
     steps:
       - name: Checkout code
-        uses: actions/checkout@v4
+        uses: actions/checkout@8e8c483db84b4bee98b60c0593521ed34d9990e8 # ratchet:actions/checkout@v6
+        with:
+          persist-credentials: false
 
       - name: Discover test directories
         id: set-matrix
         run: |
           # Find all leaf-level directories in both test directories
-          tests_dirs=$(find backend/tests/integration/tests -mindepth 1 -maxdepth 1 -type d ! -name "__pycache__" -exec basename {} \; | sort)
+          tests_dirs=$(find backend/tests/integration/tests -mindepth 1 -maxdepth 1 -type d ! -name "__pycache__" ! -name "mcp" -exec basename {} \; | sort)
           connector_dirs=$(find backend/tests/integration/connector_job_tests -mindepth 1 -maxdepth 1 -type d ! -name "__pycache__" -exec basename {} \; | sort)
 
           # Create JSON array with directory info
@@ -56,141 +64,189 @@ jobs:
           all_dirs="[${all_dirs%,}]"
           echo "test-dirs=$all_dirs" >> $GITHUB_OUTPUT
 
-  prepare-build:
-    runs-on: blacksmith-2vcpu-ubuntu-2404-arm
-    steps:
-      - name: Checkout code
-        uses: actions/checkout@v4
-
-      - name: Setup Python
-        uses: actions/setup-python@v5
-        with:
-          python-version: "3.11"
-          cache: "pip"
-          cache-dependency-path: |
-            backend/requirements/default.txt
-            backend/requirements/dev.txt
-
-      - name: Install Python dependencies
-        run: |
-          python -m pip install --upgrade pip
-          pip install --retries 5 --timeout 30 -r backend/requirements/default.txt
-          pip install --retries 5 --timeout 30 -r backend/requirements/dev.txt
-
-      - name: Generate OpenAPI schema
-        working-directory: ./backend
-        env:
-          PYTHONPATH: "."
-        run: |
-          python scripts/onyx_openapi_schema.py --filename generated/openapi.json
-
-      - name: Generate OpenAPI Python client
-        working-directory: ./backend
-        run: |
-          docker run --rm \
-            -v "${{ github.workspace }}/backend/generated:/local" \
-            openapitools/openapi-generator-cli generate \
-            -i /local/openapi.json \
-            -g python \
-            -o /local/onyx_openapi_client \
-            --package-name onyx_openapi_client \
-            --skip-validate-spec \
-            --openapi-normalizer "SIMPLIFY_ONEOF_ANYOF=true,SET_OAS3_NULLABLE=true"
-
-      - name: Upload OpenAPI artifacts
-        uses: actions/upload-artifact@v4
-        with:
-          name: openapi-artifacts
-          path: backend/generated/
-
   build-backend-image:
-    runs-on: blacksmith-16vcpu-ubuntu-2404-arm
+    runs-on:
+      [
+        runs-on,
+        runner=1cpu-linux-arm64,
+        "run-id=${{ github.run_id }}-build-backend-image",
+        "extras=ecr-cache",
+      ]
+    timeout-minutes: 45
     steps:
+      - uses: runs-on/action@cd2b598b0515d39d78c38a02d529db87d2196d1e # ratchet:runs-on/action@v2
       - name: Checkout code
-        uses: actions/checkout@v4
-
-      - name: Login to Private Registry
-        uses: docker/login-action@v3
+        uses: actions/checkout@8e8c483db84b4bee98b60c0593521ed34d9990e8 # ratchet:actions/checkout@v6
         with:
-          registry: ${{ env.PRIVATE_REGISTRY }}
-          username: ${{ env.PRIVATE_REGISTRY_USERNAME }}
-          password: ${{ env.PRIVATE_REGISTRY_PASSWORD }}
+          persist-credentials: false
+
+      - name: Format branch name for cache
+        id: format-branch
+        env:
+          PR_NUMBER: ${{ github.event.pull_request.number }}
+          REF_NAME: ${{ github.ref_name }}
+        run: |
+          if [ -n "${PR_NUMBER}" ]; then
+            CACHE_SUFFIX="${PR_NUMBER}"
+          else
+            # shellcheck disable=SC2001
+            CACHE_SUFFIX=$(echo "${REF_NAME}" | sed 's/[^A-Za-z0-9._-]/-/g')
+          fi
+          echo "cache-suffix=${CACHE_SUFFIX}" >> $GITHUB_OUTPUT
 
       - name: Set up Docker Buildx
-        uses: useblacksmith/setup-docker-builder@v1
+        uses: docker/setup-buildx-action@e468171a9de216ec08956ac3ada2f0791b6bd435 # ratchet:docker/setup-buildx-action@v3
+
+      # needed for pulling Vespa, Redis, Postgres, and Minio images
+      # otherwise, we hit the "Unauthenticated users" limit
+      # https://docs.docker.com/docker-hub/usage/
+      - name: Login to Docker Hub
+        uses: docker/login-action@5e57cd118135c172c3672efd75eb46360885c0ef # ratchet:docker/login-action@v3
+        with:
+          username: ${{ secrets.DOCKER_USERNAME }}
+          password: ${{ secrets.DOCKER_TOKEN }}
 
       - name: Build and push Backend Docker image
-        uses: useblacksmith/build-push-action@v2
+        uses: docker/build-push-action@263435318d21b8e681c14492fe198d362a7d2c83 # ratchet:docker/build-push-action@v6
         with:
           context: ./backend
           file: ./backend/Dockerfile
-          platforms: linux/arm64
-          tags: ${{ env.PRIVATE_REGISTRY }}/integration-test-onyx-backend:test-${{ github.run_id }}
           push: true
-          outputs: type=registry
-          no-cache: true
+          tags: ${{ env.RUNS_ON_ECR_CACHE }}:integration-test-backend-test-${{ github.run_id }}
+          cache-from: |
+            type=registry,ref=${{ env.RUNS_ON_ECR_CACHE }}:backend-cache-${{ github.event.pull_request.head.sha || github.sha }}
+            type=registry,ref=${{ env.RUNS_ON_ECR_CACHE }}:backend-cache-${{ steps.format-branch.outputs.cache-suffix }}
+            type=registry,ref=${{ env.RUNS_ON_ECR_CACHE }}:backend-cache
+            type=registry,ref=onyxdotapp/onyx-backend:latest
+          cache-to: |
+            type=registry,ref=${{ env.RUNS_ON_ECR_CACHE }}:backend-cache-${{ github.event.pull_request.head.sha || github.sha }},mode=max
+            type=registry,ref=${{ env.RUNS_ON_ECR_CACHE }}:backend-cache-${{ steps.format-branch.outputs.cache-suffix }},mode=max
+            type=registry,ref=${{ env.RUNS_ON_ECR_CACHE }}:backend-cache,mode=max
+          no-cache: ${{ vars.DOCKER_NO_CACHE == 'true' }}
 
   build-model-server-image:
-    runs-on: blacksmith-16vcpu-ubuntu-2404-arm
+    runs-on:
+      [
+        runs-on,
+        runner=1cpu-linux-arm64,
+        "run-id=${{ github.run_id }}-build-model-server-image",
+        "extras=ecr-cache",
+      ]
+    timeout-minutes: 45
     steps:
+      - uses: runs-on/action@cd2b598b0515d39d78c38a02d529db87d2196d1e # ratchet:runs-on/action@v2
       - name: Checkout code
-        uses: actions/checkout@v4
-
-      - name: Login to Private Registry
-        uses: docker/login-action@v3
+        uses: actions/checkout@8e8c483db84b4bee98b60c0593521ed34d9990e8 # ratchet:actions/checkout@v6
         with:
-          registry: ${{ env.PRIVATE_REGISTRY }}
-          username: ${{ env.PRIVATE_REGISTRY_USERNAME }}
-          password: ${{ env.PRIVATE_REGISTRY_PASSWORD }}
+          persist-credentials: false
+
+      - name: Format branch name for cache
+        id: format-branch
+        env:
+          PR_NUMBER: ${{ github.event.pull_request.number }}
+          REF_NAME: ${{ github.ref_name }}
+        run: |
+          if [ -n "${PR_NUMBER}" ]; then
+            CACHE_SUFFIX="${PR_NUMBER}"
+          else
+            # shellcheck disable=SC2001
+            CACHE_SUFFIX=$(echo "${REF_NAME}" | sed 's/[^A-Za-z0-9._-]/-/g')
+          fi
+          echo "cache-suffix=${CACHE_SUFFIX}" >> $GITHUB_OUTPUT
 
       - name: Set up Docker Buildx
-        uses: useblacksmith/setup-docker-builder@v1
+        uses: docker/setup-buildx-action@e468171a9de216ec08956ac3ada2f0791b6bd435 # ratchet:docker/setup-buildx-action@v3
+
+      # needed for pulling Vespa, Redis, Postgres, and Minio images
+      # otherwise, we hit the "Unauthenticated users" limit
+      # https://docs.docker.com/docker-hub/usage/
+      - name: Login to Docker Hub
+        uses: docker/login-action@5e57cd118135c172c3672efd75eb46360885c0ef # ratchet:docker/login-action@v3
+        with:
+          username: ${{ secrets.DOCKER_USERNAME }}
+          password: ${{ secrets.DOCKER_TOKEN }}
 
       - name: Build and push Model Server Docker image
-        uses: useblacksmith/build-push-action@v2
+        uses: docker/build-push-action@263435318d21b8e681c14492fe198d362a7d2c83 # ratchet:docker/build-push-action@v6
         with:
           context: ./backend
           file: ./backend/Dockerfile.model_server
-          platforms: linux/arm64
-          tags: ${{ env.PRIVATE_REGISTRY }}/integration-test-onyx-model-server:test-${{ github.run_id }}
           push: true
-          outputs: type=registry
-          provenance: false
-          no-cache: true
+          tags: ${{ env.RUNS_ON_ECR_CACHE }}:integration-test-model-server-test-${{ github.run_id }}
+          cache-from: |
+            type=registry,ref=${{ env.RUNS_ON_ECR_CACHE }}:model-server-cache-${{ github.event.pull_request.head.sha || github.sha }}
+            type=registry,ref=${{ env.RUNS_ON_ECR_CACHE }}:model-server-cache-${{ steps.format-branch.outputs.cache-suffix }}
+            type=registry,ref=${{ env.RUNS_ON_ECR_CACHE }}:model-server-cache
+            type=registry,ref=onyxdotapp/onyx-model-server:latest
+          cache-to: |
+            type=registry,ref=${{ env.RUNS_ON_ECR_CACHE }}:model-server-cache-${{ github.event.pull_request.head.sha || github.sha }},mode=max
+            type=registry,ref=${{ env.RUNS_ON_ECR_CACHE }}:model-server-cache-${{ steps.format-branch.outputs.cache-suffix }},mode=max
+            type=registry,ref=${{ env.RUNS_ON_ECR_CACHE }}:model-server-cache,mode=max
 
   build-integration-image:
-    needs: prepare-build
-    runs-on: blacksmith-16vcpu-ubuntu-2404-arm
+    runs-on:
+      [
+        runs-on,
+        runner=2cpu-linux-arm64,
+        "run-id=${{ github.run_id }}-build-integration-image",
+        "extras=ecr-cache",
+      ]
+    timeout-minutes: 45
     steps:
+      - uses: runs-on/action@cd2b598b0515d39d78c38a02d529db87d2196d1e # ratchet:runs-on/action@v2
       - name: Checkout code
-        uses: actions/checkout@v4
-
-      - name: Login to Private Registry
-        uses: docker/login-action@v3
+        uses: actions/checkout@8e8c483db84b4bee98b60c0593521ed34d9990e8 # ratchet:actions/checkout@v6
         with:
-          registry: ${{ env.PRIVATE_REGISTRY }}
-          username: ${{ env.PRIVATE_REGISTRY_USERNAME }}
-          password: ${{ env.PRIVATE_REGISTRY_PASSWORD }}
+          persist-credentials: false
 
-      - name: Download OpenAPI artifacts
-        uses: actions/download-artifact@v4
-        with:
-          name: openapi-artifacts
-          path: backend/generated/
+      - name: Format branch name for cache
+        id: format-branch
+        env:
+          PR_NUMBER: ${{ github.event.pull_request.number }}
+          REF_NAME: ${{ github.ref_name }}
+        run: |
+          if [ -n "${PR_NUMBER}" ]; then
+            CACHE_SUFFIX="${PR_NUMBER}"
+          else
+            # shellcheck disable=SC2001
+            CACHE_SUFFIX=$(echo "${REF_NAME}" | sed 's/[^A-Za-z0-9._-]/-/g')
+          fi
+          echo "cache-suffix=${CACHE_SUFFIX}" >> $GITHUB_OUTPUT
 
       - name: Set up Docker Buildx
-        uses: useblacksmith/setup-docker-builder@v1
+        uses: docker/setup-buildx-action@e468171a9de216ec08956ac3ada2f0791b6bd435 # ratchet:docker/setup-buildx-action@v3
 
-      - name: Build and push integration test Docker image
-        uses: useblacksmith/build-push-action@v2
+      # needed for pulling openapitools/openapi-generator-cli
+      # otherwise, we hit the "Unauthenticated users" limit
+      # https://docs.docker.com/docker-hub/usage/
+      - name: Login to Docker Hub
+        uses: docker/login-action@5e57cd118135c172c3672efd75eb46360885c0ef # ratchet:docker/login-action@v3
         with:
-          context: ./backend
-          file: ./backend/tests/integration/Dockerfile
-          platforms: linux/arm64
-          tags: ${{ env.PRIVATE_REGISTRY }}/integration-test-onyx-integration:test-${{ github.run_id }}
-          push: true
-          outputs: type=registry
-          no-cache: true
+          username: ${{ secrets.DOCKER_USERNAME }}
+          password: ${{ secrets.DOCKER_TOKEN }}
+
+      - name: Build and push integration test image with Docker Bake
+        env:
+          INTEGRATION_REPOSITORY: ${{ env.RUNS_ON_ECR_CACHE }}
+          TAG: integration-test-${{ github.run_id }}
+          CACHE_SUFFIX: ${{ steps.format-branch.outputs.cache-suffix }}
+          HEAD_SHA: ${{ github.event.pull_request.head.sha || github.sha }}
+        run: |
+          docker buildx bake --push \
+            --set backend.cache-from=type=registry,ref=${RUNS_ON_ECR_CACHE}:backend-cache-${HEAD_SHA} \
+            --set backend.cache-from=type=registry,ref=${RUNS_ON_ECR_CACHE}:backend-cache-${CACHE_SUFFIX} \
+            --set backend.cache-from=type=registry,ref=${RUNS_ON_ECR_CACHE}:backend-cache \
+            --set backend.cache-from=type=registry,ref=onyxdotapp/onyx-backend:latest \
+            --set backend.cache-to=type=registry,ref=${RUNS_ON_ECR_CACHE}:backend-cache-${HEAD_SHA},mode=max \
+            --set backend.cache-to=type=registry,ref=${RUNS_ON_ECR_CACHE}:backend-cache-${CACHE_SUFFIX},mode=max \
+            --set backend.cache-to=type=registry,ref=${RUNS_ON_ECR_CACHE}:backend-cache,mode=max \
+            --set integration.cache-from=type=registry,ref=${RUNS_ON_ECR_CACHE}:integration-cache-${HEAD_SHA} \
+            --set integration.cache-from=type=registry,ref=${RUNS_ON_ECR_CACHE}:integration-cache-${CACHE_SUFFIX} \
+            --set integration.cache-from=type=registry,ref=${RUNS_ON_ECR_CACHE}:integration-cache \
+            --set integration.cache-to=type=registry,ref=${RUNS_ON_ECR_CACHE}:integration-cache-${HEAD_SHA},mode=max \
+            --set integration.cache-to=type=registry,ref=${RUNS_ON_ECR_CACHE}:integration-cache-${CACHE_SUFFIX},mode=max \
+            --set integration.cache-to=type=registry,ref=${RUNS_ON_ECR_CACHE}:integration-cache,mode=max \
+            integration
 
   integration-tests-mit:
     needs:
@@ -200,8 +256,12 @@ jobs:
         build-model-server-image,
         build-integration-image,
       ]
-    # See https://docs.blacksmith.sh/blacksmith-runners/overview
-    runs-on: blacksmith-8vcpu-ubuntu-2404-arm
+    runs-on:
+      - runs-on
+      - runner=4cpu-linux-arm64
+      - ${{ format('run-id={0}-integration-tests-mit-job-{1}', github.run_id, strategy['job-index']) }}
+      - extras=ecr-cache
+    timeout-minutes: 45
 
     strategy:
       fail-fast: false
@@ -209,56 +269,45 @@ jobs:
         test-dir: ${{ fromJson(needs.discover-test-dirs.outputs.test-dirs) }}
 
     steps:
+      - uses: runs-on/action@cd2b598b0515d39d78c38a02d529db87d2196d1e # ratchet:runs-on/action@v2
       - name: Checkout code
-        uses: actions/checkout@v4
-
-      - name: Login to Private Registry
-        uses: docker/login-action@v3
+        uses: actions/checkout@8e8c483db84b4bee98b60c0593521ed34d9990e8 # ratchet:actions/checkout@v6
         with:
-          registry: ${{ env.PRIVATE_REGISTRY }}
-          username: ${{ env.PRIVATE_REGISTRY_USERNAME }}
-          password: ${{ env.PRIVATE_REGISTRY_PASSWORD }}
+          persist-credentials: false
 
       # needed for pulling Vespa, Redis, Postgres, and Minio images
       # otherwise, we hit the "Unauthenticated users" limit
       # https://docs.docker.com/docker-hub/usage/
       - name: Login to Docker Hub
-        uses: docker/login-action@v3
+        uses: docker/login-action@5e57cd118135c172c3672efd75eb46360885c0ef # ratchet:docker/login-action@v3
         with:
           username: ${{ secrets.DOCKER_USERNAME }}
           password: ${{ secrets.DOCKER_TOKEN }}
 
-      - name: Pull Docker images
-        run: |
-          # Pull all images from registry in parallel
-          echo "Pulling Docker images in parallel..."
-          # Pull images from private registry
-          (docker pull --platform linux/arm64 ${{ env.PRIVATE_REGISTRY }}/integration-test-onyx-backend:test-${{ github.run_id }}) &
-          (docker pull --platform linux/arm64 ${{ env.PRIVATE_REGISTRY }}/integration-test-onyx-model-server:test-${{ github.run_id }}) &
-          (docker pull --platform linux/arm64 ${{ env.PRIVATE_REGISTRY }}/integration-test-onyx-integration:test-${{ github.run_id }}) &
-
-          # Wait for all background jobs to complete
-          wait
-          echo "All Docker images pulled successfully"
-
-          # Re-tag to remove registry prefix for docker-compose
-          docker tag ${{ env.PRIVATE_REGISTRY }}/integration-test-onyx-backend:test-${{ github.run_id }} onyxdotapp/onyx-backend:test
-          docker tag ${{ env.PRIVATE_REGISTRY }}/integration-test-onyx-model-server:test-${{ github.run_id }} onyxdotapp/onyx-model-server:test
-          docker tag ${{ env.PRIVATE_REGISTRY }}/integration-test-onyx-integration:test-${{ github.run_id }} onyxdotapp/onyx-integration:test
-
       # NOTE: Use pre-ping/null pool to reduce flakiness due to dropped connections
       # NOTE: don't need web server for integration tests
+      - name: Create .env file for Docker Compose
+        env:
+          ECR_CACHE: ${{ env.RUNS_ON_ECR_CACHE }}
+          RUN_ID: ${{ github.run_id }}
+        run: |
+          cat <<EOF > deployment/docker_compose/.env
+          AUTH_TYPE=basic
+          POSTGRES_POOL_PRE_PING=true
+          POSTGRES_USE_NULL_POOL=true
+          REQUIRE_EMAIL_VERIFICATION=false
+          DISABLE_TELEMETRY=true
+          ONYX_BACKEND_IMAGE=${ECR_CACHE}:integration-test-backend-test-${RUN_ID}
+          ONYX_MODEL_SERVER_IMAGE=${ECR_CACHE}:integration-test-model-server-test-${RUN_ID}
+          INTEGRATION_TESTS_MODE=true
+          MCP_SERVER_ENABLED=true
+          AUTO_LLM_UPDATE_INTERVAL_SECONDS=1
+          EOF
+
       - name: Start Docker containers
         run: |
           cd deployment/docker_compose
-          AUTH_TYPE=basic \
-          POSTGRES_POOL_PRE_PING=true \
-          POSTGRES_USE_NULL_POOL=true \
-          REQUIRE_EMAIL_VERIFICATION=false \
-          DISABLE_TELEMETRY=true \
-          IMAGE_TAG=test \
-          INTEGRATION_TESTS_MODE=true \
-          docker compose up \
+          docker compose -f docker-compose.yml -f docker-compose.dev.yml up \
             relational_db \
             index \
             cache \
@@ -270,39 +319,45 @@ jobs:
             -d
         id: start_docker
 
-      - name: Wait for service to be ready
+      - name: Wait for services to be ready
         run: |
           echo "Starting wait-for-service script..."
 
-          docker logs -f onyx-api_server-1 &
+          wait_for_service() {
+            local url=$1
+            local label=$2
+            local timeout=${3:-300}  # default 5 minutes
+            local start_time
+            start_time=$(date +%s)
 
-          start_time=$(date +%s)
-          timeout=300  # 5 minutes in seconds
+            while true; do
+              local current_time
+              current_time=$(date +%s)
+              local elapsed_time=$((current_time - start_time))
 
-          while true; do
-            current_time=$(date +%s)
-            elapsed_time=$((current_time - start_time))
-            
-            if [ $elapsed_time -ge $timeout ]; then
-              echo "Timeout reached. Service did not become ready in 5 minutes."
-              exit 1
-            fi
-            
-            # Use curl with error handling to ignore specific exit code 56
-            response=$(curl -s -o /dev/null -w "%{http_code}" http://localhost:8080/health || echo "curl_error")
-            
-            if [ "$response" = "200" ]; then
-              echo "Service is ready!"
-              break
-            elif [ "$response" = "curl_error" ]; then
-              echo "Curl encountered an error, possibly exit code 56. Continuing to retry..."
-            else
-              echo "Service not ready yet (HTTP status $response). Retrying in 5 seconds..."
-            fi
-            
-            sleep 5
-          done
-          echo "Finished waiting for service."
+              if [ $elapsed_time -ge $timeout ]; then
+                echo "Timeout reached. ${label} did not become ready in $timeout seconds."
+                exit 1
+              fi
+
+              local response
+              response=$(curl -s -o /dev/null -w "%{http_code}" "$url" || echo "curl_error")
+
+              if [ "$response" = "200" ]; then
+                echo "${label} is ready!"
+                break
+              elif [ "$response" = "curl_error" ]; then
+                echo "Curl encountered an error while checking ${label}. Retrying in 5 seconds..."
+              else
+                echo "${label} not ready yet (HTTP status $response). Retrying in 5 seconds..."
+              fi
+
+              sleep 5
+            done
+          }
+
+          wait_for_service "http://localhost:8080/health" "API server"
+          echo "Finished waiting for services."
 
       - name: Start Mock Services
         run: |
@@ -312,7 +367,7 @@ jobs:
 
       # NOTE: Use pre-ping/null to reduce flakiness due to dropped connections
       - name: Run Integration Tests for ${{ matrix.test-dir.name }}
-        uses: nick-fields/retry@v3
+        uses: nick-fields/retry@ce71cc2ab81d554ebbe88c79ab5975992d79ba08 # ratchet:nick-fields/retry@v3
         with:
           timeout_minutes: 20
           max_attempts: 3
@@ -333,13 +388,16 @@ jobs:
               -e REDIS_HOST=cache \
               -e API_SERVER_HOST=api_server \
               -e OPENAI_API_KEY=${OPENAI_API_KEY} \
+              -e EXA_API_KEY=${EXA_API_KEY} \
               -e SLACK_BOT_TOKEN=${SLACK_BOT_TOKEN} \
               -e CONFLUENCE_TEST_SPACE_URL=${CONFLUENCE_TEST_SPACE_URL} \
               -e CONFLUENCE_USER_NAME=${CONFLUENCE_USER_NAME} \
               -e CONFLUENCE_ACCESS_TOKEN=${CONFLUENCE_ACCESS_TOKEN} \
+              -e CONFLUENCE_ACCESS_TOKEN_SCOPED=${CONFLUENCE_ACCESS_TOKEN_SCOPED} \
               -e JIRA_BASE_URL=${JIRA_BASE_URL} \
               -e JIRA_USER_EMAIL=${JIRA_USER_EMAIL} \
               -e JIRA_API_TOKEN=${JIRA_API_TOKEN} \
+              -e JIRA_API_TOKEN_SCOPED=${JIRA_API_TOKEN_SCOPED} \
               -e PERM_SYNC_SHAREPOINT_CLIENT_ID=${PERM_SYNC_SHAREPOINT_CLIENT_ID} \
               -e PERM_SYNC_SHAREPOINT_PRIVATE_KEY="${PERM_SYNC_SHAREPOINT_PRIVATE_KEY}" \
               -e PERM_SYNC_SHAREPOINT_CERTIFICATE_PASSWORD=${PERM_SYNC_SHAREPOINT_CERTIFICATE_PASSWORD} \
@@ -347,7 +405,7 @@ jobs:
               -e TEST_WEB_HOSTNAME=test-runner \
               -e MOCK_CONNECTOR_SERVER_HOST=mock_connector_server \
               -e MOCK_CONNECTOR_SERVER_PORT=8001 \
-              onyxdotapp/onyx-integration:test \
+              ${{ env.RUNS_ON_ECR_CACHE }}:integration-test-${{ github.run_id }} \
               /app/tests/integration/${{ matrix.test-dir.path }}
 
       # ------------------------------------------------------------
@@ -366,31 +424,19 @@ jobs:
 
       - name: Upload logs
         if: always()
-        uses: actions/upload-artifact@v4
+        uses: actions/upload-artifact@330a01c490aca151604b8cf639adc76d48f6c5d4 # ratchet:actions/upload-artifact@v4
         with:
           name: docker-all-logs-${{ matrix.test-dir.name }}
           path: ${{ github.workspace }}/docker-compose.log
       # ------------------------------------------------------------
 
-      - name: Stop Docker containers
-        if: always()
-        run: |
-          cd deployment/docker_compose
-          docker compose down -v
-
-  
-  required: 
-    runs-on: blacksmith-2vcpu-ubuntu-2404-arm
+  required:
+    # NOTE: Github-hosted runners have about 20s faster queue times and are preferred here.
+    runs-on: ubuntu-slim
+    timeout-minutes: 45
     needs: [integration-tests-mit]
     if: ${{ always() }}
     steps:
-      - uses: actions/github-script@v7
-        with:
-          script: |
-            const needs = ${{ toJSON(needs) }};
-            const failed = Object.values(needs).some(n => n.result !== 'success');
-            if (failed) {
-              core.setFailed('One or more upstream jobs failed or were cancelled.');
-            } else {
-              core.notice('All required jobs succeeded.');
-            }
+      - name: Check job status
+        if: ${{ contains(needs.*.result, 'failure') || contains(needs.*.result, 'cancelled') || contains(needs.*.result, 'skipped') }}
+        run: exit 1

.github/workflows/pr-playwright-tests.yml

@@ -3,16 +3,20 @@ concurrency:
   group: Run-Playwright-Tests-${{ github.workflow }}-${{ github.head_ref || github.event.workflow_run.head_branch || github.run_id }}
   cancel-in-progress: true
 
-on: push
+on:
+  merge_group:
+  pull_request:
+    branches:
+      - main
+      - "release/**"
+  push:
+    tags:
+      - "v*.*.*"
+
+permissions:
+  contents: read
 
 env:
-  # AWS ECR Configuration
-  AWS_REGION: ${{ secrets.AWS_REGION || 'us-west-2' }}
-  ECR_REGISTRY: ${{ secrets.ECR_REGISTRY }}
-  AWS_ACCESS_KEY_ID: ${{ secrets.AWS_ACCESS_KEY_ID_ECR }}
-  AWS_SECRET_ACCESS_KEY: ${{ secrets.AWS_SECRET_ACCESS_KEY_ECR }}
-  BUILDX_NO_DEFAULT_ATTESTATIONS: 1
-  
   # Test Environment Variables
   OPENAI_API_KEY: ${{ secrets.OPENAI_API_KEY }}
   SLACK_BOT_TOKEN: ${{ secrets.SLACK_BOT_TOKEN }}
@@ -23,173 +27,295 @@ env:
   SLACK_CLIENT_ID: ${{ secrets.SLACK_CLIENT_ID }}
   SLACK_CLIENT_SECRET: ${{ secrets.SLACK_CLIENT_SECRET }}
 
+  # for MCP Oauth tests
+  MCP_OAUTH_CLIENT_ID: ${{ secrets.MCP_OAUTH_CLIENT_ID }}
+  MCP_OAUTH_CLIENT_SECRET: ${{ secrets.MCP_OAUTH_CLIENT_SECRET }}
+  MCP_OAUTH_ISSUER: ${{ secrets.MCP_OAUTH_ISSUER }}
+  MCP_OAUTH_JWKS_URI: ${{ secrets.MCP_OAUTH_JWKS_URI }}
+  MCP_OAUTH_USERNAME: ${{ vars.MCP_OAUTH_USERNAME }}
+  MCP_OAUTH_PASSWORD: ${{ secrets.MCP_OAUTH_PASSWORD }}
+
+  # for MCP API Key tests
+  MCP_API_KEY: test-api-key-12345
+  MCP_API_KEY_TEST_PORT: 8005
+  MCP_API_KEY_TEST_URL: http://host.docker.internal:8005/mcp
+  MCP_API_KEY_SERVER_HOST: 0.0.0.0
+  MCP_API_KEY_SERVER_PUBLIC_HOST: host.docker.internal
+
   MOCK_LLM_RESPONSE: true
+  MCP_TEST_SERVER_PORT: 8004
+  MCP_TEST_SERVER_URL: http://host.docker.internal:8004/mcp
+  MCP_TEST_SERVER_PUBLIC_URL: http://host.docker.internal:8004/mcp
+  MCP_TEST_SERVER_BIND_HOST: 0.0.0.0
+  MCP_TEST_SERVER_PUBLIC_HOST: host.docker.internal
+  MCP_SERVER_HOST: 0.0.0.0
+  MCP_SERVER_PUBLIC_HOST: host.docker.internal
+  MCP_SERVER_PUBLIC_URL: http://host.docker.internal:8004/mcp
 
 jobs:
   build-web-image:
-    runs-on: blacksmith-8vcpu-ubuntu-2404-arm
+    runs-on:
+      [
+        runs-on,
+        runner=4cpu-linux-arm64,
+        "run-id=${{ github.run_id }}-build-web-image",
+        "extras=ecr-cache",
+      ]
+    timeout-minutes: 45
     steps:
+      - uses: runs-on/action@cd2b598b0515d39d78c38a02d529db87d2196d1e # ratchet:runs-on/action@v2
+
       - name: Checkout code
-        uses: actions/checkout@v4
-
-      - name: Configure AWS credentials
-        uses: aws-actions/configure-aws-credentials@v4
+        uses: actions/checkout@8e8c483db84b4bee98b60c0593521ed34d9990e8 # ratchet:actions/checkout@v6
         with:
-          aws-access-key-id: ${{ env.AWS_ACCESS_KEY_ID }}
-          aws-secret-access-key: ${{ env.AWS_SECRET_ACCESS_KEY }}
-          aws-region: ${{ env.AWS_REGION }}
+          persist-credentials: false
 
-      - name: Login to Amazon ECR
-        id: login-ecr
-        uses: aws-actions/amazon-ecr-login@v2
+      - name: Format branch name for cache
+        id: format-branch
+        env:
+          PR_NUMBER: ${{ github.event.pull_request.number }}
+          REF_NAME: ${{ github.ref_name }}
+        run: |
+          if [ -n "${PR_NUMBER}" ]; then
+            CACHE_SUFFIX="${PR_NUMBER}"
+          else
+            # shellcheck disable=SC2001
+            CACHE_SUFFIX=$(echo "${REF_NAME}" | sed 's/[^A-Za-z0-9._-]/-/g')
+          fi
+          echo "cache-suffix=${CACHE_SUFFIX}" >> $GITHUB_OUTPUT
 
       - name: Set up Docker Buildx
-        uses: useblacksmith/setup-docker-builder@v1
+        uses: docker/setup-buildx-action@e468171a9de216ec08956ac3ada2f0791b6bd435 # ratchet:docker/setup-buildx-action@v3
 
-      - name: Build and push Web Docker image
-        uses: useblacksmith/build-push-action@v2
-        with:
-          context: ./web
-          file: ./web/Dockerfile
-          platforms: linux/arm64
-          tags: ${{ env.ECR_REGISTRY }}/integration-test-onyx-web-server:playwright-test-${{ github.run_id }}
-          provenance: false
-          sbom: false
-          push: true
-
-  build-backend-image:
-    runs-on: blacksmith-8vcpu-ubuntu-2404-arm
-    steps:
-      - name: Checkout code
-        uses: actions/checkout@v4
-
-      - name: Configure AWS credentials
-        uses: aws-actions/configure-aws-credentials@v4
-        with:
-          aws-access-key-id: ${{ env.AWS_ACCESS_KEY_ID }}
-          aws-secret-access-key: ${{ env.AWS_SECRET_ACCESS_KEY }}
-          aws-region: ${{ env.AWS_REGION }}
-
-      - name: Login to Amazon ECR
-        id: login-ecr
-        uses: aws-actions/amazon-ecr-login@v2
-
-      - name: Set up Docker Buildx
-        uses: useblacksmith/setup-docker-builder@v1
-
-      - name: Build and push Backend Docker image
-        uses: useblacksmith/build-push-action@v2
-        with:
-          context: ./backend
-          file: ./backend/Dockerfile
-          platforms: linux/arm64
-          tags: ${{ env.ECR_REGISTRY }}/integration-test-onyx-backend:playwright-test-${{ github.run_id }}
-          provenance: false
-          sbom: false
-          push: true
-
-  build-model-server-image:
-    runs-on: blacksmith-8vcpu-ubuntu-2404-arm
-    steps:
-      - name: Checkout code
-        uses: actions/checkout@v4
-
-      - name: Configure AWS credentials
-        uses: aws-actions/configure-aws-credentials@v4
-        with:
-          aws-access-key-id: ${{ env.AWS_ACCESS_KEY_ID }}
-          aws-secret-access-key: ${{ env.AWS_SECRET_ACCESS_KEY }}
-          aws-region: ${{ env.AWS_REGION }}
-
-      - name: Login to Amazon ECR
-        id: login-ecr
-        uses: aws-actions/amazon-ecr-login@v2
-
-      - name: Set up Docker Buildx
-        uses: useblacksmith/setup-docker-builder@v1
-
-      - name: Build and push Model Server Docker image
-        uses: useblacksmith/build-push-action@v2
-        with:
-          context: ./backend
-          file: ./backend/Dockerfile.model_server
-          platforms: linux/arm64
-          tags: ${{ env.ECR_REGISTRY }}/integration-test-onyx-model-server:playwright-test-${{ github.run_id }}
-          provenance: false
-          sbom: false
-          push: true
-
-  playwright-tests:
-    needs: [build-web-image, build-backend-image, build-model-server-image]
-    name: Playwright Tests
-    runs-on: blacksmith-8vcpu-ubuntu-2404-arm
-    steps:
-      - name: Checkout code
-        uses: actions/checkout@v4
-        with:
-          fetch-depth: 0
-
-      - name: Configure AWS credentials
-        uses: aws-actions/configure-aws-credentials@v4
-        with:
-          aws-access-key-id: ${{ env.AWS_ACCESS_KEY_ID }}
-          aws-secret-access-key: ${{ env.AWS_SECRET_ACCESS_KEY }}
-          aws-region: ${{ env.AWS_REGION }}
-
-      - name: Login to Amazon ECR
-        id: login-ecr
-        uses: aws-actions/amazon-ecr-login@v2
-
-      # needed for pulling Vespa, Redis, Postgres, and Minio images
-      # otherwise, we hit the "Unauthenticated users" limit
+      # needed for pulling external images otherwise, we hit the "Unauthenticated users" limit
       # https://docs.docker.com/docker-hub/usage/
       - name: Login to Docker Hub
-        uses: docker/login-action@v3
+        uses: docker/login-action@5e57cd118135c172c3672efd75eb46360885c0ef # ratchet:docker/login-action@v3
         with:
           username: ${{ secrets.DOCKER_USERNAME }}
           password: ${{ secrets.DOCKER_TOKEN }}
 
-      - name: Pull Docker images
+      - name: Build and push Web Docker image
+        uses: docker/build-push-action@263435318d21b8e681c14492fe198d362a7d2c83 # ratchet:docker/build-push-action@v6
+        with:
+          context: ./web
+          file: ./web/Dockerfile
+          platforms: linux/arm64
+          tags: ${{ env.RUNS_ON_ECR_CACHE }}:playwright-test-web-${{ github.run_id }}
+          push: true
+          cache-from: |
+            type=registry,ref=${{ env.RUNS_ON_ECR_CACHE }}:web-cache-${{ github.event.pull_request.head.sha || github.sha }}
+            type=registry,ref=${{ env.RUNS_ON_ECR_CACHE }}:web-cache-${{ steps.format-branch.outputs.cache-suffix }}
+            type=registry,ref=${{ env.RUNS_ON_ECR_CACHE }}:web-cache
+            type=registry,ref=onyxdotapp/onyx-web-server:latest
+          cache-to: |
+            type=registry,ref=${{ env.RUNS_ON_ECR_CACHE }}:web-cache-${{ github.event.pull_request.head.sha || github.sha }},mode=max
+            type=registry,ref=${{ env.RUNS_ON_ECR_CACHE }}:web-cache-${{ steps.format-branch.outputs.cache-suffix }},mode=max
+            type=registry,ref=${{ env.RUNS_ON_ECR_CACHE }}:web-cache,mode=max
+          no-cache: ${{ vars.DOCKER_NO_CACHE == 'true' }}
+
+  build-backend-image:
+    runs-on:
+      [
+        runs-on,
+        runner=1cpu-linux-arm64,
+        "run-id=${{ github.run_id }}-build-backend-image",
+        "extras=ecr-cache",
+      ]
+    timeout-minutes: 45
+    steps:
+      - uses: runs-on/action@cd2b598b0515d39d78c38a02d529db87d2196d1e # ratchet:runs-on/action@v2
+
+      - name: Checkout code
+        uses: actions/checkout@8e8c483db84b4bee98b60c0593521ed34d9990e8 # ratchet:actions/checkout@v6
+        with:
+          persist-credentials: false
+
+      - name: Format branch name for cache
+        id: format-branch
+        env:
+          PR_NUMBER: ${{ github.event.pull_request.number }}
+          REF_NAME: ${{ github.ref_name }}
         run: |
-          # Pull all images from ECR in parallel
-          echo "Pulling Docker images in parallel..."
-          (docker pull ${{ env.ECR_REGISTRY }}/integration-test-onyx-web-server:playwright-test-${{ github.run_id }}) &
-          (docker pull ${{ env.ECR_REGISTRY }}/integration-test-onyx-backend:playwright-test-${{ github.run_id }}) &
-          (docker pull ${{ env.ECR_REGISTRY }}/integration-test-onyx-model-server:playwright-test-${{ github.run_id }}) &
+          if [ -n "${PR_NUMBER}" ]; then
+            CACHE_SUFFIX="${PR_NUMBER}"
+          else
+            # shellcheck disable=SC2001
+            CACHE_SUFFIX=$(echo "${REF_NAME}" | sed 's/[^A-Za-z0-9._-]/-/g')
+          fi
+          echo "cache-suffix=${CACHE_SUFFIX}" >> $GITHUB_OUTPUT
 
-          # Wait for all background jobs to complete
-          wait
-          echo "All Docker images pulled successfully"
+      - name: Set up Docker Buildx
+        uses: docker/setup-buildx-action@e468171a9de216ec08956ac3ada2f0791b6bd435 # ratchet:docker/setup-buildx-action@v3
 
-          # Re-tag with expected names for docker-compose
-          docker tag ${{ env.ECR_REGISTRY }}/integration-test-onyx-web-server:playwright-test-${{ github.run_id }} onyxdotapp/onyx-web-server:test
-          docker tag ${{ env.ECR_REGISTRY }}/integration-test-onyx-backend:playwright-test-${{ github.run_id }} onyxdotapp/onyx-backend:test
-          docker tag ${{ env.ECR_REGISTRY }}/integration-test-onyx-model-server:playwright-test-${{ github.run_id }} onyxdotapp/onyx-model-server:test
+      # needed for pulling external images otherwise, we hit the "Unauthenticated users" limit
+      # https://docs.docker.com/docker-hub/usage/
+      - name: Login to Docker Hub
+        uses: docker/login-action@5e57cd118135c172c3672efd75eb46360885c0ef # ratchet:docker/login-action@v3
+        with:
+          username: ${{ secrets.DOCKER_USERNAME }}
+          password: ${{ secrets.DOCKER_TOKEN }}
+
+      - name: Build and push Backend Docker image
+        uses: docker/build-push-action@263435318d21b8e681c14492fe198d362a7d2c83 # ratchet:docker/build-push-action@v6
+        with:
+          context: ./backend
+          file: ./backend/Dockerfile
+          platforms: linux/arm64
+          tags: ${{ env.RUNS_ON_ECR_CACHE }}:playwright-test-backend-${{ github.run_id }}
+          push: true
+          cache-from: |
+            type=registry,ref=${{ env.RUNS_ON_ECR_CACHE }}:backend-cache-${{ github.event.pull_request.head.sha || github.sha }}
+            type=registry,ref=${{ env.RUNS_ON_ECR_CACHE }}:backend-cache-${{ steps.format-branch.outputs.cache-suffix }}
+            type=registry,ref=${{ env.RUNS_ON_ECR_CACHE }}:backend-cache
+            type=registry,ref=onyxdotapp/onyx-backend:latest
+          cache-to: |
+            type=registry,ref=${{ env.RUNS_ON_ECR_CACHE }}:backend-cache-${{ github.event.pull_request.head.sha || github.sha }},mode=max
+            type=registry,ref=${{ env.RUNS_ON_ECR_CACHE }}:backend-cache-${{ steps.format-branch.outputs.cache-suffix }},mode=max
+            type=registry,ref=${{ env.RUNS_ON_ECR_CACHE }}:backend-cache,mode=max
+          no-cache: ${{ vars.DOCKER_NO_CACHE == 'true' }}
+
+  build-model-server-image:
+    runs-on:
+      [
+        runs-on,
+        runner=1cpu-linux-arm64,
+        "run-id=${{ github.run_id }}-build-model-server-image",
+        "extras=ecr-cache",
+      ]
+    timeout-minutes: 45
+    steps:
+      - uses: runs-on/action@cd2b598b0515d39d78c38a02d529db87d2196d1e # ratchet:runs-on/action@v2
+
+      - name: Checkout code
+        uses: actions/checkout@8e8c483db84b4bee98b60c0593521ed34d9990e8 # ratchet:actions/checkout@v6
+        with:
+          persist-credentials: false
+
+      - name: Format branch name for cache
+        id: format-branch
+        env:
+          PR_NUMBER: ${{ github.event.pull_request.number }}
+          REF_NAME: ${{ github.ref_name }}
+        run: |
+          if [ -n "${PR_NUMBER}" ]; then
+            CACHE_SUFFIX="${PR_NUMBER}"
+          else
+            # shellcheck disable=SC2001
+            CACHE_SUFFIX=$(echo "${REF_NAME}" | sed 's/[^A-Za-z0-9._-]/-/g')
+          fi
+          echo "cache-suffix=${CACHE_SUFFIX}" >> $GITHUB_OUTPUT
+
+      - name: Set up Docker Buildx
+        uses: docker/setup-buildx-action@e468171a9de216ec08956ac3ada2f0791b6bd435 # ratchet:docker/setup-buildx-action@v3
+
+      # needed for pulling external images otherwise, we hit the "Unauthenticated users" limit
+      # https://docs.docker.com/docker-hub/usage/
+      - name: Login to Docker Hub
+        uses: docker/login-action@5e57cd118135c172c3672efd75eb46360885c0ef # ratchet:docker/login-action@v3
+        with:
+          username: ${{ secrets.DOCKER_USERNAME }}
+          password: ${{ secrets.DOCKER_TOKEN }}
+
+      - name: Build and push Model Server Docker image
+        uses: docker/build-push-action@263435318d21b8e681c14492fe198d362a7d2c83 # ratchet:docker/build-push-action@v6
+        with:
+          context: ./backend
+          file: ./backend/Dockerfile.model_server
+          platforms: linux/arm64
+          tags: ${{ env.RUNS_ON_ECR_CACHE }}:playwright-test-model-server-${{ github.run_id }}
+          push: true
+          cache-from: |
+            type=registry,ref=${{ env.RUNS_ON_ECR_CACHE }}:model-server-cache-${{ github.event.pull_request.head.sha || github.sha }}
+            type=registry,ref=${{ env.RUNS_ON_ECR_CACHE }}:model-server-cache-${{ steps.format-branch.outputs.cache-suffix }}
+            type=registry,ref=${{ env.RUNS_ON_ECR_CACHE }}:model-server-cache
+            type=registry,ref=onyxdotapp/onyx-model-server:latest
+          cache-to: |
+            type=registry,ref=${{ env.RUNS_ON_ECR_CACHE }}:model-server-cache-${{ github.event.pull_request.head.sha || github.sha }},mode=max
+            type=registry,ref=${{ env.RUNS_ON_ECR_CACHE }}:model-server-cache-${{ steps.format-branch.outputs.cache-suffix }},mode=max
+            type=registry,ref=${{ env.RUNS_ON_ECR_CACHE }}:model-server-cache,mode=max
+          no-cache: ${{ vars.DOCKER_NO_CACHE == 'true' }}
+
+  playwright-tests:
+    needs: [build-web-image, build-backend-image, build-model-server-image]
+    name: Playwright Tests (${{ matrix.project }})
+    runs-on:
+      - runs-on
+      - runner=8cpu-linux-arm64
+      - "run-id=${{ github.run_id }}-playwright-tests-${{ matrix.project }}"
+      - "extras=ecr-cache"
+      - volume=50gb
+    timeout-minutes: 45
+    strategy:
+      fail-fast: false
+      matrix:
+        project: [admin, no-auth, exclusive]
+    steps:
+      - uses: runs-on/action@cd2b598b0515d39d78c38a02d529db87d2196d1e # ratchet:runs-on/action@v2
+
+      - name: Checkout code
+        uses: actions/checkout@8e8c483db84b4bee98b60c0593521ed34d9990e8 # ratchet:actions/checkout@v6
+        with:
+          persist-credentials: false
 
       - name: Setup node
-        uses: actions/setup-node@v4
+        uses: actions/setup-node@395ad3262231945c25e8478fd5baf05154b1d79f # ratchet:actions/setup-node@v4
         with:
           node-version: 22
+          cache: "npm"
+          cache-dependency-path: ./web/package-lock.json
 
       - name: Install node dependencies
         working-directory: ./web
         run: npm ci
 
+      - name: Cache playwright cache
+        uses: runs-on/cache@50350ad4242587b6c8c2baa2e740b1bc11285ff4 # ratchet:runs-on/cache@v4
+        with:
+          path: ~/.cache/ms-playwright
+          key: ${{ runner.os }}-playwright-npm-${{ hashFiles('web/package-lock.json') }}
+          restore-keys: |
+            ${{ runner.os }}-playwright-npm-
+
       - name: Install playwright browsers
         working-directory: ./web
         run: npx playwright install --with-deps
 
+      - name: Create .env file for Docker Compose
+        env:
+          OPENAI_API_KEY_VALUE: ${{ env.OPENAI_API_KEY }}
+          EXA_API_KEY_VALUE: ${{ env.EXA_API_KEY }}
+          ECR_CACHE: ${{ env.RUNS_ON_ECR_CACHE }}
+          RUN_ID: ${{ github.run_id }}
+        run: |
+          cat <<EOF > deployment/docker_compose/.env
+          ENABLE_PAID_ENTERPRISE_EDITION_FEATURES=true
+          AUTH_TYPE=basic
+          GEN_AI_API_KEY=${OPENAI_API_KEY_VALUE}
+          EXA_API_KEY=${EXA_API_KEY_VALUE}
+          REQUIRE_EMAIL_VERIFICATION=false
+          DISABLE_TELEMETRY=true
+          ONYX_BACKEND_IMAGE=${ECR_CACHE}:playwright-test-backend-${RUN_ID}
+          ONYX_MODEL_SERVER_IMAGE=${ECR_CACHE}:playwright-test-model-server-${RUN_ID}
+          ONYX_WEB_SERVER_IMAGE=${ECR_CACHE}:playwright-test-web-${RUN_ID}
+          EOF
+          if [ "${{ matrix.project }}" = "no-auth" ]; then
+            echo "PLAYWRIGHT_FORCE_EMPTY_LLM_PROVIDERS=true" >> deployment/docker_compose/.env
+          fi
+
+      # needed for pulling Vespa, Redis, Postgres, and Minio images
+      # otherwise, we hit the "Unauthenticated users" limit
+      # https://docs.docker.com/docker-hub/usage/
+      - name: Login to Docker Hub
+        uses: docker/login-action@5e57cd118135c172c3672efd75eb46360885c0ef # ratchet:docker/login-action@v3
+        with:
+          username: ${{ secrets.DOCKER_USERNAME }}
+          password: ${{ secrets.DOCKER_TOKEN }}
+
       - name: Start Docker containers
         run: |
           cd deployment/docker_compose
-          ENABLE_PAID_ENTERPRISE_EDITION_FEATURES=true \
-          AUTH_TYPE=basic \
-          GEN_AI_API_KEY=${{ env.OPENAI_API_KEY }} \
-          EXA_API_KEY=${{ env.EXA_API_KEY }} \
-          REQUIRE_EMAIL_VERIFICATION=false \
-          DISABLE_TELEMETRY=true \
-          IMAGE_TAG=test \
-          docker compose up -d
+          docker compose -f docker-compose.yml -f docker-compose.dev.yml -f docker-compose.mcp-oauth-test.yml -f docker-compose.mcp-api-key-test.yml up -d
         id: start_docker
 
       - name: Wait for service to be ready
@@ -204,15 +330,15 @@ jobs:
           while true; do
             current_time=$(date +%s)
             elapsed_time=$((current_time - start_time))
-            
+
             if [ $elapsed_time -ge $timeout ]; then
               echo "Timeout reached. Service did not become ready in 5 minutes."
               exit 1
             fi
-            
+
             # Use curl with error handling to ignore specific exit code 56
             response=$(curl -s -o /dev/null -w "%{http_code}" http://localhost:8080/health || echo "curl_error")
-            
+
             if [ "$response" = "200" ]; then
               echo "Service is ready!"
               break
@@ -221,43 +347,129 @@ jobs:
             else
               echo "Service not ready yet (HTTP status $response). Retrying in 5 seconds..."
             fi
-            
+
             sleep 5
           done
           echo "Finished waiting for service."
 
+      - name: Wait for MCP OAuth mock server
+        run: |
+          echo "Waiting for MCP OAuth mock server on port ${MCP_TEST_SERVER_PORT:-8004}..."
+          start_time=$(date +%s)
+          timeout=120
+
+          while true; do
+            current_time=$(date +%s)
+            elapsed_time=$((current_time - start_time))
+
+            if [ $elapsed_time -ge $timeout ]; then
+              echo "Timeout reached. MCP OAuth mock server did not become ready in ${timeout}s."
+              exit 1
+            fi
+
+            if curl -sf "http://localhost:${MCP_TEST_SERVER_PORT:-8004}/healthz" > /dev/null; then
+              echo "MCP OAuth mock server is ready!"
+              break
+            fi
+
+            sleep 3
+          done
+
+      - name: Wait for MCP API Key mock server
+        run: |
+          echo "Waiting for MCP API Key mock server on port ${MCP_API_KEY_TEST_PORT:-8005}..."
+          start_time=$(date +%s)
+          timeout=120
+
+          while true; do
+            current_time=$(date +%s)
+            elapsed_time=$((current_time - start_time))
+
+            if [ $elapsed_time -ge $timeout ]; then
+              echo "Timeout reached. MCP API Key mock server did not become ready in ${timeout}s."
+              exit 1
+            fi
+
+            if curl -sf "http://localhost:${MCP_API_KEY_TEST_PORT:-8005}/healthz" > /dev/null; then
+              echo "MCP API Key mock server is ready!"
+              break
+            fi
+
+            sleep 3
+          done
+
+      - name: Wait for web server to be ready
+        run: |
+          echo "Waiting for web server on port 3000..."
+          start_time=$(date +%s)
+          timeout=120
+
+          while true; do
+            current_time=$(date +%s)
+            elapsed_time=$((current_time - start_time))
+
+            if [ $elapsed_time -ge $timeout ]; then
+              echo "Timeout reached. Web server did not become ready in ${timeout}s."
+              exit 1
+            fi
+
+            if curl -sf "http://localhost:3000/api/health" > /dev/null 2>&1 || \
+               curl -sf "http://localhost:3000/" > /dev/null 2>&1; then
+              echo "Web server is ready!"
+              break
+            fi
+
+            echo "Web server not ready yet. Retrying in 3 seconds..."
+            sleep 3
+          done
+
       - name: Run Playwright tests
         working-directory: ./web
-        run: npx playwright test
+        env:
+          PROJECT: ${{ matrix.project }}
+        run: |
+          # Create test-results directory to ensure it exists for artifact upload
+          mkdir -p test-results
+          if [ "${PROJECT}" = "no-auth" ]; then
+            export PLAYWRIGHT_FORCE_EMPTY_LLM_PROVIDERS=true
+          fi
+          npx playwright test --project ${PROJECT}
 
-      - uses: actions/upload-artifact@v4
+      - uses: actions/upload-artifact@330a01c490aca151604b8cf639adc76d48f6c5d4 # ratchet:actions/upload-artifact@v4
         if: always()
         with:
-          # Chromatic automatically defaults to the test-results directory.
-          # Replace with the path to your custom directory and adjust the CHROMATIC_ARCHIVE_LOCATION environment variable accordingly.
-          name: test-results
-          path: ./web/test-results
+          # Includes test results and trace.zip files
+          name: playwright-test-results-${{ matrix.project }}-${{ github.run_id }}
+          path: ./web/test-results/
           retention-days: 30
 
       # save before stopping the containers so the logs can be captured
       - name: Save Docker logs
         if: success() || failure()
+        env:
+          WORKSPACE: ${{ github.workspace }}
         run: |
           cd deployment/docker_compose
           docker compose logs > docker-compose.log
-          mv docker-compose.log ${{ github.workspace }}/docker-compose.log
+          mv docker-compose.log ${WORKSPACE}/docker-compose.log
 
       - name: Upload logs
         if: success() || failure()
-        uses: actions/upload-artifact@v4
+        uses: actions/upload-artifact@330a01c490aca151604b8cf639adc76d48f6c5d4 # ratchet:actions/upload-artifact@v4
         with:
-          name: docker-logs
+          name: docker-logs-${{ matrix.project }}-${{ github.run_id }}
           path: ${{ github.workspace }}/docker-compose.log
 
-      - name: Stop Docker containers
-        run: |
-          cd deployment/docker_compose
-          docker compose down -v
+  playwright-required:
+    # NOTE: Github-hosted runners have about 20s faster queue times and are preferred here.
+    runs-on: ubuntu-slim
+    timeout-minutes: 45
+    needs: [playwright-tests]
+    if: ${{ always() }}
+    steps:
+      - name: Check job status
+        if: ${{ contains(needs.*.result, 'failure') || contains(needs.*.result, 'cancelled') || contains(needs.*.result, 'skipped') }}
+        run: exit 1
 
 # NOTE: Chromatic UI diff testing is currently disabled.
 # We are using Playwright for local and CI testing without visual regression checks.
@@ -276,12 +488,12 @@ jobs:
 #     ]
 #   steps:
 #     - name: Checkout code
-#       uses: actions/checkout@v4
+#       uses: actions/checkout@8e8c483db84b4bee98b60c0593521ed34d9990e8 # ratchet:actions/checkout@v6
 #       with:
 #         fetch-depth: 0
 
 #     - name: Setup node
-#       uses: actions/setup-node@v4
+#       uses: actions/setup-node@395ad3262231945c25e8478fd5baf05154b1d79f # ratchet:actions/setup-node@v4
 #       with:
 #         node-version: 22
 
@@ -290,7 +502,7 @@ jobs:
 #       run: npm ci
 
 #     - name: Download Playwright test results
-#       uses: actions/download-artifact@v4
+#       uses: actions/download-artifact@d3f86a106a0bac45b974a628896c90dbdf5c8093 # ratchet:actions/download-artifact@v4
 #       with:
 #         name: test-results
 #         path: ./web/test-results

.github/workflows/pr-python-checks.yml

@@ -1,4 +1,7 @@
 name: Python Checks
+concurrency:
+  group: Python-Checks-${{ github.workflow }}-${{ github.head_ref || github.event.workflow_run.head_branch || github.run_id }}
+  cancel-in-progress: true
 
 on:
   merge_group:
@@ -6,62 +9,60 @@ on:
     branches:
       - main
       - 'release/**'
+  push:
+    tags:
+      - "v*.*.*"
+
+permissions:
+  contents: read
 
 jobs:
   mypy-check:
     # See https://runs-on.com/runners/linux/
-    runs-on: [runs-on,runner=8cpu-linux-x64,"run-id=${{ github.run_id }}"]
+    # Note: Mypy seems quite optimized for x64 compared to arm64.
+    # Similarly, mypy is single-threaded and incremental, so 2cpu is sufficient.
+    runs-on: [runs-on, runner=2cpu-linux-x64, "run-id=${{ github.run_id }}-mypy-check", "extras=s3-cache"]
+    timeout-minutes: 45
 
     steps:
-    - name: Checkout code
-      uses: actions/checkout@v4
+      - uses: runs-on/action@cd2b598b0515d39d78c38a02d529db87d2196d1e # ratchet:runs-on/action@v2
+      - name: Checkout code
+        uses: actions/checkout@8e8c483db84b4bee98b60c0593521ed34d9990e8 # ratchet:actions/checkout@v6
+        with:
+          persist-credentials: false
 
-    - name: Set up Python
-      uses: actions/setup-python@v5
-      with:
-        python-version: '3.11'
-        cache: 'pip'
-        cache-dependency-path: |
-          backend/requirements/default.txt
-          backend/requirements/dev.txt
-          backend/requirements/model_server.txt
-    - run: |
-        python -m pip install --upgrade pip
-        pip install --retries 5 --timeout 30 -r backend/requirements/default.txt
-        pip install --retries 5 --timeout 30 -r backend/requirements/dev.txt
-        pip install --retries 5 --timeout 30 -r backend/requirements/model_server.txt
+      - name: Setup Python and Install Dependencies
+        uses: ./.github/actions/setup-python-and-install-dependencies
+        with:
+          requirements: |
+            backend/requirements/default.txt
+            backend/requirements/dev.txt
+            backend/requirements/model_server.txt
+            backend/requirements/ee.txt
 
-    - name: Generate OpenAPI schema
-      working-directory: ./backend
-      env:
-        PYTHONPATH: "."
-      run: |
-        python scripts/onyx_openapi_schema.py --filename generated/openapi.json
+      - name: Generate OpenAPI schema and Python client
+        shell: bash
+        run: |
+          ods openapi all
 
-    - name: Generate OpenAPI Python client
-      working-directory: ./backend
-      run: |
-        docker run --rm \
-          -v "${{ github.workspace }}/backend/generated:/local" \
-          openapitools/openapi-generator-cli generate \
-          -i /local/openapi.json \
-          -g python \
-          -o /local/onyx_openapi_client \
-          --package-name onyx_openapi_client \
-          --skip-validate-spec \
-          --openapi-normalizer "SIMPLIFY_ONEOF_ANYOF=true,SET_OAS3_NULLABLE=true"
-            
-    - name: Run MyPy
-      run: |
-        cd backend
-        mypy .
+      - name: Cache mypy cache
+        if: ${{ vars.DISABLE_MYPY_CACHE != 'true' }}
+        uses: runs-on/cache@50350ad4242587b6c8c2baa2e740b1bc11285ff4 # ratchet:runs-on/cache@v4
+        with:
+          path: backend/.mypy_cache
+          key: mypy-${{ runner.os }}-${{ hashFiles('**/*.py', '**/*.pyi', 'backend/pyproject.toml') }}
+          restore-keys: |
+            mypy-${{ runner.os }}-
 
-    - name: Check import order with reorder-python-imports
-      run: |
-        cd backend
-        find ./onyx -name "*.py" | xargs reorder-python-imports --py311-plus
+      - name: Run MyPy
+        working-directory: ./backend
+        env:
+          MYPY_FORCE_COLOR: 1
+          TERM: xterm-256color
+        run: mypy .
 
-    - name: Check code formatting with Black
-      run: |
-        cd backend
-        black --check .
+      - name: Run MyPy (tools/)
+        env:
+          MYPY_FORCE_COLOR: 1
+          TERM: xterm-256color
+        run: mypy tools/

.github/workflows/pr-python-connector-tests.yml

@@ -1,30 +1,49 @@
 name: Connector Tests
+concurrency:
+  group: Connector-Tests-${{ github.workflow }}-${{ github.head_ref || github.event.workflow_run.head_branch || github.run_id }}
+  cancel-in-progress: true
 
 on:
   merge_group:
   pull_request:
     branches: [main]
+  push:
+    tags:
+      - "v*.*.*"
   schedule:
     # This cron expression runs the job daily at 16:00 UTC (9am PT)
     - cron: "0 16 * * *"
 
+permissions:
+  contents: read
+
 env:
   # AWS
   AWS_ACCESS_KEY_ID_DAILY_CONNECTOR_TESTS: ${{ secrets.AWS_ACCESS_KEY_ID_DAILY_CONNECTOR_TESTS }}
   AWS_SECRET_ACCESS_KEY_DAILY_CONNECTOR_TESTS: ${{ secrets.AWS_SECRET_ACCESS_KEY_DAILY_CONNECTOR_TESTS }}
 
+  # Cloudflare R2
+  R2_ACCOUNT_ID_DAILY_CONNECTOR_TESTS: ${{ vars.R2_ACCOUNT_ID_DAILY_CONNECTOR_TESTS }}
+  R2_ACCESS_KEY_ID_DAILY_CONNECTOR_TESTS: ${{ secrets.R2_ACCESS_KEY_ID_DAILY_CONNECTOR_TESTS }}
+  R2_SECRET_ACCESS_KEY_DAILY_CONNECTOR_TESTS: ${{ secrets.R2_SECRET_ACCESS_KEY_DAILY_CONNECTOR_TESTS }}
+
+  # Google Cloud Storage
+  GCS_ACCESS_KEY_ID_DAILY_CONNECTOR_TESTS: ${{ secrets.GCS_ACCESS_KEY_ID_DAILY_CONNECTOR_TESTS }}
+  GCS_SECRET_ACCESS_KEY_DAILY_CONNECTOR_TESTS: ${{ secrets.GCS_SECRET_ACCESS_KEY_DAILY_CONNECTOR_TESTS }}
+
   # Confluence
-  CONFLUENCE_TEST_SPACE_URL: ${{ secrets.CONFLUENCE_TEST_SPACE_URL }}
-  CONFLUENCE_TEST_SPACE: ${{ secrets.CONFLUENCE_TEST_SPACE }}
+  CONFLUENCE_TEST_SPACE_URL: ${{ vars.CONFLUENCE_TEST_SPACE_URL }}
+  CONFLUENCE_TEST_SPACE: ${{ vars.CONFLUENCE_TEST_SPACE }}
   CONFLUENCE_TEST_PAGE_ID: ${{ secrets.CONFLUENCE_TEST_PAGE_ID }}
-  CONFLUENCE_IS_CLOUD: ${{ secrets.CONFLUENCE_IS_CLOUD }}
-  CONFLUENCE_USER_NAME: ${{ secrets.CONFLUENCE_USER_NAME }}
+  CONFLUENCE_USER_NAME: ${{ vars.CONFLUENCE_USER_NAME }}
   CONFLUENCE_ACCESS_TOKEN: ${{ secrets.CONFLUENCE_ACCESS_TOKEN }}
+  CONFLUENCE_ACCESS_TOKEN_SCOPED: ${{ secrets.CONFLUENCE_ACCESS_TOKEN_SCOPED }}
 
   # Jira
   JIRA_BASE_URL: ${{ secrets.JIRA_BASE_URL }}
   JIRA_USER_EMAIL: ${{ secrets.JIRA_USER_EMAIL }}
   JIRA_API_TOKEN: ${{ secrets.JIRA_API_TOKEN }}
+  JIRA_API_TOKEN_SCOPED: ${{ secrets.JIRA_API_TOKEN_SCOPED }}
 
   # Gong
   GONG_ACCESS_KEY: ${{ secrets.GONG_ACCESS_KEY }}
@@ -54,22 +73,22 @@ env:
   HUBSPOT_ACCESS_TOKEN: ${{ secrets.HUBSPOT_ACCESS_TOKEN }}
 
   # IMAP
-  IMAP_HOST: ${{ secrets.IMAP_HOST }}
-  IMAP_USERNAME: ${{ secrets.IMAP_USERNAME }}
+  IMAP_HOST: ${{ vars.IMAP_HOST }}
+  IMAP_USERNAME: ${{ vars.IMAP_USERNAME }}
   IMAP_PASSWORD: ${{ secrets.IMAP_PASSWORD }}
-  IMAP_MAILBOXES: ${{ secrets.IMAP_MAILBOXES }}
+  IMAP_MAILBOXES: ${{ vars.IMAP_MAILBOXES }}
 
   # Airtable
-  AIRTABLE_TEST_BASE_ID: ${{ secrets.AIRTABLE_TEST_BASE_ID }}
-  AIRTABLE_TEST_TABLE_ID: ${{ secrets.AIRTABLE_TEST_TABLE_ID }}
-  AIRTABLE_TEST_TABLE_NAME: ${{ secrets.AIRTABLE_TEST_TABLE_NAME }}
+  AIRTABLE_TEST_BASE_ID: ${{ vars.AIRTABLE_TEST_BASE_ID }}
+  AIRTABLE_TEST_TABLE_ID: ${{ vars.AIRTABLE_TEST_TABLE_ID }}
+  AIRTABLE_TEST_TABLE_NAME: ${{ vars.AIRTABLE_TEST_TABLE_NAME }}
   AIRTABLE_ACCESS_TOKEN: ${{ secrets.AIRTABLE_ACCESS_TOKEN }}
 
   # Sharepoint
-  SHAREPOINT_CLIENT_ID: ${{ secrets.SHAREPOINT_CLIENT_ID }}
+  SHAREPOINT_CLIENT_ID: ${{ vars.SHAREPOINT_CLIENT_ID }}
   SHAREPOINT_CLIENT_SECRET: ${{ secrets.SHAREPOINT_CLIENT_SECRET }}
-  SHAREPOINT_CLIENT_DIRECTORY_ID: ${{ secrets.SHAREPOINT_CLIENT_DIRECTORY_ID }}
-  SHAREPOINT_SITE: ${{ secrets.SHAREPOINT_SITE }}
+  SHAREPOINT_CLIENT_DIRECTORY_ID: ${{ vars.SHAREPOINT_CLIENT_DIRECTORY_ID }}
+  SHAREPOINT_SITE: ${{ vars.SHAREPOINT_SITE }}
 
   # Github
   ACCESS_TOKEN_GITHUB: ${{ secrets.ACCESS_TOKEN_GITHUB }}
@@ -96,36 +115,66 @@ env:
   TEAMS_DIRECTORY_ID: ${{ secrets.TEAMS_DIRECTORY_ID }}
   TEAMS_SECRET: ${{ secrets.TEAMS_SECRET }}
 
+  # Bitbucket
+  BITBUCKET_WORKSPACE: ${{ secrets.BITBUCKET_WORKSPACE }}
+  BITBUCKET_REPOSITORIES: ${{ secrets.BITBUCKET_REPOSITORIES }}
+  BITBUCKET_PROJECTS: ${{ secrets.BITBUCKET_PROJECTS }}
+  BITBUCKET_EMAIL: ${{ vars.BITBUCKET_EMAIL }}
+  BITBUCKET_API_TOKEN: ${{ secrets.BITBUCKET_API_TOKEN }}
+
+  # Fireflies
+  FIREFLIES_API_KEY: ${{ secrets.FIREFLIES_API_KEY }}
+
 jobs:
   connectors-check:
     # See https://runs-on.com/runners/linux/
-    runs-on: [runs-on, runner=8cpu-linux-x64, "run-id=${{ github.run_id }}"]
+    runs-on: [runs-on, runner=8cpu-linux-x64, "run-id=${{ github.run_id }}-connectors-check", "extras=s3-cache"]
+    timeout-minutes: 45
 
     env:
       PYTHONPATH: ./backend
+      DISABLE_TELEMETRY: "true"
 
     steps:
-      - name: Checkout code
-        uses: actions/checkout@v4
+      - uses: runs-on/action@cd2b598b0515d39d78c38a02d529db87d2196d1e # ratchet:runs-on/action@v2
 
-      - name: Set up Python
-        uses: actions/setup-python@v5
+      - name: Checkout code
+        uses: actions/checkout@8e8c483db84b4bee98b60c0593521ed34d9990e8 # ratchet:actions/checkout@v6
         with:
-          python-version: "3.11"
-          cache: "pip"
-          cache-dependency-path: |
+          persist-credentials: false
+
+      - name: Setup Python and Install Dependencies
+        uses: ./.github/actions/setup-python-and-install-dependencies
+        with:
+          requirements: |
             backend/requirements/default.txt
             backend/requirements/dev.txt
 
-      - name: Install Dependencies
-        run: |
-          python -m pip install --upgrade pip
-          pip install --retries 5 --timeout 30 -r backend/requirements/default.txt
-          pip install --retries 5 --timeout 30 -r backend/requirements/dev.txt
-          playwright install chromium
-          playwright install-deps chromium
+      - name: Setup Playwright
+        uses: ./.github/actions/setup-playwright
 
-      - name: Run Tests
+      - name: Detect Connector changes
+        id: changes
+        uses: dorny/paths-filter@de90cc6fb38fc0963ad72b210f1f284cd68cea36 # ratchet:dorny/paths-filter@v3
+        with:
+          filters: |
+            hubspot:
+              - 'backend/onyx/connectors/hubspot/**'
+              - 'backend/tests/daily/connectors/hubspot/**'
+              - 'uv.lock'
+            salesforce:
+              - 'backend/onyx/connectors/salesforce/**'
+              - 'backend/tests/daily/connectors/salesforce/**'
+              - 'uv.lock'
+            github:
+              - 'backend/onyx/connectors/github/**'
+              - 'backend/tests/daily/connectors/github/**'
+              - 'uv.lock'
+            file_processing:
+              - 'backend/onyx/file_processing/**'
+              - 'uv.lock'
+
+      - name: Run Tests (excluding HubSpot, Salesforce, GitHub, and Coda)
         shell: script -q -e -c "bash --noprofile --norc -eo pipefail {0}"
         run: |
           py.test \
@@ -135,14 +184,59 @@ jobs:
             -o junit_family=xunit2 \
             -xv \
             --ff \
-            backend/tests/daily/connectors
+            backend/tests/daily/connectors \
+            --ignore backend/tests/daily/connectors/hubspot \
+            --ignore backend/tests/daily/connectors/salesforce \
+            --ignore backend/tests/daily/connectors/github \
+            --ignore backend/tests/daily/connectors/coda
+
+      - name: Run HubSpot Connector Tests
+        if: ${{ github.event_name == 'schedule' || steps.changes.outputs.hubspot == 'true' || steps.changes.outputs.file_processing == 'true' }}
+        shell: script -q -e -c "bash --noprofile --norc -eo pipefail {0}"
+        run: |
+          py.test \
+            -n 8 \
+            --dist loadfile \
+            --durations=8 \
+            -o junit_family=xunit2 \
+            -xv \
+            --ff \
+            backend/tests/daily/connectors/hubspot
+
+      - name: Run Salesforce Connector Tests
+        if: ${{ github.event_name == 'schedule' || steps.changes.outputs.salesforce == 'true' || steps.changes.outputs.file_processing == 'true' }}
+        shell: script -q -e -c "bash --noprofile --norc -eo pipefail {0}"
+        run: |
+          py.test \
+            -n 8 \
+            --dist loadfile \
+            --durations=8 \
+            -o junit_family=xunit2 \
+            -xv \
+            --ff \
+            backend/tests/daily/connectors/salesforce
+
+      - name: Run GitHub Connector Tests
+        if: ${{ github.event_name == 'schedule' || steps.changes.outputs.github == 'true' || steps.changes.outputs.file_processing == 'true' }}
+        shell: script -q -e -c "bash --noprofile --norc -eo pipefail {0}"
+        run: |
+          py.test \
+            -n 8 \
+            --dist loadfile \
+            --durations=8 \
+            -o junit_family=xunit2 \
+            -xv \
+            --ff \
+            backend/tests/daily/connectors/github
 
       - name: Alert on Failure
         if: failure() && github.event_name == 'schedule'
         env:
           SLACK_WEBHOOK: ${{ secrets.SLACK_WEBHOOK }}
+          REPO: ${{ github.repository }}
+          RUN_ID: ${{ github.run_id }}
         run: |
           curl -X POST \
             -H 'Content-type: application/json' \
-            --data '{"text":"Scheduled Connector Tests failed! Check the run at: https://github.com/${{ github.repository }}/actions/runs/${{ github.run_id }}"}' \
+            --data "{\"text\":\"Scheduled Connector Tests failed! Check the run at: https://github.com/${REPO}/actions/runs/${RUN_ID}\"}" \
             $SLACK_WEBHOOK

.github/workflows/pr-python-model-tests.yml

@@ -10,12 +10,15 @@ on:
         description: 'Branch to run the workflow on'
         required: false
         default: 'main'
-        
+
+permissions:
+  contents: read
+
 env:
   # Bedrock
   AWS_ACCESS_KEY_ID: ${{ secrets.AWS_ACCESS_KEY_ID }}
   AWS_SECRET_ACCESS_KEY: ${{ secrets.AWS_SECRET_ACCESS_KEY }}
-  AWS_REGION_NAME: ${{ secrets.AWS_REGION_NAME }}
+  AWS_REGION_NAME: ${{ vars.AWS_REGION_NAME }}
 
   # API keys for testing
   COHERE_API_KEY: ${{ secrets.COHERE_API_KEY }}
@@ -23,22 +26,25 @@ env:
   LITELLM_API_URL: ${{ secrets.LITELLM_API_URL }}
   OPENAI_API_KEY: ${{ secrets.OPENAI_API_KEY }}
   AZURE_API_KEY: ${{ secrets.AZURE_API_KEY }}
-  AZURE_API_URL: ${{ secrets.AZURE_API_URL }}
+  AZURE_API_URL: ${{ vars.AZURE_API_URL }}
 
 jobs:
   model-check:
     # See https://runs-on.com/runners/linux/
-    runs-on: [runs-on,runner=8cpu-linux-x64,"run-id=${{ github.run_id }}"]
+    runs-on: [runs-on,runner=8cpu-linux-x64,"run-id=${{ github.run_id }}-model-check"]
+    timeout-minutes: 45
 
     env:
       PYTHONPATH: ./backend
 
     steps:
       - name: Checkout code
-        uses: actions/checkout@v4
+        uses: actions/checkout@8e8c483db84b4bee98b60c0593521ed34d9990e8 # ratchet:actions/checkout@v6
+        with:
+          persist-credentials: false
 
       - name: Login to Docker Hub
-        uses: docker/login-action@v3
+        uses: docker/login-action@5e57cd118135c172c3672efd75eb46360885c0ef # ratchet:docker/login-action@v3
         with:
           username: ${{ secrets.DOCKER_USERNAME }}
           password: ${{ secrets.DOCKER_TOKEN }}
@@ -53,9 +59,9 @@ jobs:
         run: |
           docker pull onyxdotapp/onyx-model-server:latest
           docker tag onyxdotapp/onyx-model-server:latest onyxdotapp/onyx-model-server:test
-          
+
       - name: Set up Python
-        uses: actions/setup-python@v5
+        uses: actions/setup-python@83679a892e2d95755f2dac6acb0bfd1e9ac5d548 # ratchet:actions/setup-python@v6
         with:
           python-version: "3.11"
           cache: "pip"
@@ -90,15 +96,15 @@ jobs:
           while true; do
             current_time=$(date +%s)
             elapsed_time=$((current_time - start_time))
-            
+
             if [ $elapsed_time -ge $timeout ]; then
               echo "Timeout reached. Service did not become ready in 5 minutes."
               exit 1
             fi
-            
+
             # Use curl with error handling to ignore specific exit code 56
             response=$(curl -s -o /dev/null -w "%{http_code}" http://localhost:9000/api/health || echo "curl_error")
-            
+
             if [ "$response" = "200" ]; then
               echo "Service is ready!"
               break
@@ -107,11 +113,11 @@ jobs:
             else
               echo "Service not ready yet (HTTP status $response). Retrying in 5 seconds..."
             fi
-            
+
             sleep 5
           done
           echo "Finished waiting for service."
-          
+
       - name: Run Tests
         shell: script -q -e -c "bash --noprofile --norc -eo pipefail {0}"
         run: |
@@ -122,12 +128,14 @@ jobs:
         if: failure() && github.event_name == 'schedule'
         env:
           SLACK_WEBHOOK: ${{ secrets.SLACK_WEBHOOK }}
+          REPO: ${{ github.repository }}
+          RUN_ID: ${{ github.run_id }}
         run: |
           curl -X POST \
             -H 'Content-type: application/json' \
-            --data '{"text":"Scheduled Model Tests failed! Check the run at: https://github.com/${{ github.repository }}/actions/runs/${{ github.run_id }}"}' \
+            --data "{\"text\":\"Scheduled Model Tests failed! Check the run at: https://github.com/${REPO}/actions/runs/${RUN_ID}\"}" \
             $SLACK_WEBHOOK
-            
+
       - name: Dump all-container logs (optional)
         if: always()
         run: |
@@ -136,14 +144,7 @@ jobs:
 
       - name: Upload logs
         if: always()
-        uses: actions/upload-artifact@v4
+        uses: actions/upload-artifact@330a01c490aca151604b8cf639adc76d48f6c5d4 # ratchet:actions/upload-artifact@v4
         with:
           name: docker-all-logs
           path: ${{ github.workspace }}/docker-compose.log
-          
-      - name: Stop Docker containers
-        if: always()
-        run: |
-          cd deployment/docker_compose
-          docker compose -f docker-compose.model-server-test.yml down -v
-          

.github/workflows/pr-python-tests.yml

@@ -1,4 +1,7 @@
 name: Python Unit Tests
+concurrency:
+  group: Python-Unit-Tests-${{ github.workflow }}-${{ github.head_ref || github.event.workflow_run.head_branch || github.run_id }}
+  cancel-in-progress: true
 
 on:
   merge_group:
@@ -6,39 +9,41 @@ on:
     branches:
       - main
       - 'release/**'
+  push:
+    tags:
+      - "v*.*.*"
+
+permissions:
+  contents: read
 
 jobs:
   backend-check:
     # See https://runs-on.com/runners/linux/
-    runs-on: [runs-on,runner=8cpu-linux-x64,"run-id=${{ github.run_id }}"]
+    runs-on: [runs-on, runner=2cpu-linux-arm64, "run-id=${{ github.run_id }}-backend-check"]
+    timeout-minutes: 45
+
 
     env:
       PYTHONPATH: ./backend
       REDIS_CLOUD_PYTEST_PASSWORD: ${{ secrets.REDIS_CLOUD_PYTEST_PASSWORD }}
-      SF_USERNAME: ${{ secrets.SF_USERNAME }}
-      SF_PASSWORD: ${{ secrets.SF_PASSWORD }}
-      SF_SECURITY_TOKEN: ${{ secrets.SF_SECURITY_TOKEN }}
-      
-    steps:
-    - name: Checkout code
-      uses: actions/checkout@v4
+      DISABLE_TELEMETRY: "true"
 
-    - name: Set up Python
-      uses: actions/setup-python@v5
+    steps:
+    - uses: runs-on/action@cd2b598b0515d39d78c38a02d529db87d2196d1e # ratchet:runs-on/action@v2
+
+    - name: Checkout code
+      uses: actions/checkout@8e8c483db84b4bee98b60c0593521ed34d9990e8 # ratchet:actions/checkout@v6
       with:
-        python-version: '3.11'
-        cache: 'pip'
-        cache-dependency-path: |
+        persist-credentials: false
+
+    - name: Setup Python and Install Dependencies
+      uses: ./.github/actions/setup-python-and-install-dependencies
+      with:
+        requirements: |
           backend/requirements/default.txt
           backend/requirements/dev.txt
           backend/requirements/model_server.txt
-
-    - name: Install Dependencies
-      run: |
-        python -m pip install --upgrade pip
-        pip install --retries 5 --timeout 30 -r backend/requirements/default.txt
-        pip install --retries 5 --timeout 30 -r backend/requirements/dev.txt
-        pip install --retries 5 --timeout 30 -r backend/requirements/model_server.txt
+          backend/requirements/ee.txt
 
     - name: Run Tests
       shell: script -q -e -c "bash --noprofile --norc -eo pipefail {0}"

.github/workflows/pr-quality-checks.yml

@@ -6,18 +6,44 @@ concurrency:
 on:
   merge_group:
   pull_request: null
+  push:
+    branches:
+      - main
+    tags:
+      - "v*.*.*"
+
+permissions:
+  contents: read
 
 jobs:
   quality-checks:
-    # See https://runs-on.com/runners/linux/
-    runs-on: [runs-on,runner=8cpu-linux-x64,"run-id=${{ github.run_id }}"]
+    runs-on: ubuntu-latest
+    timeout-minutes: 45
     steps:
-      - uses: actions/checkout@v4
+      - uses: actions/checkout@8e8c483db84b4bee98b60c0593521ed34d9990e8 # ratchet:actions/checkout@v6
         with:
           fetch-depth: 0
-      - uses: actions/setup-python@v5
+          persist-credentials: false
+      - uses: actions/setup-python@83679a892e2d95755f2dac6acb0bfd1e9ac5d548 # ratchet:actions/setup-python@v6
         with:
           python-version: "3.11"
-      - uses: pre-commit/action@v3.0.1
+      - name: Setup Terraform
+        uses: hashicorp/setup-terraform@b9cd54a3c349d3f38e8881555d616ced269862dd # ratchet:hashicorp/setup-terraform@v3
+      - name: Setup node
+        uses: actions/setup-node@395ad3262231945c25e8478fd5baf05154b1d79f # ratchet:actions/setup-node@v6
+        with: # zizmor: ignore[cache-poisoning]
+          node-version: 22
+          cache: "npm"
+          cache-dependency-path: ./web/package-lock.json
+      - name: Install node dependencies
+        working-directory: ./web
+        run: npm ci
+      - uses: j178/prek-action@91fd7d7cf70ae1dee9f4f44e7dfa5d1073fe6623 # ratchet:j178/prek-action@v1
         with:
-          extra_args: ${{ github.event_name == 'pull_request' && format('--from-ref {0} --to-ref {1}', github.event.pull_request.base.sha, github.event.pull_request.head.sha) || '' }}
+          prek-version: '0.2.21'
+          extra-args: ${{ github.event_name == 'pull_request' && format('--from-ref {0} --to-ref {1}', github.event.pull_request.base.sha, github.event.pull_request.head.sha) || github.event_name == 'merge_group' && format('--from-ref {0} --to-ref {1}', github.event.merge_group.base_sha, github.event.merge_group.head_sha) || github.ref_name == 'main' && '--all-files' || '' }}
+      - name: Check Actions
+        uses: giner/check-actions@28d366c7cbbe235f9624a88aa31a628167eee28c # ratchet:giner/check-actions@v1.0.1
+        with:
+          check_permissions: false
+          check_versions: false

.github/workflows/release-devtools.yml

@@ -0,0 +1,41 @@
+name: Release Devtools
+
+on:
+  push:
+    tags:
+      - "ods/v*.*.*"
+
+jobs:
+  pypi:
+    runs-on: ubuntu-latest
+    environment:
+      name: release-devtools
+    permissions:
+      id-token: write
+    timeout-minutes: 10
+    strategy:
+      matrix:
+        os-arch:
+          - { goos: "linux", goarch: "amd64" }
+          - { goos: "linux", goarch: "arm64" }
+          - { goos: "windows", goarch: "amd64" }
+          - { goos: "windows", goarch: "arm64" }
+          - { goos: "darwin", goarch: "amd64" }
+          - { goos: "darwin", goarch: "arm64" }
+          - { goos: "", goarch: "" }
+    steps:
+      - uses: actions/checkout@8e8c483db84b4bee98b60c0593521ed34d9990e8 # ratchet:actions/checkout@v6
+        with:
+          persist-credentials: false
+          fetch-depth: 0
+      - uses: astral-sh/setup-uv@ed21f2f24f8dd64503750218de024bcf64c7250a # ratchet:astral-sh/setup-uv@v7
+        with:
+          enable-cache: false
+          version: "0.9.9"
+      - run: |
+          GOOS="${{ matrix.os-arch.goos }}" \
+          GOARCH="${{ matrix.os-arch.goarch }}" \
+          uv build --wheel
+        working-directory: tools/ods
+      - run: uv publish
+        working-directory: tools/ods

.github/workflows/sync_foss.yml

@@ -0,0 +1,49 @@
+name: Sync FOSS Repo
+
+on:
+  schedule:
+    # Run daily at 3am PT (11am UTC during PST)
+    - cron: '0 11 * * *'
+  workflow_dispatch:
+
+jobs:
+  sync-foss:
+    runs-on: ubuntu-latest
+    timeout-minutes: 45
+    permissions:
+      contents: read
+    steps:
+      - name: Checkout main Onyx repo
+        uses: actions/checkout@8e8c483db84b4bee98b60c0593521ed34d9990e8 # ratchet:actions/checkout@v6
+        with:
+          fetch-depth: 0
+          persist-credentials: false
+
+      - name: Install git-filter-repo
+        run: |
+          sudo apt-get update && sudo apt-get install -y git-filter-repo
+
+      - name: Configure SSH for deploy key
+        env:
+          FOSS_REPO_DEPLOY_KEY: ${{ secrets.FOSS_REPO_DEPLOY_KEY }}
+        run: |
+          mkdir -p ~/.ssh
+          echo "$FOSS_REPO_DEPLOY_KEY" > ~/.ssh/id_ed25519
+          chmod 600 ~/.ssh/id_ed25519
+          ssh-keyscan github.com >> ~/.ssh/known_hosts
+
+      - name: Set Git config
+        run: |
+          git config --global user.name "onyx-bot"
+          git config --global user.email "bot@onyx.app"
+
+      - name: Build FOSS version
+        run: bash backend/scripts/make_foss_repo.sh
+
+      - name: Push to FOSS repo
+        env:
+          FOSS_REPO_URL: git@github.com:onyx-dot-app/onyx-foss.git
+        run: |
+          cd /tmp/foss_repo
+          git remote add public "$FOSS_REPO_URL"
+          git push --force public main

.github/workflows/tag-nightly.yml

@@ -3,27 +3,30 @@ name: Nightly Tag Push
 on:
   schedule:
     - cron: "0 10 * * *" # Runs every day at 2 AM PST / 3 AM PDT / 10 AM UTC
+  workflow_dispatch:
 
 permissions:
   contents: write # Allows pushing tags to the repository
 
 jobs:
   create-and-push-tag:
-    runs-on: [runs-on, runner=2cpu-linux-x64, "run-id=${{ github.run_id }}"]
+    runs-on: ubuntu-slim
+    timeout-minutes: 45
 
     steps:
       # actions using GITHUB_TOKEN cannot trigger another workflow, but we do want this to trigger docker pushes
       # see https://github.com/orgs/community/discussions/27028#discussioncomment-3254367 for the workaround we
       # implement here which needs an actual user's deploy key
       - name: Checkout code
-        uses: actions/checkout@v4
+        uses: actions/checkout@8e8c483db84b4bee98b60c0593521ed34d9990e8 # ratchet:actions/checkout@v6
         with:
-          ssh-key: "${{ secrets.RKUO_DEPLOY_KEY }}"
+          ssh-key: "${{ secrets.DEPLOY_KEY }}"
+          persist-credentials: true
 
       - name: Set up Git user
         run: |
-          git config user.name "Richard Kuo [bot]"
-          git config user.email "rkuo[bot]@onyx.app"
+          git config user.name "Onyx Bot [bot]"
+          git config user.email "onyx-bot[bot]@onyx.app"
 
       - name: Check for existing nightly tag
         id: check_tag
@@ -51,3 +54,12 @@ jobs:
         run: |
           TAG_NAME="nightly-latest-$(date +'%Y%m%d')"
           git push origin $TAG_NAME
+
+      - name: Send Slack notification
+        if: failure()
+        uses: ./.github/actions/slack-notify
+        with:
+          webhook-url: ${{ secrets.MONITOR_DEPLOYMENTS_WEBHOOK }}
+          title: "🚨 Nightly Tag Push Failed"
+          ref-name: ${{ github.ref_name }}
+          failed-jobs: "create-and-push-tag"

.github/workflows/zizmor.yml

@@ -0,0 +1,50 @@
+name: Run Zizmor
+
+on:
+  push:
+    branches: ["main"]
+  pull_request:
+    branches: ["**"]
+
+permissions: {}
+
+jobs:
+  zizmor:
+    name: zizmor
+    runs-on: ubuntu-slim
+    timeout-minutes: 45
+    permissions:
+      security-events: write # needed for SARIF uploads
+    steps:
+      - name: Checkout repository
+        uses: actions/checkout@8e8c483db84b4bee98b60c0593521ed34d9990e8 # ratchet:actions/checkout@v6.0.1
+        with:
+          persist-credentials: false
+
+      - name: Detect changes
+        id: filter
+        uses: dorny/paths-filter@de90cc6fb38fc0963ad72b210f1f284cd68cea36 # ratchet:dorny/paths-filter@v3
+        with:
+          filters: |
+            zizmor:
+              - '.github/**'
+
+      - name: Install the latest version of uv
+        if: steps.filter.outputs.zizmor == 'true' || github.ref_name == 'main'
+        uses: astral-sh/setup-uv@ed21f2f24f8dd64503750218de024bcf64c7250a # ratchet:astral-sh/setup-uv@v7
+        with:
+          enable-cache: false
+          version: "0.9.9"
+
+      - name: Run zizmor
+        if: steps.filter.outputs.zizmor == 'true' || github.ref_name == 'main'
+        run: uv run --no-sync --with zizmor zizmor --format=sarif . > results.sarif
+        env:
+          GH_TOKEN: ${{ secrets.GITHUB_TOKEN }}
+
+      - name: Upload SARIF file
+        if: steps.filter.outputs.zizmor == 'true' || github.ref_name == 'main'
+        uses: github/codeql-action/upload-sarif@ba454b8ab46733eb6145342877cd148270bb77ab # ratchet:github/codeql-action/upload-sarif@codeql-bundle-v2.23.5
+        with:
+          sarif_file: results.sarif
+          category: zizmor

.gitignore

@@ -1,6 +1,7 @@
 # editors
 .vscode
 .zed
+.cursor
 
 # macos
 .DS_store
@@ -18,6 +19,7 @@ backend/tests/regression/search_quality/eval-*
 backend/tests/regression/search_quality/search_eval_config.yaml
 backend/tests/regression/search_quality/*.json
 backend/onyx/evals/data/
+backend/onyx/evals/one_off/*.json
 *.log
 
 # secret files
@@ -27,10 +29,16 @@ settings.json
 
 # others
 /deployment/data/nginx/app.conf
+/deployment/data/nginx/mcp.conf.inc
+/deployment/data/nginx/mcp_upstream.conf.inc
 *.sw?
 /backend/tests/regression/answer_quality/search_test_config.yaml
 *.egg-info
 
+# Claude
+AGENTS.md
+CLAUDE.md
+
 # Local .terraform directories
 **/.terraform/*
 
@@ -40,3 +48,11 @@ settings.json
 
 # Local .terraform.lock.hcl file
 .terraform.lock.hcl
+
+node_modules
+
+# MCP configs
+.playwright-mcp
+
+# plans
+plans/

.mcp.json.template

@@ -1,8 +0,0 @@
-{
-  "mcpServers": {
-    "onyx-mcp": {
-      "type": "http",
-      "url": "http://localhost:8000/mcp"
-    }
-  }
-}

.pre-commit-config.yaml

@@ -1,75 +1,159 @@
+default_install_hook_types:
+  - pre-commit
+  - post-checkout
+  - post-merge
+  - post-rewrite
 repos:
-  - repo: https://github.com/psf/black
-    rev: 25.1.0
+  - repo: https://github.com/astral-sh/uv-pre-commit
+    # From: https://github.com/astral-sh/uv-pre-commit/pull/53/commits/d30b4298e4fb63ce8609e29acdbcf4c9018a483c
+    rev: d30b4298e4fb63ce8609e29acdbcf4c9018a483c
     hooks:
-    - id: black
-      language_version: python3.11
+      - id: uv-sync
+        args: ["--locked", "--all-extras"]
+      - id: uv-lock
+        files: ^pyproject\.toml$
+      - id: uv-export
+        name: uv-export default.txt
+        args:
+          [
+            "--no-emit-project",
+            "--no-default-groups",
+            "--no-hashes",
+            "--extra",
+            "backend",
+            "-o",
+            "backend/requirements/default.txt",
+          ]
+        files: ^(pyproject\.toml|uv\.lock|backend/requirements/.*\.txt)$
+      - id: uv-export
+        name: uv-export dev.txt
+        args:
+          [
+            "--no-emit-project",
+            "--no-default-groups",
+            "--no-hashes",
+            "--extra",
+            "dev",
+            "-o",
+            "backend/requirements/dev.txt",
+          ]
+        files: ^(pyproject\.toml|uv\.lock|backend/requirements/.*\.txt)$
+      - id: uv-export
+        name: uv-export ee.txt
+        args:
+          [
+            "--no-emit-project",
+            "--no-default-groups",
+            "--no-hashes",
+            "--extra",
+            "ee",
+            "-o",
+            "backend/requirements/ee.txt",
+          ]
+        files: ^(pyproject\.toml|uv\.lock|backend/requirements/.*\.txt)$
+      - id: uv-export
+        name: uv-export model_server.txt
+        args:
+          [
+            "--no-emit-project",
+            "--no-default-groups",
+            "--no-hashes",
+            "--extra",
+            "model_server",
+            "-o",
+            "backend/requirements/model_server.txt",
+          ]
+        files: ^(pyproject\.toml|uv\.lock|backend/requirements/.*\.txt)$
+      - id: uv-run
+        name: Check lazy imports
+        args: ["--active", "--with=onyx-devtools", "ods", "check-lazy-imports"]
+        files: ^backend/(?!\.venv/).*\.py$
+      # NOTE: This takes ~6s on a single, large module which is prohibitively slow.
+      # - id: uv-run
+      #   name: mypy
+      #   args: ["--all-extras", "mypy"]
+      #   pass_filenames: true
+      #   files: ^backend/.*\.py$
+
+  - repo: https://github.com/rhysd/actionlint
+    rev: a443f344ff32813837fa49f7aa6cbc478d770e62 # frozen: v1.7.9
+    hooks:
+      - id: actionlint
+
+  - repo: https://github.com/psf/black
+    rev: 8a737e727ac5ab2f1d4cf5876720ed276dc8dc4b # frozen: 25.1.0
+    hooks:
+      - id: black
+        language_version: python3.11
 
   # this is a fork which keeps compatibility with black
   - repo: https://github.com/wimglenn/reorder-python-imports-black
-    rev: v3.14.0
+    rev: f55cd27f90f0cf0ee775002c2383ce1c7820013d # frozen: v3.14.0
     hooks:
-    - id: reorder-python-imports
-      args: ['--py311-plus', '--application-directories=backend/']
-      # need to ignore alembic files, since reorder-python-imports gets confused
-      # and thinks that alembic is a local package since there is a folder
-      # in the backend directory called `alembic`
-      exclude: ^backend/alembic/
+      - id: reorder-python-imports
+        args: ["--py311-plus", "--application-directories=backend/"]
+        # need to ignore alembic files, since reorder-python-imports gets confused
+        # and thinks that alembic is a local package since there is a folder
+        # in the backend directory called `alembic`
+        exclude: ^backend/alembic/
 
   # These settings will remove unused imports with side effects
   # Note: The repo currently does not and should not have imports with side effects
   - repo: https://github.com/PyCQA/autoflake
-    rev: v2.3.1
+    rev: 0544741e2b4a22b472d9d93e37d4ea9153820bb1 # frozen: v2.3.1
     hooks:
       - id: autoflake
-        args: [ '--remove-all-unused-imports', '--remove-unused-variables', '--in-place' , '--recursive']
+        args:
+          [
+            "--remove-all-unused-imports",
+            "--remove-unused-variables",
+            "--in-place",
+            "--recursive",
+          ]
+
+  - repo: https://github.com/golangci/golangci-lint
+    rev: 9f61b0f53f80672872fced07b6874397c3ed197b # frozen: v2.7.2
+    hooks:
+      - id: golangci-lint
+        entry: bash -c "find tools/ -name go.mod -print0 | xargs -0 -I{} bash -c 'cd \"$(dirname {})\" && golangci-lint run ./...'"
 
   - repo: https://github.com/astral-sh/ruff-pre-commit
     # Ruff version.
-    rev: v0.11.4
+    rev: 971923581912ef60a6b70dbf0c3e9a39563c9d47 # frozen: v0.11.4
     hooks:
       - id: ruff
+
   - repo: https://github.com/pre-commit/mirrors-prettier
-    rev: v3.1.0
+    rev: ffb6a759a979008c0e6dff86e39f4745a2d9eac4 # frozen: v3.1.0
     hooks:
-    - id: prettier
-      types_or: [html, css, javascript, ts, tsx]
-      additional_dependencies:
-      - prettier
+      - id: prettier
+        types_or: [html, css, javascript, ts, tsx]
+        language_version: system
+
+  - repo: https://github.com/sirwart/ripsecrets
+    rev: 7d94620933e79b8acaa0cd9e60e9864b07673d86 # frozen: v0.1.11
+    hooks:
+      - id: ripsecrets
+        args:
+          - --additional-pattern
+          - ^sk-[A-Za-z0-9_\-]{20,}$
 
   - repo: local
     hooks:
-      - id: check-lazy-imports
-        name: Check lazy imports are not directly imported
-        entry: python3 backend/scripts/check_lazy_imports.py
+      - id: terraform-fmt
+        name: terraform fmt
+        entry: terraform fmt -recursive
         language: system
-        files: ^backend/(?!\.venv/).*\.py$
         pass_filenames: false
+        files: \.tf$
 
-  # We would like to have a mypy pre-commit hook, but due to the fact that
-  # pre-commit runs in it's own isolated environment, we would need to install
-  # and keep in sync all dependencies so mypy has access to the appropriate type
-  # stubs. This does not seem worth it at the moment, so for now we will stick to
-  # having mypy run via Github Actions / manually by contributors
-  # - repo: https://github.com/pre-commit/mirrors-mypy
-  #   rev: v1.1.1
-  #   hooks:
-  #     - id: mypy
-  #       exclude: ^tests/
-  #       # below are needed for type stubs since pre-commit runs in it's own
-  #       # isolated environment. Unfortunately, this needs to be kept in sync
-  #       # with requirements/dev.txt + requirements/default.txt
-  #       additional_dependencies: [
-  #         alembic==1.10.4,
-  #         types-beautifulsoup4==4.12.0.3,
-  #         types-html5lib==1.1.11.13,
-  #         types-oauthlib==3.2.0.9,
-  #         types-psycopg2==2.9.21.10,
-  #         types-python-dateutil==2.8.19.13,
-  #         types-regex==2023.3.23.1,
-  #         types-requests==2.28.11.17,
-  #         types-retry==0.9.9.3,
-  #         types-urllib3==1.26.25.11
-  #       ]
-  #       # TODO: add back once errors are addressed
-  #       # args: [--strict]
+      # Uses tsgo (TypeScript's native Go compiler) for ~10x faster type checking.
+      # This is a preview package - if it breaks:
+      #   1. Try updating: cd web && npm update @typescript/native-preview
+      #   2. Or fallback to tsc: replace 'tsgo' with 'tsc' below
+      - id: typescript-check
+        name: TypeScript type check
+        entry: bash -c 'cd web && npx tsgo --noEmit --project tsconfig.types.json'
+        language: system
+        pass_filenames: false
+        files: ^web/.*\.(ts|tsx)$

.vscode/env_template.txt

@@ -1,66 +1,59 @@
-# Copy this file to .env in the .vscode folder
-# Fill in the <REPLACE THIS> values as needed, it is recommended to set the GEN_AI_API_KEY value to avoid having to set up an LLM in the UI
-# Also check out danswer/backend/scripts/restart_containers.sh for a script to restart the containers which Danswer relies on outside of VSCode/Cursor processes
+# Copy this file to .env in the .vscode folder.
+# Fill in the <REPLACE THIS> values as needed; it is recommended to set the
+# GEN_AI_API_KEY value to avoid having to set up an LLM in the UI.
+# Also check out onyx/backend/scripts/restart_containers.sh for a script to
+# restart the containers which Onyx relies on outside of VSCode/Cursor
+# processes.
 
-# For local dev, often user Authentication is not needed
+
+# For local dev, often user Authentication is not needed.
 AUTH_TYPE=disabled
 
-# Skip warm up for dev
-SKIP_WARM_UP=True
 
-# Always keep these on for Dev
-# Logs all model prompts to stdout
+# Always keep these on for Dev.
+# Logs model prompts, reasoning, and answer to stdout.
 LOG_ONYX_MODEL_INTERACTIONS=True
 # More verbose logging
 LOG_LEVEL=debug
 
 
-# This passes top N results to LLM an additional time for reranking prior to answer generation
-# This step is quite heavy on token usage so we disable it for dev generally
+# This passes top N results to LLM an additional time for reranking prior to
+# answer generation.
+# This step is quite heavy on token usage so we disable it for dev generally.
 DISABLE_LLM_DOC_RELEVANCE=False
 
 
-# Useful if you want to toggle auth on/off (google_oauth/OIDC specifically)
+# Useful if you want to toggle auth on/off (google_oauth/OIDC specifically).
 OAUTH_CLIENT_ID=<REPLACE THIS>
 OAUTH_CLIENT_SECRET=<REPLACE THIS>
 OPENID_CONFIG_URL=<REPLACE THIS>
 SAML_CONF_DIR=/<ABSOLUTE PATH TO ONYX>/onyx/backend/ee/onyx/configs/saml_config
 
-# Generally not useful for dev, we don't generally want to set up an SMTP server for dev
+
+# Generally not useful for dev, we don't generally want to set up an SMTP server
+# for dev.
 REQUIRE_EMAIL_VERIFICATION=False
 
 
-# Set these so if you wipe the DB, you don't end up having to go through the UI every time
+# Set these so if you wipe the DB, you don't end up having to go through the UI
+# every time.
 GEN_AI_API_KEY=<REPLACE THIS>
 OPENAI_API_KEY=<REPLACE THIS>
-# If answer quality isn't important for dev, use gpt-4o-mini since it's cheaper
+# If answer quality isn't important for dev, use gpt-4o-mini since it's cheaper.
 GEN_AI_MODEL_VERSION=gpt-4o
 FAST_GEN_AI_MODEL_VERSION=gpt-4o
 
-# For Danswer Slack Bot, overrides the UI values so no need to set this up via UI every time
-# Only needed if using DanswerBot
-#ONYX_BOT_SLACK_APP_TOKEN=<REPLACE THIS>
-#ONYX_BOT_SLACK_BOT_TOKEN=<REPLACE THIS>
-
 
 # Python stuff
 PYTHONPATH=../backend
 PYTHONUNBUFFERED=1
 
 
-# Internet Search
-EXA_API_KEY=<REPLACE THIS>
-
-
-# Enable the full set of Danswer Enterprise Edition features
-# NOTE: DO NOT ENABLE THIS UNLESS YOU HAVE A PAID ENTERPRISE LICENSE (or if you are using this for local testing/development)
+# Enable the full set of Danswer Enterprise Edition features.
+# NOTE: DO NOT ENABLE THIS UNLESS YOU HAVE A PAID ENTERPRISE LICENSE (or if you
+# are using this for local testing/development).
 ENABLE_PAID_ENTERPRISE_EDITION_FEATURES=False
 
-# Agent Search configs  # TODO: Remove give proper namings
-AGENT_RETRIEVAL_STATS=False   # Note: This setting will incur substantial re-ranking effort
-AGENT_RERANKING_STATS=True
-AGENT_MAX_QUERY_RETRIEVAL_RESULTS=20
-AGENT_RERANKING_MAX_QUERY_RETRIEVAL_RESULTS=20
 
 # S3 File Store Configuration (MinIO for local development)
 S3_ENDPOINT_URL=http://localhost:9004
@@ -68,11 +61,24 @@ S3_FILE_STORE_BUCKET_NAME=onyx-file-store-bucket
 S3_AWS_ACCESS_KEY_ID=minioadmin
 S3_AWS_SECRET_ACCESS_KEY=minioadmin
 
-# Show extra/uncommon connectors
+
+# Show extra/uncommon connectors.
 SHOW_EXTRA_CONNECTORS=True
 
+
 # Local langsmith tracing
 LANGSMITH_TRACING="true"
 LANGSMITH_ENDPOINT="https://api.smith.langchain.com"
 LANGSMITH_API_KEY=<REPLACE_THIS>
-LANGSMITH_PROJECT=<REPLACE_THIS>
+LANGSMITH_PROJECT=<REPLACE_THIS>
+
+
+# Local Confluence OAuth testing
+# OAUTH_CONFLUENCE_CLOUD_CLIENT_ID=<REPLACE_THIS>
+# OAUTH_CONFLUENCE_CLOUD_CLIENT_SECRET=<REPLACE_THIS>
+# NEXT_PUBLIC_TEST_ENV=True
+
+
+# OpenSearch
+# Arbitrary password is fine for local development.
+OPENSEARCH_INITIAL_ADMIN_PASSWORD=<REPLACE THIS>

AGENTS.md.template

@@ -1,13 +1,13 @@
 # AGENTS.md
 
-This file provides guidance to Codex when working with code in this repository.
+This file provides guidance to AI agents when working with code in this repository.
 
 ## KEY NOTES
 
-- If you run into any missing python dependency errors, try running your command with `source backend/.venv/bin/activate` \
+- If you run into any missing python dependency errors, try running your command with `source .venv/bin/activate` \
 to assume the python venv.
 - To make tests work, check the `.env` file at the root of the project to find an OpenAI key.
-- If using `playwright` to explore the frontend, you can usually log in with username `a@test.com` and password
+- If using `playwright` to explore the frontend, you can usually log in with username `a@example.com` and password
 `a`. The app can be accessed at `http://localhost:3000`.
 - You should assume that all Onyx services are running. To verify, you can check the `backend/log` directory to
 make sure we see logs coming out from the relevant service.
@@ -70,7 +70,12 @@ Onyx uses Celery for asynchronous task processing with multiple specialized work
    - Single thread (monitoring doesn't need parallelism)
    - Cloud-specific monitoring tasks
 
-8. **Beat Worker** (`beat`)
+8. **User File Processing Worker** (`user_file_processing`)
+   - Processes user-uploaded files
+   - Handles user file indexing and project synchronization
+   - Configurable concurrency
+
+9. **Beat Worker** (`beat`)
    - Celery's scheduler for periodic tasks
    - Uses DynamicTenantScheduler for multi-tenant support
    - Schedules tasks like:
@@ -82,6 +87,31 @@ Onyx uses Celery for asynchronous task processing with multiple specialized work
      - Monitoring tasks (every 5 minutes)
      - Cleanup tasks (hourly)
 
+#### Worker Deployment Modes
+
+Onyx supports two deployment modes for background workers, controlled by the `USE_LIGHTWEIGHT_BACKGROUND_WORKER` environment variable:
+
+**Lightweight Mode** (default, `USE_LIGHTWEIGHT_BACKGROUND_WORKER=true`):
+- Runs a single consolidated `background` worker that handles all background tasks:
+  - Pruning operations (from `heavy` worker)
+  - Knowledge graph processing (from `kg_processing` worker)
+  - Monitoring tasks (from `monitoring` worker)
+  - User file processing (from `user_file_processing` worker)
+- Lower resource footprint (single worker process)
+- Suitable for smaller deployments or development environments
+- Default concurrency: 6 threads
+
+**Standard Mode** (`USE_LIGHTWEIGHT_BACKGROUND_WORKER=false`):
+- Runs separate specialized workers as documented above (heavy, kg_processing, monitoring, user_file_processing)
+- Better isolation and scalability
+- Can scale individual workers independently based on workload
+- Suitable for production deployments with higher load
+
+The deployment mode affects:
+- **Backend**: Worker processes spawned by supervisord or dev scripts
+- **Helm**: Which Kubernetes deployments are created
+- **Dev Environment**: Which workers `dev_run_background_jobs.py` spawns
+
 #### Key Features
 
 - **Thread-based Workers**: All workers use thread pools (not processes) for stability
@@ -151,6 +181,286 @@ web/
 └── src/lib/                     # Utilities & business logic
 ```
 
+## Frontend Standards
+
+### 1. Import Standards
+
+**Always use absolute imports with the `@` prefix.**
+
+**Reason:** Moving files around becomes easier since you don't also have to update those import statements. This makes modifications to the codebase much nicer.
+
+```typescript
+// ✅ Good
+import { Button } from "@/components/ui/button";
+import { useAuth } from "@/hooks/useAuth";
+import { Text } from "@/refresh-components/texts/Text";
+
+// ❌ Bad
+import { Button } from "../../../components/ui/button";
+import { useAuth } from "./hooks/useAuth";
+```
+
+### 2. React Component Functions
+
+**Prefer regular functions over arrow functions for React components.**
+
+**Reason:** Functions just become easier to read.
+
+```typescript
+// ✅ Good
+function UserProfile({ userId }: UserProfileProps) {
+  return <div>User Profile</div>
+}
+
+// ❌ Bad
+const UserProfile = ({ userId }: UserProfileProps) => {
+  return <div>User Profile</div>
+}
+```
+
+### 3. Props Interface Extraction
+
+**Extract prop types into their own interface definitions.**
+
+**Reason:** Functions just become easier to read.
+
+```typescript
+// ✅ Good
+interface UserCardProps {
+  user: User
+  showActions?: boolean
+  onEdit?: (userId: string) => void
+}
+
+function UserCard({ user, showActions = false, onEdit }: UserCardProps) {
+  return <div>User Card</div>
+}
+
+// ❌ Bad
+function UserCard({
+  user,
+  showActions = false,
+  onEdit
+}: {
+  user: User
+  showActions?: boolean
+  onEdit?: (userId: string) => void
+}) {
+  return <div>User Card</div>
+}
+```
+
+### 4. Spacing Guidelines
+
+**Prefer padding over margins for spacing.**
+
+**Reason:** We want to consolidate usage to paddings instead of margins.
+
+```typescript
+// ✅ Good
+<div className="p-4 space-y-2">
+  <div className="p-2">Content</div>
+</div>
+
+// ❌ Bad
+<div className="m-4 space-y-2">
+  <div className="m-2">Content</div>
+</div>
+```
+
+### 5. Tailwind Dark Mode
+
+**Strictly forbid using the `dark:` modifier in Tailwind classes, except for logo icon handling.**
+
+**Reason:** The `colors.css` file already, VERY CAREFULLY, defines what the exact opposite colour of each light-mode colour is. Overriding this behaviour is VERY bad and will lead to horrible UI breakages.
+
+**Exception:** The `createLogoIcon` helper in `web/src/components/icons/icons.tsx` uses `dark:` modifiers (`dark:invert`, `dark:hidden`, `dark:block`) to handle third-party logo icons that cannot automatically adapt through `colors.css`. This is the ONLY acceptable use of dark mode modifiers.
+
+```typescript
+// ✅ Good - Standard components use `web/tailwind-themes/tailwind.config.js` / `web/src/app/css/colors.css`
+<div className="bg-background-neutral-03 text-text-02">
+  Content
+</div>
+
+// ✅ Good - Logo icons with dark mode handling via createLogoIcon
+export const GithubIcon = createLogoIcon(githubLightIcon, {
+  monochromatic: true,  // Will apply dark:invert internally
+});
+
+export const GitbookIcon = createLogoIcon(gitbookLightIcon, {
+  darkSrc: gitbookDarkIcon,  // Will use dark:hidden/dark:block internally
+});
+
+// ❌ Bad - Manual dark mode overrides
+<div className="bg-white dark:bg-black text-black dark:text-white">
+  Content
+</div>
+```
+
+### 6. Class Name Utilities
+
+**Use the `cn` utility instead of raw string formatting for classNames.**
+
+**Reason:** `cn`s are easier to read. They also allow for more complex types (i.e., string-arrays) to get formatted properly (it flattens each element in that string array down). As a result, it can allow things such as conditionals (i.e., `myCondition && "some-tailwind-class"`, which evaluates to `false` when `myCondition` is `false`) to get filtered out.
+
+```typescript
+import { cn } from '@/lib/utils'
+
+// ✅ Good
+<div className={cn(
+  'base-class',
+  isActive && 'active-class',
+  className
+)}>
+  Content
+</div>
+
+// ❌ Bad
+<div className={`base-class ${isActive ? 'active-class' : ''} ${className}`}>
+  Content
+</div>
+```
+
+### 7. Custom Hooks Organization
+
+**Follow a "hook-per-file" layout. Each hook should live in its own file within `web/src/hooks`.**
+
+**Reason:** This is just a layout preference. Keeps code clean.
+
+```typescript
+// web/src/hooks/useUserData.ts
+export function useUserData(userId: string) {
+  // hook implementation
+}
+
+// web/src/hooks/useLocalStorage.ts
+export function useLocalStorage<T>(key: string, initialValue: T) {
+  // hook implementation
+}
+```
+
+### 8. Icon Usage
+
+**ONLY use icons from the `web/src/icons` directory. Do NOT use icons from `react-icons`, `lucide`, or other external libraries.**
+
+**Reason:** We have a very carefully curated selection of icons that match our Onyx guidelines. We do NOT want to muddy those up with different aesthetic stylings.
+
+```typescript
+// ✅ Good
+import SvgX from "@/icons/x";
+import SvgMoreHorizontal from "@/icons/more-horizontal";
+
+// ❌ Bad
+import { User } from "lucide-react";
+import { FiSearch } from "react-icons/fi";
+```
+
+**Missing Icons**: If an icon is needed but doesn't exist in the `web/src/icons` directory, import it from Figma using the Figma MCP tool and add it to the icons directory.
+If you need help with this step, reach out to `raunak@onyx.app`.
+
+### 9. Text Rendering
+
+**Prefer using the `refresh-components/texts/Text` component for all text rendering. Avoid "naked" text nodes.**
+
+**Reason:** The `Text` component is fully compliant with the stylings provided in Figma. It provides easy utilities to specify the text-colour and font-size in the form of flags. Super duper easy.
+
+```typescript
+// ✅ Good
+import { Text } from '@/refresh-components/texts/Text'
+
+function UserCard({ name }: { name: string }) {
+  return (
+    <Text
+      {/* The `text03` flag makes the text it renders to be coloured the 3rd-scale grey */}
+      text03
+      {/* The `mainAction` flag makes the text it renders to be "main-action" font + line-height + weightage, as described in the Figma */}
+      mainAction
+    >
+      {name}
+    </Text>
+  )
+}
+
+// ❌ Bad
+function UserCard({ name }: { name: string }) {
+  return (
+    <div>
+      <h2>{name}</h2>
+      <p>User details</p>
+    </div>
+  )
+}
+```
+
+### 10. Component Usage
+
+**Heavily avoid raw HTML input components. Always use components from the `web/src/refresh-components` or `web/lib/opal/src` directory.**
+
+**Reason:** We've put in a lot of effort to unify the components that are rendered in the Onyx app. Using raw components breaks the entire UI of the application, and leaves it in a muddier state than before.
+
+```typescript
+// ✅ Good
+import Button from '@/refresh-components/buttons/Button'
+import InputTypeIn from '@/refresh-components/inputs/InputTypeIn'
+import SvgPlusCircle from '@/icons/plus-circle'
+
+function ContactForm() {
+  return (
+    <form>
+      <InputTypeIn placeholder="Search..." />
+      <Button type="submit" leftIcon={SvgPlusCircle}>Submit</Button>
+    </form>
+  )
+}
+
+// ❌ Bad
+function ContactForm() {
+  return (
+    <form>
+      <input placeholder="Name" />
+      <textarea placeholder="Message" />
+      <button type="submit">Submit</button>
+    </form>
+  )
+}
+```
+
+### 11. Colors
+
+**Always use custom overrides for colors and borders rather than built in Tailwind CSS colors. These overrides live in `web/tailwind-themes/tailwind.config.js`.**
+
+**Reason:** Our custom color system uses CSS variables that automatically handle dark mode and maintain design consistency across the app. Standard Tailwind colors bypass this system.
+
+**Available color categories:**
+- **Text:** `text-01` through `text-05`, `text-inverted-XX`
+- **Backgrounds:** `background-neutral-XX`, `background-tint-XX` (and inverted variants)
+- **Borders:** `border-01` through `border-05`, `border-inverted-XX`
+- **Actions:** `action-link-XX`, `action-danger-XX`
+- **Status:** `status-info-XX`, `status-success-XX`, `status-warning-XX`, `status-error-XX`
+- **Theme:** `theme-primary-XX`, `theme-red-XX`, `theme-blue-XX`, etc.
+
+```typescript
+// ✅ Good - Use custom Onyx color classes
+<div className="bg-background-neutral-01 border border-border-02" />
+<div className="bg-background-tint-02 border border-border-01" />
+<div className="bg-status-success-01" />
+<div className="bg-action-link-01" />
+<div className="bg-theme-primary-05" />
+
+// ❌ Bad - Do NOT use standard Tailwind colors
+<div className="bg-gray-100 border border-gray-300 text-gray-600" />
+<div className="bg-white border border-slate-200" />
+<div className="bg-green-100 text-green-700" />
+<div className="bg-blue-100 text-blue-600" />
+<div className="bg-indigo-500" />
+```
+
+### 12. Data Fetching
+
+**Prefer using `useSWR` for data fetching. Data should generally be fetched on the client side. Components that need data should display a loader / placeholder while waiting for that data. Prefer loading data within the component that needs it rather than at the top level and passing it down.**
+
+**Reason:** Client side fetching allows us to load the skeleton of the page without waiting for data to load, leading to a snappier UX. Loading data where needed reduces dependencies between a component and its parent component(s).
+
 ## Database & Migrations
 
 ### Running Migrations
@@ -164,13 +474,15 @@ alembic -n schema_private upgrade head
 
 ### Creating Migrations
 ```bash
-# Auto-generate migration
-alembic revision --autogenerate -m "description"
+# Create migration
+alembic revision -m "description"
 
 # Multi-tenant migration
-alembic -n schema_private revision --autogenerate -m "description"
+alembic -n schema_private revision -m "description"
 ```
 
+Write the migration manually and place it in the file that alembic creates when running the above command.
+
 ## Testing Strategy
 
 There are 4 main types of tests within Onyx:
@@ -263,14 +575,6 @@ will be tailing their logs to this file.
 - Token management and rate limiting
 - Custom prompts and agent actions
 
-## UI/UX Patterns
-
-- Tailwind CSS with design system in `web/src/components/ui/`
-- Radix UI and Headless UI for accessible components
-- SWR for data fetching and caching
-- Form validation with react-hook-form
-- Error handling with popup notifications
-
 ## Creating a Plan
 When creating a plan in the `plans` directory, make sure to include at least these elements:
 

CLAUDE.md

@@ -1,295 +0,0 @@
-# CLAUDE.md
-
-This file provides guidance to Claude Code (claude.ai/code) when working with code in this repository.
-
-## KEY NOTES
-
-- If you run into any missing python dependency errors, try running your command with `source backend/.venv/bin/activate` \
-to assume the python venv.
-- To make tests work, check the `.env` file at the root of the project to find an OpenAI key.
-- If using `playwright` to explore the frontend, you can usually log in with username `a@test.com` and password
-`a`. The app can be accessed at `http://localhost:3000`.
-- You should assume that all Onyx services are running. To verify, you can check the `backend/log` directory to
-make sure we see logs coming out from the relevant service.
-- To connect to the Postgres database, use: `docker exec -it onyx-relational_db-1 psql -U postgres -c "<SQL>"`
-- When making calls to the backend, always go through the frontend. E.g. make a call to `http://localhost:3000/api/persona` not `http://localhost:8080/api/persona`
-- Put ALL db operations under the `backend/onyx/db` / `backend/ee/onyx/db` directories. Don't run queries
-outside of those directories.
-
-## Project Overview
-
-**Onyx** (formerly Danswer) is an open-source Gen-AI and Enterprise Search platform that connects to company documents, apps, and people. It features a modular architecture with both Community Edition (MIT licensed) and Enterprise Edition offerings.
-
-
-### Background Workers (Celery)
-
-Onyx uses Celery for asynchronous task processing with multiple specialized workers:
-
-#### Worker Types
-
-1. **Primary Worker** (`celery_app.py`)
-   - Coordinates core background tasks and system-wide operations
-   - Handles connector management, document sync, pruning, and periodic checks
-   - Runs with 4 threads concurrency
-   - Tasks: connector deletion, vespa sync, pruning, LLM model updates, user file sync
-
-2. **Docfetching Worker** (`docfetching`)
-   - Fetches documents from external data sources (connectors)
-   - Spawns docprocessing tasks for each document batch
-   - Implements watchdog monitoring for stuck connectors
-   - Configurable concurrency (default from env)
-
-3. **Docprocessing Worker** (`docprocessing`)
-   - Processes fetched documents through the indexing pipeline:
-     - Upserts documents to PostgreSQL
-     - Chunks documents and adds contextual information
-     - Embeds chunks via model server
-     - Writes chunks to Vespa vector database
-     - Updates document metadata
-   - Configurable concurrency (default from env)
-
-4. **Light Worker** (`light`)
-   - Handles lightweight, fast operations
-   - Tasks: vespa operations, document permissions sync, external group sync
-   - Higher concurrency for quick tasks
-
-5. **Heavy Worker** (`heavy`)
-   - Handles resource-intensive operations
-   - Primary task: document pruning operations
-   - Runs with 4 threads concurrency
-
-6. **KG Processing Worker** (`kg_processing`)
-   - Handles Knowledge Graph processing and clustering
-   - Builds relationships between documents
-   - Runs clustering algorithms
-   - Configurable concurrency
-
-7. **Monitoring Worker** (`monitoring`)
-   - System health monitoring and metrics collection
-   - Monitors Celery queues, process memory, and system status
-   - Single thread (monitoring doesn't need parallelism)
-   - Cloud-specific monitoring tasks
-
-8. **Beat Worker** (`beat`)
-   - Celery's scheduler for periodic tasks
-   - Uses DynamicTenantScheduler for multi-tenant support
-   - Schedules tasks like:
-     - Indexing checks (every 15 seconds)
-     - Connector deletion checks (every 20 seconds)
-     - Vespa sync checks (every 20 seconds)
-     - Pruning checks (every 20 seconds)
-     - KG processing (every 60 seconds)
-     - Monitoring tasks (every 5 minutes)
-     - Cleanup tasks (hourly)
-
-#### Key Features
-
-- **Thread-based Workers**: All workers use thread pools (not processes) for stability
-- **Tenant Awareness**: Multi-tenant support with per-tenant task isolation. There is a 
-middleware layer that automatically finds the appropriate tenant ID when sending tasks 
-via Celery Beat.
-- **Task Prioritization**: High, Medium, Low priority queues
-- **Monitoring**: Built-in heartbeat and liveness checking
-- **Failure Handling**: Automatic retry and failure recovery mechanisms
-- **Redis Coordination**: Inter-process communication via Redis
-- **PostgreSQL State**: Task state and metadata stored in PostgreSQL
-
-
-#### Important Notes
-
-**Defining Tasks**: 
-- Always use `@shared_task` rather than `@celery_app`
-- Put tasks under `background/celery/tasks/` or `ee/background/celery/tasks`
-
-**Defining APIs**:
-When creating new FastAPI APIs, do NOT use the `response_model` field. Instead, just type the
-function.
-
-**Testing Updates**:
-If you make any updates to a celery worker and you want to test these changes, you will need
-to ask me to restart the celery worker. There is no auto-restart on code-change mechanism.
-
-### Code Quality
-```bash
-# Install and run pre-commit hooks
-pre-commit install
-pre-commit run --all-files
-```
-
-NOTE: Always make sure everything is strictly typed (both in Python and Typescript).
-
-## Architecture Overview
-
-### Technology Stack
-- **Backend**: Python 3.11, FastAPI, SQLAlchemy, Alembic, Celery
-- **Frontend**: Next.js 15+, React 18, TypeScript, Tailwind CSS
-- **Database**: PostgreSQL with Redis caching
-- **Search**: Vespa vector database
-- **Auth**: OAuth2, SAML, multi-provider support
-- **AI/ML**: LangChain, LiteLLM, multiple embedding models
-
-### Directory Structure
-
-```
-backend/
-├── onyx/
-│   ├── auth/                    # Authentication & authorization
-│   ├── chat/                    # Chat functionality & LLM interactions
-│   ├── connectors/              # Data source connectors
-│   ├── db/                      # Database models & operations
-│   ├── document_index/          # Vespa integration
-│   ├── federated_connectors/    # External search connectors
-│   ├── llm/                     # LLM provider integrations
-│   └── server/                  # API endpoints & routers
-├── ee/                          # Enterprise Edition features
-├── alembic/                     # Database migrations
-└── tests/                       # Test suites
-
-web/
-├── src/app/                     # Next.js app router pages
-├── src/components/              # Reusable React components
-└── src/lib/                     # Utilities & business logic
-```
-
-## Database & Migrations
-
-### Running Migrations
-```bash
-# Standard migrations
-alembic upgrade head
-
-# Multi-tenant (Enterprise)
-alembic -n schema_private upgrade head
-```
-
-### Creating Migrations
-```bash
-# Auto-generate migration
-alembic revision --autogenerate -m "description"
-
-# Multi-tenant migration
-alembic -n schema_private revision --autogenerate -m "description"
-```
-
-## Testing Strategy
-
-There are 4 main types of tests within Onyx:
-
-### Unit Tests
-These should not assume any Onyx/external services are available to be called.
-Interactions with the outside world should be mocked using `unittest.mock`. Generally, only 
-write these for complex, isolated modules e.g. `citation_processing.py`.
-
-To run them:
-
-```bash
-python -m dotenv -f .vscode/.env run -- pytest -xv backend/tests/unit
-```
-
-### External Dependency Unit Tests
-These tests assume that all external dependencies of Onyx are available and callable (e.g. Postgres, Redis, 
-MinIO/S3, Vespa are running + OpenAI can be called + any request to the internet is fine + etc.).
-
-However, the actual Onyx containers are not running and with these tests we call the function to test directly.
-We can also mock components/calls at will. 
-
-The goal with these tests are to minimize mocking while giving some flexibility to mock things that are flakey, 
-need strictly controlled behavior, or need to have their internal behavior validated (e.g. verify a function is called
-with certain args, something that would be impossible with proper integration tests).
-
-A great example of this type of test is `backend/tests/external_dependency_unit/connectors/confluence/test_confluence_group_sync.py`.
-
-To run them:
-
-```bash
-python -m dotenv -f .vscode/.env run -- pytest backend/tests/external_dependency_unit
-```
-
-### Integration Tests
-Standard integration tests. Every test in `backend/tests/integration` runs against a real Onyx deployment. We cannot 
-mock anything in these tests. Prefer writing integration tests (or External Dependency Unit Tests if mocking/internal 
-verification is necessary) over any other type of test.
-
-Tests are parallelized at a directory level.
-
-When writing integration tests, make sure to check the root `conftest.py` for useful fixtures + the `backend/tests/integration/common_utils` directory for utilities. Prefer (if one exists), calling the appropriate Manager 
-class in the utils over directly calling the APIs with a library like `requests`. Prefer using fixtures rather than
-calling the utilities directly (e.g. do NOT create admin users with 
-`admin_user = UserManager.create(name="admin_user")`, instead use the `admin_user` fixture).
-
-A great example of this type of test is `backend/tests/integration/dev_apis/test_simple_chat_api.py`.
-
-To run them:
-
-```bash
-python -m dotenv -f .vscode/.env run -- pytest backend/tests/integration
-```
-
-### Playwright (E2E) Tests
-These tests are an even more complete version of the Integration Tests mentioned above. Has all services of Onyx 
-running, *including* the Web Server.
-
-Use these tests for anything that requires significant frontend <-> backend coordination.
-
-Tests are located at `web/tests/e2e`. Tests are written in TypeScript.
-
-To run them:
-
-```bash
-npx playwright test <TEST_NAME>
-```
-
-
-## Logs
-
-When (1) writing integration tests or (2) doing live tests (e.g. curl / playwright) you can get access
-to logs via the `backend/log/<service_name>_debug.log` file. All Onyx services (api_server, web_server, celery_X)
-will be tailing their logs to this file. 
-
-
-## Security Considerations
-
-- Never commit API keys or secrets to repository
-- Use encrypted credential storage for connector credentials
-- Follow RBAC patterns for new features
-- Implement proper input validation with Pydantic models
-- Use parameterized queries to prevent SQL injection
-
-## AI/LLM Integration
-
-- Multiple LLM providers supported via LiteLLM
-- Configurable models per feature (chat, search, embeddings)
-- Streaming support for real-time responses
-- Token management and rate limiting
-- Custom prompts and agent actions
-
-## UI/UX Patterns
-
-- Tailwind CSS with design system in `web/src/components/ui/`
-- Radix UI and Headless UI for accessible components
-- SWR for data fetching and caching
-- Form validation with react-hook-form
-- Error handling with popup notifications
-
-## Creating a Plan
-When creating a plan in the `plans` directory, make sure to include at least these elements:
-
-**Issues to Address**
-What the change is meant to do.
-
-**Important Notes**
-Things you come across in your research that are important to the implementation.
-
-**Implementation strategy**
-How you are going to make the changes happen. High level approach.
-
-**Tests**
-What unit (use rarely), external dependency unit, integration, and playwright tests you plan to write to 
-verify the correct behavior. Don't overtest. Usually, a given change only needs one type of test.
-
-Do NOT include these: *Timeline*, *Rollback plan*
-
-This is a minimal list - feel free to include more. Do NOT write code as part of your plan.
-Keep it high level. You can reference certain files or functions though.
-
-Before writing your plan, make sure to do research. Explore the relevant sections in the codebase.

CLAUDE.md.template

@@ -0,0 +1,604 @@
+# CLAUDE.md
+
+This file provides guidance to Claude Code (claude.ai/code) when working with code in this repository.
+
+## KEY NOTES
+
+- If you run into any missing python dependency errors, try running your command with `source .venv/bin/activate` \
+to assume the python venv.
+- To make tests work, check the `.env` file at the root of the project to find an OpenAI key.
+- If using `playwright` to explore the frontend, you can usually log in with username `a@example.com` and password
+`a`. The app can be accessed at `http://localhost:3000`.
+- You should assume that all Onyx services are running. To verify, you can check the `backend/log` directory to
+make sure we see logs coming out from the relevant service.
+- To connect to the Postgres database, use: `docker exec -it onyx-relational_db-1 psql -U postgres -c "<SQL>"`
+- When making calls to the backend, always go through the frontend. E.g. make a call to `http://localhost:3000/api/persona` not `http://localhost:8080/api/persona`
+- Put ALL db operations under the `backend/onyx/db` / `backend/ee/onyx/db` directories. Don't run queries
+outside of those directories.
+
+## Project Overview
+
+**Onyx** (formerly Danswer) is an open-source Gen-AI and Enterprise Search platform that connects to company documents, apps, and people. It features a modular architecture with both Community Edition (MIT licensed) and Enterprise Edition offerings.
+
+
+### Background Workers (Celery)
+
+Onyx uses Celery for asynchronous task processing with multiple specialized workers:
+
+#### Worker Types
+
+1. **Primary Worker** (`celery_app.py`)
+   - Coordinates core background tasks and system-wide operations
+   - Handles connector management, document sync, pruning, and periodic checks
+   - Runs with 4 threads concurrency
+   - Tasks: connector deletion, vespa sync, pruning, LLM model updates, user file sync
+
+2. **Docfetching Worker** (`docfetching`)
+   - Fetches documents from external data sources (connectors)
+   - Spawns docprocessing tasks for each document batch
+   - Implements watchdog monitoring for stuck connectors
+   - Configurable concurrency (default from env)
+
+3. **Docprocessing Worker** (`docprocessing`)
+   - Processes fetched documents through the indexing pipeline:
+     - Upserts documents to PostgreSQL
+     - Chunks documents and adds contextual information
+     - Embeds chunks via model server
+     - Writes chunks to Vespa vector database
+     - Updates document metadata
+   - Configurable concurrency (default from env)
+
+4. **Light Worker** (`light`)
+   - Handles lightweight, fast operations
+   - Tasks: vespa operations, document permissions sync, external group sync
+   - Higher concurrency for quick tasks
+
+5. **Heavy Worker** (`heavy`)
+   - Handles resource-intensive operations
+   - Primary task: document pruning operations
+   - Runs with 4 threads concurrency
+
+6. **KG Processing Worker** (`kg_processing`)
+   - Handles Knowledge Graph processing and clustering
+   - Builds relationships between documents
+   - Runs clustering algorithms
+   - Configurable concurrency
+
+7. **Monitoring Worker** (`monitoring`)
+   - System health monitoring and metrics collection
+   - Monitors Celery queues, process memory, and system status
+   - Single thread (monitoring doesn't need parallelism)
+   - Cloud-specific monitoring tasks
+
+8. **User File Processing Worker** (`user_file_processing`)
+   - Processes user-uploaded files
+   - Handles user file indexing and project synchronization
+   - Configurable concurrency
+
+9. **Beat Worker** (`beat`)
+   - Celery's scheduler for periodic tasks
+   - Uses DynamicTenantScheduler for multi-tenant support
+   - Schedules tasks like:
+     - Indexing checks (every 15 seconds)
+     - Connector deletion checks (every 20 seconds)
+     - Vespa sync checks (every 20 seconds)
+     - Pruning checks (every 20 seconds)
+     - KG processing (every 60 seconds)
+     - Monitoring tasks (every 5 minutes)
+     - Cleanup tasks (hourly)
+
+#### Worker Deployment Modes
+
+Onyx supports two deployment modes for background workers, controlled by the `USE_LIGHTWEIGHT_BACKGROUND_WORKER` environment variable:
+
+**Lightweight Mode** (default, `USE_LIGHTWEIGHT_BACKGROUND_WORKER=true`):
+- Runs a single consolidated `background` worker that handles all background tasks:
+  - Light worker tasks (Vespa operations, permissions sync, deletion)
+  - Document processing (indexing pipeline)
+  - Document fetching (connector data retrieval)
+  - Pruning operations (from `heavy` worker)
+  - Knowledge graph processing (from `kg_processing` worker)
+  - Monitoring tasks (from `monitoring` worker)
+  - User file processing (from `user_file_processing` worker)
+- Lower resource footprint (fewer worker processes)
+- Suitable for smaller deployments or development environments
+- Default concurrency: 20 threads (increased to handle combined workload)
+
+**Standard Mode** (`USE_LIGHTWEIGHT_BACKGROUND_WORKER=false`):
+- Runs separate specialized workers as documented above (light, docprocessing, docfetching, heavy, kg_processing, monitoring, user_file_processing)
+- Better isolation and scalability
+- Can scale individual workers independently based on workload
+- Suitable for production deployments with higher load
+
+The deployment mode affects:
+- **Backend**: Worker processes spawned by supervisord or dev scripts
+- **Helm**: Which Kubernetes deployments are created
+- **Dev Environment**: Which workers `dev_run_background_jobs.py` spawns
+
+#### Key Features
+
+- **Thread-based Workers**: All workers use thread pools (not processes) for stability
+- **Tenant Awareness**: Multi-tenant support with per-tenant task isolation. There is a
+middleware layer that automatically finds the appropriate tenant ID when sending tasks
+via Celery Beat.
+- **Task Prioritization**: High, Medium, Low priority queues
+- **Monitoring**: Built-in heartbeat and liveness checking
+- **Failure Handling**: Automatic retry and failure recovery mechanisms
+- **Redis Coordination**: Inter-process communication via Redis
+- **PostgreSQL State**: Task state and metadata stored in PostgreSQL
+
+
+#### Important Notes
+
+**Defining Tasks**: 
+- Always use `@shared_task` rather than `@celery_app`
+- Put tasks under `background/celery/tasks/` or `ee/background/celery/tasks`
+
+**Defining APIs**:
+When creating new FastAPI APIs, do NOT use the `response_model` field. Instead, just type the
+function.
+
+**Testing Updates**:
+If you make any updates to a celery worker and you want to test these changes, you will need
+to ask me to restart the celery worker. There is no auto-restart on code-change mechanism.
+
+### Code Quality
+```bash
+# Install and run pre-commit hooks
+pre-commit install
+pre-commit run --all-files
+```
+
+NOTE: Always make sure everything is strictly typed (both in Python and Typescript).
+
+## Architecture Overview
+
+### Technology Stack
+- **Backend**: Python 3.11, FastAPI, SQLAlchemy, Alembic, Celery
+- **Frontend**: Next.js 15+, React 18, TypeScript, Tailwind CSS
+- **Database**: PostgreSQL with Redis caching
+- **Search**: Vespa vector database
+- **Auth**: OAuth2, SAML, multi-provider support
+- **AI/ML**: LangChain, LiteLLM, multiple embedding models
+
+### Directory Structure
+
+```
+backend/
+├── onyx/
+│   ├── auth/                    # Authentication & authorization
+│   ├── chat/                    # Chat functionality & LLM interactions
+│   ├── connectors/              # Data source connectors
+│   ├── db/                      # Database models & operations
+│   ├── document_index/          # Vespa integration
+│   ├── federated_connectors/    # External search connectors
+│   ├── llm/                     # LLM provider integrations
+│   └── server/                  # API endpoints & routers
+├── ee/                          # Enterprise Edition features
+├── alembic/                     # Database migrations
+└── tests/                       # Test suites
+
+web/
+├── src/app/                     # Next.js app router pages
+├── src/components/              # Reusable React components
+└── src/lib/                     # Utilities & business logic
+```
+
+## Frontend Standards
+
+### 1. Import Standards
+
+**Always use absolute imports with the `@` prefix.**
+
+**Reason:** Moving files around becomes easier since you don't also have to update those import statements. This makes modifications to the codebase much nicer.
+
+```typescript
+// ✅ Good
+import { Button } from "@/components/ui/button";
+import { useAuth } from "@/hooks/useAuth";
+import { Text } from "@/refresh-components/texts/Text";
+
+// ❌ Bad
+import { Button } from "../../../components/ui/button";
+import { useAuth } from "./hooks/useAuth";
+```
+
+### 2. React Component Functions
+
+**Prefer regular functions over arrow functions for React components.**
+
+**Reason:** Functions just become easier to read.
+
+```typescript
+// ✅ Good
+function UserProfile({ userId }: UserProfileProps) {
+  return <div>User Profile</div>
+}
+
+// ❌ Bad
+const UserProfile = ({ userId }: UserProfileProps) => {
+  return <div>User Profile</div>
+}
+```
+
+### 3. Props Interface Extraction
+
+**Extract prop types into their own interface definitions.**
+
+**Reason:** Functions just become easier to read.
+
+```typescript
+// ✅ Good
+interface UserCardProps {
+  user: User
+  showActions?: boolean
+  onEdit?: (userId: string) => void
+}
+
+function UserCard({ user, showActions = false, onEdit }: UserCardProps) {
+  return <div>User Card</div>
+}
+
+// ❌ Bad
+function UserCard({
+  user,
+  showActions = false,
+  onEdit
+}: {
+  user: User
+  showActions?: boolean
+  onEdit?: (userId: string) => void
+}) {
+  return <div>User Card</div>
+}
+```
+
+### 4. Spacing Guidelines
+
+**Prefer padding over margins for spacing.**
+
+**Reason:** We want to consolidate usage to paddings instead of margins.
+
+```typescript
+// ✅ Good
+<div className="p-4 space-y-2">
+  <div className="p-2">Content</div>
+</div>
+
+// ❌ Bad
+<div className="m-4 space-y-2">
+  <div className="m-2">Content</div>
+</div>
+```
+
+### 5. Tailwind Dark Mode
+
+**Strictly forbid using the `dark:` modifier in Tailwind classes, except for logo icon handling.**
+
+**Reason:** The `colors.css` file already, VERY CAREFULLY, defines what the exact opposite colour of each light-mode colour is. Overriding this behaviour is VERY bad and will lead to horrible UI breakages.
+
+**Exception:** The `createLogoIcon` helper in `web/src/components/icons/icons.tsx` uses `dark:` modifiers (`dark:invert`, `dark:hidden`, `dark:block`) to handle third-party logo icons that cannot automatically adapt through `colors.css`. This is the ONLY acceptable use of dark mode modifiers.
+
+```typescript
+// ✅ Good - Standard components use `tailwind-themes/tailwind.config.js` / `src/app/css/colors.css`
+<div className="bg-background-neutral-03 text-text-02">
+  Content
+</div>
+
+// ✅ Good - Logo icons with dark mode handling via createLogoIcon
+export const GithubIcon = createLogoIcon(githubLightIcon, {
+  monochromatic: true,  // Will apply dark:invert internally
+});
+
+export const GitbookIcon = createLogoIcon(gitbookLightIcon, {
+  darkSrc: gitbookDarkIcon,  // Will use dark:hidden/dark:block internally
+});
+
+// ❌ Bad - Manual dark mode overrides
+<div className="bg-white dark:bg-black text-black dark:text-white">
+  Content
+</div>
+```
+
+### 6. Class Name Utilities
+
+**Use the `cn` utility instead of raw string formatting for classNames.**
+
+**Reason:** `cn`s are easier to read. They also allow for more complex types (i.e., string-arrays) to get formatted properly (it flattens each element in that string array down). As a result, it can allow things such as conditionals (i.e., `myCondition && "some-tailwind-class"`, which evaluates to `false` when `myCondition` is `false`) to get filtered out.
+
+```typescript
+import { cn } from '@/lib/utils'
+
+// ✅ Good
+<div className={cn(
+  'base-class',
+  isActive && 'active-class',
+  className
+)}>
+  Content
+</div>
+
+// ❌ Bad
+<div className={`base-class ${isActive ? 'active-class' : ''} ${className}`}>
+  Content
+</div>
+```
+
+### 7. Custom Hooks Organization
+
+**Follow a "hook-per-file" layout. Each hook should live in its own file within `web/src/hooks`.**
+
+**Reason:** This is just a layout preference. Keeps code clean.
+
+```typescript
+// web/src/hooks/useUserData.ts
+export function useUserData(userId: string) {
+  // hook implementation
+}
+
+// web/src/hooks/useLocalStorage.ts
+export function useLocalStorage<T>(key: string, initialValue: T) {
+  // hook implementation
+}
+```
+
+### 8. Icon Usage
+
+**ONLY use icons from the `web/src/icons` directory. Do NOT use icons from `react-icons`, `lucide`, or other external libraries.**
+
+**Reason:** We have a very carefully curated selection of icons that match our Onyx guidelines. We do NOT want to muddy those up with different aesthetic stylings.
+
+```typescript
+// ✅ Good
+import SvgX from "@/icons/x";
+import SvgMoreHorizontal from "@/icons/more-horizontal";
+
+// ❌ Bad
+import { User } from "lucide-react";
+import { FiSearch } from "react-icons/fi";
+```
+
+**Missing Icons**: If an icon is needed but doesn't exist in the `web/src/icons` directory, import it from Figma using the Figma MCP tool and add it to the icons directory.
+If you need help with this step, reach out to `raunak@onyx.app`.
+
+### 9. Text Rendering
+
+**Prefer using the `refresh-components/texts/Text` component for all text rendering. Avoid "naked" text nodes.**
+
+**Reason:** The `Text` component is fully compliant with the stylings provided in Figma. It provides easy utilities to specify the text-colour and font-size in the form of flags. Super duper easy.
+
+```typescript
+// ✅ Good
+import { Text } from '@/refresh-components/texts/Text'
+
+function UserCard({ name }: { name: string }) {
+  return (
+    <Text
+      {/* The `text03` flag makes the text it renders to be coloured the 3rd-scale grey */}
+      text03
+      {/* The `mainAction` flag makes the text it renders to be "main-action" font + line-height + weightage, as described in the Figma */}
+      mainAction
+    >
+      {name}
+    </Text>
+  )
+}
+
+// ❌ Bad
+function UserCard({ name }: { name: string }) {
+  return (
+    <div>
+      <h2>{name}</h2>
+      <p>User details</p>
+    </div>
+  )
+}
+```
+
+### 10. Component Usage
+
+**Heavily avoid raw HTML input components. Always use components from the `web/src/refresh-components` or `web/lib/opal/src` directory.**
+
+**Reason:** We've put in a lot of effort to unify the components that are rendered in the Onyx app. Using raw components breaks the entire UI of the application, and leaves it in a muddier state than before.
+
+```typescript
+// ✅ Good
+import Button from '@/refresh-components/buttons/Button'
+import InputTypeIn from '@/refresh-components/inputs/InputTypeIn'
+import SvgPlusCircle from '@/icons/plus-circle'
+
+function ContactForm() {
+  return (
+    <form>
+      <InputTypeIn placeholder="Search..." />
+      <Button type="submit" leftIcon={SvgPlusCircle}>Submit</Button>
+    </form>
+  )
+}
+
+// ❌ Bad
+function ContactForm() {
+  return (
+    <form>
+      <input placeholder="Name" />
+      <textarea placeholder="Message" />
+      <button type="submit">Submit</button>
+    </form>
+  )
+}
+```
+
+### 11. Colors
+
+**Always use custom overrides for colors and borders rather than built in Tailwind CSS colors. These overrides live in `web/tailwind-themes/tailwind.config.js`.**
+
+**Reason:** Our custom color system uses CSS variables that automatically handle dark mode and maintain design consistency across the app. Standard Tailwind colors bypass this system.
+
+**Available color categories:**
+- **Text:** `text-01` through `text-05`, `text-inverted-XX`
+- **Backgrounds:** `background-neutral-XX`, `background-tint-XX` (and inverted variants)
+- **Borders:** `border-01` through `border-05`, `border-inverted-XX`
+- **Actions:** `action-link-XX`, `action-danger-XX`
+- **Status:** `status-info-XX`, `status-success-XX`, `status-warning-XX`, `status-error-XX`
+- **Theme:** `theme-primary-XX`, `theme-red-XX`, `theme-blue-XX`, etc.
+
+```typescript
+// ✅ Good - Use custom Onyx color classes
+<div className="bg-background-neutral-01 border border-border-02" />
+<div className="bg-background-tint-02 border border-border-01" />
+<div className="bg-status-success-01" />
+<div className="bg-action-link-01" />
+<div className="bg-theme-primary-05" />
+
+// ❌ Bad - Do NOT use standard Tailwind colors
+<div className="bg-gray-100 border border-gray-300 text-gray-600" />
+<div className="bg-white border border-slate-200" />
+<div className="bg-green-100 text-green-700" />
+<div className="bg-blue-100 text-blue-600" />
+<div className="bg-indigo-500" />
+```
+
+### 12. Data Fetching
+
+**Prefer using `useSWR` for data fetching. Data should generally be fetched on the client side. Components that need data should display a loader / placeholder while waiting for that data. Prefer loading data within the component that needs it rather than at the top level and passing it down.**
+
+**Reason:** Client side fetching allows us to load the skeleton of the page without waiting for data to load, leading to a snappier UX. Loading data where needed reduces dependencies between a component and its parent component(s).
+
+## Database & Migrations
+
+### Running Migrations
+```bash
+# Standard migrations
+alembic upgrade head
+
+# Multi-tenant (Enterprise)
+alembic -n schema_private upgrade head
+```
+
+### Creating Migrations
+```bash
+# Create migration
+alembic revision -m "description"
+
+# Multi-tenant migration
+alembic -n schema_private revision -m "description"
+```
+
+Write the migration manually and place it in the file that alembic creates when running the above command.
+
+## Testing Strategy
+
+First, you must activate the virtual environment with `source .venv/bin/activate`. 
+
+There are 4 main types of tests within Onyx:
+
+### Unit Tests
+These should not assume any Onyx/external services are available to be called.
+Interactions with the outside world should be mocked using `unittest.mock`. Generally, only 
+write these for complex, isolated modules e.g. `citation_processing.py`.
+
+To run them:
+
+```bash
+pytest -xv backend/tests/unit
+```
+
+### External Dependency Unit Tests
+These tests assume that all external dependencies of Onyx are available and callable (e.g. Postgres, Redis, 
+MinIO/S3, Vespa are running + OpenAI can be called + any request to the internet is fine + etc.).
+
+However, the actual Onyx containers are not running and with these tests we call the function to test directly.
+We can also mock components/calls at will. 
+
+The goal with these tests are to minimize mocking while giving some flexibility to mock things that are flakey, 
+need strictly controlled behavior, or need to have their internal behavior validated (e.g. verify a function is called
+with certain args, something that would be impossible with proper integration tests).
+
+A great example of this type of test is `backend/tests/external_dependency_unit/connectors/confluence/test_confluence_group_sync.py`.
+
+To run them:
+
+```bash
+python -m dotenv -f .vscode/.env run -- pytest backend/tests/external_dependency_unit
+```
+
+### Integration Tests
+Standard integration tests. Every test in `backend/tests/integration` runs against a real Onyx deployment. We cannot 
+mock anything in these tests. Prefer writing integration tests (or External Dependency Unit Tests if mocking/internal 
+verification is necessary) over any other type of test.
+
+Tests are parallelized at a directory level.
+
+When writing integration tests, make sure to check the root `conftest.py` for useful fixtures + the `backend/tests/integration/common_utils` directory for utilities. Prefer (if one exists), calling the appropriate Manager 
+class in the utils over directly calling the APIs with a library like `requests`. Prefer using fixtures rather than
+calling the utilities directly (e.g. do NOT create admin users with 
+`admin_user = UserManager.create(name="admin_user")`, instead use the `admin_user` fixture).
+
+A great example of this type of test is `backend/tests/integration/dev_apis/test_simple_chat_api.py`.
+
+To run them:
+
+```bash
+python -m dotenv -f .vscode/.env run -- pytest backend/tests/integration
+```
+
+### Playwright (E2E) Tests
+These tests are an even more complete version of the Integration Tests mentioned above. Has all services of Onyx 
+running, *including* the Web Server.
+
+Use these tests for anything that requires significant frontend <-> backend coordination.
+
+Tests are located at `web/tests/e2e`. Tests are written in TypeScript.
+
+To run them:
+
+```bash
+npx playwright test <TEST_NAME>
+```
+
+
+## Logs
+
+When (1) writing integration tests or (2) doing live tests (e.g. curl / playwright) you can get access
+to logs via the `backend/log/<service_name>_debug.log` file. All Onyx services (api_server, web_server, celery_X)
+will be tailing their logs to this file. 
+
+
+## Security Considerations
+
+- Never commit API keys or secrets to repository
+- Use encrypted credential storage for connector credentials
+- Follow RBAC patterns for new features
+- Implement proper input validation with Pydantic models
+- Use parameterized queries to prevent SQL injection
+
+## AI/LLM Integration
+
+- Multiple LLM providers supported via LiteLLM
+- Configurable models per feature (chat, search, embeddings)
+- Streaming support for real-time responses
+- Token management and rate limiting
+- Custom prompts and agent actions
+
+## Creating a Plan
+When creating a plan in the `plans` directory, make sure to include at least these elements:
+
+**Issues to Address**
+What the change is meant to do.
+
+**Important Notes**
+Things you come across in your research that are important to the implementation.
+
+**Implementation strategy**
+How you are going to make the changes happen. High level approach.
+
+**Tests**
+What unit (use rarely), external dependency unit, integration, and playwright tests you plan to write to 
+verify the correct behavior. Don't overtest. Usually, a given change only needs one type of test.
+
+Do NOT include these: *Timeline*, *Rollback plan*
+
+This is a minimal list - feel free to include more. Do NOT write code as part of your plan.
+Keep it high level. You can reference certain files or functions though.
+
+Before writing your plan, make sure to do research. Explore the relevant sections in the codebase.

CONTRIBUTING.md

@@ -13,8 +13,7 @@ As an open source project in a rapidly changing space, we welcome all contributi
 The [GitHub Issues](https://github.com/onyx-dot-app/onyx/issues) page is a great place to start for contribution ideas.
 
 To ensure that your contribution is aligned with the project's direction, please reach out to any maintainer on the Onyx team
-via [Slack](https://join.slack.com/t/onyx-dot-app/shared_invite/zt-34lu4m7xg-TsKGO6h8PDvR5W27zTdyhA) /
-[Discord](https://discord.gg/TDJ59cGV2X) or [email](mailto:founders@onyx.app).
+via [Discord](https://discord.gg/4NA5SbzrWb) or [email](mailto:hello@onyx.app).
 
 Issues that have been explicitly approved by the maintainers (aligned with the direction of the project)
 will be marked with the `approved by maintainers` label.
@@ -28,8 +27,7 @@ Your input is vital to making sure that Onyx moves in the right direction.
 Before starting on implementation, please raise a GitHub issue.
 
 Also, always feel free to message the founders (Chris Weaver / Yuhong Sun) on
-[Slack](https://join.slack.com/t/onyx-dot-app/shared_invite/zt-34lu4m7xg-TsKGO6h8PDvR5W27zTdyhA) /
-[Discord](https://discord.gg/TDJ59cGV2X) directly about anything at all.
+[Discord](https://discord.gg/4NA5SbzrWb) directly about anything at all.
 
 ### Contributing Code
 
@@ -46,9 +44,7 @@ Our goal is to make contributing as easy as possible. If you run into any issues
 That way we can help future contributors and users can avoid the same issue.
 
 We also have support channels and generally interesting discussions on our
-[Slack](https://join.slack.com/t/onyx-dot-app/shared_invite/zt-2twesxdr6-5iQitKZQpgq~hYIZ~dv3KA)
-and
-[Discord](https://discord.gg/TDJ59cGV2X).
+[Discord](https://discord.gg/4NA5SbzrWb).
 
 We would love to see you there!
 
@@ -75,12 +71,12 @@ If using a higher version, sometimes some libraries will not be available (i.e.
 
 #### Backend: Python requirements
 
-Currently, we use pip and recommend creating a virtual environment.
+Currently, we use [uv](https://docs.astral.sh/uv/) and recommend creating a [virtual environment](https://docs.astral.sh/uv/pip/environments/#using-a-virtual-environment).
 
 For convenience here's a command for it:
 
 ```bash
-python -m venv .venv
+uv venv .venv --python 3.11
 source .venv/bin/activate
 ```
 
@@ -99,26 +95,26 @@ If using PowerShell, the command slightly differs:
 Install the required python dependencies:
 
 ```bash
-pip install -r backend/requirements/default.txt
-pip install -r backend/requirements/dev.txt
-pip install -r backend/requirements/ee.txt
-pip install -r backend/requirements/model_server.txt
+uv sync --all-extras
 ```
 
-Install Playwright for Python (headless browser required by the Web Connector)
-
-In the activated Python virtualenv, install Playwright for Python by running:
+Install Playwright for Python (headless browser required by the Web Connector):
 
 ```bash
-playwright install
+uv run playwright install
 ```
 
-You may have to deactivate and reactivate your virtualenv for `playwright` to appear on your path.
-
 #### Frontend: Node dependencies
 
-Install [Node.js and npm](https://docs.npmjs.com/downloading-and-installing-node-js-and-npm) for the frontend.
-Once the above is done, navigate to `onyx/web` run:
+Onyx uses Node v22.20.0. We highly recommend you use [Node Version Manager (nvm)](https://github.com/nvm-sh/nvm)
+to manage your Node installations. Once installed, you can run
+
+```bash
+nvm install 22 && nvm use 22
+node -v # verify your active version
+```
+
+Navigate to `onyx/web` and run:
 
 ```bash
 npm i
@@ -129,36 +125,30 @@ npm i
 ### Backend
 
 For the backend, you'll need to setup pre-commit hooks (black / reorder-python-imports).
-First, install pre-commit (if you don't have it already) following the instructions
-[here](https://pre-commit.com/#installation).
 
-With the virtual environment active, install the pre-commit library with:
+Then run:
 
 ```bash
-pip install pre-commit
-```
-
-Then, from the `onyx/backend` directory, run:
-
-```bash
-pre-commit install
+uv run pre-commit install
 ```
 
 Additionally, we use `mypy` for static type checking.
 Onyx is fully type-annotated, and we want to keep it that way!
-To run the mypy checks manually, run `python -m mypy .` from the `onyx/backend` directory.
+To run the mypy checks manually, run `uv run mypy .` from the `onyx/backend` directory.
 
 ### Web
 
-We use `prettier` for formatting. The desired version (2.8.8) will be installed via a `npm i` from the `onyx/web` directory.
+We use `prettier` for formatting. The desired version will be installed via a `npm i` from the `onyx/web` directory.
 To run the formatter, use `npx prettier --write .` from the `onyx/web` directory.
-Please double check that prettier passes before creating a pull request.
+
+Pre-commit will also run prettier automatically on files you've recently touched. If re-formatted, your commit will fail.
+Re-stage your changes and commit again.
 
 # Running the application for development
 
 ## Developing using VSCode Debugger (recommended)
 
-We highly recommend using VSCode debugger for development.
+**We highly recommend using VSCode debugger for development.**
 See [CONTRIBUTING_VSCODE.md](./CONTRIBUTING_VSCODE.md) for more details.
 
 Otherwise, you can follow the instructions below to run the application for development.
@@ -171,7 +161,7 @@ You will need Docker installed to run these containers.
 First navigate to `onyx/deployment/docker_compose`, then start up Postgres/Vespa/Redis/MinIO with:
 
 ```bash
-docker compose up -d index relational_db cache minio
+docker compose -f docker-compose.yml -f docker-compose.dev.yml up -d index relational_db cache minio
 ```
 
 (index refers to Vespa, relational_db refers to Postgres, and cache refers to Redis)

CONTRIBUTING_VSCODE.md

@@ -21,6 +21,9 @@ Before starting, make sure the Docker Daemon is running.
 5. You can set breakpoints by clicking to the left of line numbers to help debug while the app is running
 6. Use the debug toolbar to step through code, inspect variables, etc.
 
+Note: Clear and Restart External Volumes and Containers will reset your postgres and Vespa (relational-db and index).
+Only run this if you are okay with wiping your data.
+
 ## Features
 
 - Hot reload is enabled for the web server and API servers

README.md

@@ -1,29 +1,34 @@
 <a name="readme-top"></a>
 
 <h2 align="center">
-    <a href="https://www.onyx.app/"> <img width="50%" src="https://github.com/onyx-dot-app/onyx/blob/logo/OnyxLogoCropped.jpg?raw=true)" /></a>
+    <a href="https://www.onyx.app/?utm_source=onyx_repo&utm_medium=github&utm_campaign=readme"> <img width="50%" src="https://github.com/onyx-dot-app/onyx/blob/logo/OnyxLogoCropped.jpg?raw=true" /></a>
 </h2>
 
 <p align="center">Open Source AI Platform</p>
 
 <p align="center">
     <a href="https://discord.gg/TDJ59cGV2X" target="_blank">
-        <img src="https://img.shields.io/badge/discord-join-blue.svg?logo=discord&logoColor=white" alt="Discord">
+        <img src="https://img.shields.io/badge/discord-join-blue.svg?logo=discord&logoColor=white" alt="Discord" />
     </a>
-    <a href="https://docs.onyx.app/" target="_blank">
-        <img src="https://img.shields.io/badge/docs-view-blue" alt="Documentation">
+    <a href="https://docs.onyx.app/?utm_source=onyx_repo&utm_medium=github&utm_campaign=readme" target="_blank">
+        <img src="https://img.shields.io/badge/docs-view-blue" alt="Documentation" />
     </a>
-    <a href="https://docs.onyx.app/" target="_blank">
-        <img src="https://img.shields.io/website?url=https://www.onyx.app&up_message=visit&up_color=blue" alt="Documentation">
+    <a href="https://www.onyx.app/?utm_source=onyx_repo&utm_medium=github&utm_campaign=readme" target="_blank">
+        <img src="https://img.shields.io/website?url=https://www.onyx.app&up_message=visit&up_color=blue" alt="Documentation" />
     </a>
     <a href="https://github.com/onyx-dot-app/onyx/blob/main/LICENSE" target="_blank">
-        <img src="https://img.shields.io/static/v1?label=license&message=MIT&color=blue" alt="License">
+        <img src="https://img.shields.io/static/v1?label=license&message=MIT&color=blue" alt="License" />
     </a>
 </p>
 
+<p align="center">
+  <a href="https://trendshift.io/repositories/12516" target="_blank">
+    <img src="https://trendshift.io/api/badge/repositories/12516" alt="onyx-dot-app/onyx | Trendshift" style="width: 250px; height: 55px;" />
+  </a>
+</p>
 
 
-**[Onyx](https://www.onyx.app/)** is a feature-rich, self-hostable Chat UI that works with any LLM. It is easy to deploy and can run in a completely airgapped environment.
+**[Onyx](https://www.onyx.app/?utm_source=onyx_repo&utm_medium=github&utm_campaign=readme)** is a feature-rich, self-hostable Chat UI that works with any LLM. It is easy to deploy and can run in a completely airgapped environment.
 
 Onyx comes loaded with advanced features like Agents, Web Search, RAG, MCP, Deep Research, Connectors to 40+ knowledge sources, and more.
 
@@ -52,7 +57,7 @@ Onyx comes loaded with advanced features like Agents, Web Search, RAG, MCP, Deep
 
 Onyx works with all LLMs (like OpenAI, Anthropic, Gemini, etc.) and self-hosted LLMs (like Ollama, vLLM, etc.)
 
-To learn more about the features, check out our [documentation](https://docs.onyx.app/welcome)!
+To learn more about the features, check out our [documentation](https://docs.onyx.app/welcome?utm_source=onyx_repo&utm_medium=github&utm_campaign=readme)!
 
 
 
@@ -60,13 +65,13 @@ To learn more about the features, check out our [documentation](https://docs.ony
 Onyx supports deployments in Docker, Kubernetes, Terraform, along with guides for major cloud providers.
 
 See guides below:
-- [Docker](https://docs.onyx.app/deployment/local/docker) or [Quickstart](https://docs.onyx.app/deployment/getting_started/quickstart) (best for most users)
-- [Kubernetes](https://docs.onyx.app/deployment/local/kubernetes) (best for large teams)
-- [Terraform](https://docs.onyx.app/deployment/local/terraform) (best for teams already using Terraform)
-- Cloud specific guides (best if specifically using [AWS EKS](https://docs.onyx.app/deployment/cloud/aws/eks), [Azure VMs](https://docs.onyx.app/deployment/cloud/azure), etc.)
+- [Docker](https://docs.onyx.app/deployment/local/docker?utm_source=onyx_repo&utm_medium=github&utm_campaign=readme) or [Quickstart](https://docs.onyx.app/deployment/getting_started/quickstart?utm_source=onyx_repo&utm_medium=github&utm_campaign=readme) (best for most users)
+- [Kubernetes](https://docs.onyx.app/deployment/local/kubernetes?utm_source=onyx_repo&utm_medium=github&utm_campaign=readme) (best for large teams)
+- [Terraform](https://docs.onyx.app/deployment/local/terraform?utm_source=onyx_repo&utm_medium=github&utm_campaign=readme) (best for teams already using Terraform)
+- Cloud specific guides (best if specifically using [AWS EKS](https://docs.onyx.app/deployment/cloud/aws/eks?utm_source=onyx_repo&utm_medium=github&utm_campaign=readme), [Azure VMs](https://docs.onyx.app/deployment/cloud/azure?utm_source=onyx_repo&utm_medium=github&utm_campaign=readme), etc.)
 
 > [!TIP]  
-> **To try Onyx for free without deploying, check out [Onyx Cloud](https://cloud.onyx.app/signup)**.
+> **To try Onyx for free without deploying, check out [Onyx Cloud](https://cloud.onyx.app/signup?utm_source=onyx_repo&utm_medium=github&utm_campaign=readme)**.
 
 
 
@@ -90,7 +95,7 @@ There are two editions of Onyx:
 
 - Onyx Community Edition (CE) is available freely under the MIT license.
 - Onyx Enterprise Edition (EE) includes extra features that are primarily useful for larger organizations.
-For feature details, check out [our website](https://www.onyx.app/pricing).
+For feature details, check out [our website](https://www.onyx.app/pricing?utm_source=onyx_repo&utm_medium=github&utm_campaign=readme).
 
 
 

backend/.dockerignore

@@ -15,3 +15,4 @@ build/
 dist/
 .coverage
 htmlcov/
+model_server/legacy/

backend/Dockerfile

@@ -7,16 +7,20 @@ have a contract or agreement with DanswerAI, you are not permitted to use the En
 Edition features outside of personal development or testing purposes. Please reach out to \
 founders@onyx.app for more information. Please visit https://github.com/onyx-dot-app/onyx"
 
-# Default ONYX_VERSION, typically overriden during builds by GitHub Actions.
-ARG ONYX_VERSION=0.0.0-dev
 # DO_NOT_TRACK is used to disable telemetry for Unstructured
-ENV ONYX_VERSION=${ONYX_VERSION} \
-    DANSWER_RUNNING_IN_DOCKER="true" \
+ENV DANSWER_RUNNING_IN_DOCKER="true" \
     DO_NOT_TRACK="true" \
     PLAYWRIGHT_BROWSERS_PATH="/app/.cache/ms-playwright"
 
+# Create non-root user for security best practices
+RUN groupadd -g 1001 onyx && \
+    useradd -u 1001 -g onyx -m -s /bin/bash onyx && \
+    mkdir -p /var/log/onyx && \
+    chmod 755 /var/log/onyx && \
+    chown onyx:onyx /var/log/onyx
+
+COPY --from=ghcr.io/astral-sh/uv:0.9.9 /uv /uvx /bin/
 
-RUN echo "ONYX_VERSION: ${ONYX_VERSION}"
 # Install system dependencies
 # cmake needed for psycopg (postgres)
 # libpq-dev needed for psycopg (postgres)
@@ -48,22 +52,20 @@ RUN apt-get update && \
 # Remove py which is pulled in by retry, py is not needed and is a CVE
 COPY ./requirements/default.txt /tmp/requirements.txt
 COPY ./requirements/ee.txt /tmp/ee-requirements.txt
-RUN pip install --no-cache-dir --upgrade \
-        --retries 5 \
-        --timeout 30 \
+RUN uv pip install --system --no-cache-dir --upgrade \
         -r /tmp/requirements.txt \
         -r /tmp/ee-requirements.txt && \
     pip uninstall -y py && \
     playwright install chromium && \
     playwright install-deps chromium && \
-    ln -s /usr/local/bin/supervisord /usr/bin/supervisord
-
-# Cleanup for CVEs and size reduction
-# https://github.com/tornadoweb/tornado/issues/3107
-# xserver-common and xvfb included by playwright installation but not needed after
-# perl-base is part of the base Python Debian image but not needed for Onyx functionality
-# perl-base could only be removed with --allow-remove-essential
-RUN apt-get update && \
+    chown -R onyx:onyx /app && \
+    ln -s /usr/local/bin/supervisord /usr/bin/supervisord && \
+    # Cleanup for CVEs and size reduction
+    # https://github.com/tornadoweb/tornado/issues/3107
+    # xserver-common and xvfb included by playwright installation but not needed after
+    # perl-base is part of the base Python Debian image but not needed for Onyx functionality
+    # perl-base could only be removed with --allow-remove-essential
+    apt-get update && \
     apt-get remove -y --allow-remove-essential \
         perl-base \
         xserver-common \
@@ -73,15 +75,16 @@ RUN apt-get update && \
         libxmlsec1-dev \
         pkg-config \
         gcc && \
-    apt-get install -y libxmlsec1-openssl && \
+    # Install here to avoid some packages being cleaned up above
+    apt-get install -y \
+        libxmlsec1-openssl \
+        # Install postgresql-client for easy manual tests
+        postgresql-client && \
     apt-get autoremove -y && \
     rm -rf /var/lib/apt/lists/* && \
+    rm -rf ~/.cache/uv /tmp/*.txt && \
     rm -f /usr/local/lib/python3.11/site-packages/tornado/test/test.key
 
-# Install postgresql-client for easy manual tests
-# Install it here to avoid it being cleaned up above
-RUN apt-get update && apt-get install -y postgresql-client
-
 # Pre-downloading models for setups with limited egress
 RUN python -c "from tokenizers import Tokenizer; \
 Tokenizer.from_pretrained('nomic-ai/nomic-embed-text-v1')"
@@ -92,38 +95,40 @@ nltk.download('stopwords', quiet=True); \
 nltk.download('punkt_tab', quiet=True);"
 # nltk.download('wordnet', quiet=True); introduce this back if lemmatization is needed
 
+# Pre-downloading tiktoken for setups with limited egress
+RUN python -c "import tiktoken; \
+tiktoken.get_encoding('cl100k_base')"
+
 # Set up application files
 WORKDIR /app
 
 # Enterprise Version Files
-COPY ./ee /app/ee
+COPY --chown=onyx:onyx ./ee /app/ee
 COPY supervisord.conf /etc/supervisor/conf.d/supervisord.conf
 
 # Set up application files
-COPY ./onyx /app/onyx
-COPY ./shared_configs /app/shared_configs
-COPY ./alembic /app/alembic
-COPY ./alembic_tenants /app/alembic_tenants
-COPY ./alembic.ini /app/alembic.ini
+COPY --chown=onyx:onyx ./onyx /app/onyx
+COPY --chown=onyx:onyx ./shared_configs /app/shared_configs
+COPY --chown=onyx:onyx ./alembic /app/alembic
+COPY --chown=onyx:onyx ./alembic_tenants /app/alembic_tenants
+COPY --chown=onyx:onyx ./alembic.ini /app/alembic.ini
 COPY supervisord.conf /usr/etc/supervisord.conf
-COPY ./static /app/static
+COPY --chown=onyx:onyx ./static /app/static
 
 # Escape hatch scripts
-COPY ./scripts/debugging /app/scripts/debugging
-COPY ./scripts/force_delete_connector_by_id.py /app/scripts/force_delete_connector_by_id.py
+COPY --chown=onyx:onyx ./scripts/debugging /app/scripts/debugging
+COPY --chown=onyx:onyx ./scripts/force_delete_connector_by_id.py /app/scripts/force_delete_connector_by_id.py
+COPY --chown=onyx:onyx ./scripts/supervisord_entrypoint.sh /app/scripts/supervisord_entrypoint.sh
+RUN chmod +x /app/scripts/supervisord_entrypoint.sh
 
 # Put logo in assets
-COPY ./assets /app/assets
+COPY --chown=onyx:onyx ./assets /app/assets
 
 ENV PYTHONPATH=/app
 
-# Create non-root user for security best practices
-RUN groupadd -g 1001 onyx && \
-    useradd -u 1001 -g onyx -m -s /bin/bash onyx && \
-    chown -R onyx:onyx /app && \
-    mkdir -p /var/log/onyx && \
-    chmod 755 /var/log/onyx && \
-    chown onyx:onyx /var/log/onyx
+# Default ONYX_VERSION, typically overriden during builds by GitHub Actions.
+ARG ONYX_VERSION=0.0.0-dev
+ENV ONYX_VERSION=${ONYX_VERSION}
 
 # Default command which does nothing
 # This container is used by api server and background which specify their own CMD

backend/Dockerfile.model_server

@@ -1,4 +1,29 @@
-FROM python:3.11.7-slim-bookworm
+# Base stage with dependencies
+FROM python:3.11.7-slim-bookworm AS base
+
+ENV DANSWER_RUNNING_IN_DOCKER="true" \
+    HF_HOME=/app/.cache/huggingface
+
+COPY --from=ghcr.io/astral-sh/uv:0.9.9 /uv /uvx /bin/
+
+RUN mkdir -p /app/.cache/huggingface
+
+COPY ./requirements/model_server.txt /tmp/requirements.txt
+RUN uv pip install --system --no-cache-dir --upgrade \
+        -r /tmp/requirements.txt && \
+    rm -rf ~/.cache/uv /tmp/*.txt
+
+# Stage for downloading embedding models
+FROM base AS embedding-models
+RUN python -c "from huggingface_hub import snapshot_download; \
+snapshot_download('nomic-ai/nomic-embed-text-v1');"
+
+# Initialize SentenceTransformer to cache the custom architecture
+RUN python -c "from sentence_transformers import SentenceTransformer; \
+SentenceTransformer(model_name_or_path='nomic-ai/nomic-embed-text-v1', trust_remote_code=True);"
+
+# Final stage - combine all downloads
+FROM base AS final
 
 LABEL com.danswer.maintainer="founders@onyx.app"
 LABEL com.danswer.description="This image is for the Onyx model server which runs all of the \
@@ -6,68 +31,17 @@ AI models for Onyx. This container and all the code is MIT Licensed and free for
 You can find it at https://hub.docker.com/r/onyx/onyx-model-server. For more details, \
 visit https://github.com/onyx-dot-app/onyx."
 
-# Default ONYX_VERSION, typically overriden during builds by GitHub Actions.
-ARG ONYX_VERSION=0.0.0-dev
-ENV ONYX_VERSION=${ONYX_VERSION} \
-    DANSWER_RUNNING_IN_DOCKER="true" \
-    HF_HOME=/app/.cache/huggingface
-
-RUN echo "ONYX_VERSION: ${ONYX_VERSION}"
-
 # Create non-root user for security best practices
-RUN mkdir -p /app && \
-    groupadd -g 1001 onyx && \
-    useradd -u 1001 -g onyx -m -s /bin/bash onyx  && \
-    chown -R onyx:onyx /app && \
+RUN groupadd -g 1001 onyx && \
+    useradd -u 1001 -g onyx -m -s /bin/bash onyx && \
     mkdir -p /var/log/onyx && \
     chmod 755 /var/log/onyx && \
     chown onyx:onyx /var/log/onyx
 
-# --- add toolchain needed for Rust/Python builds (fastuuid) ---
-ENV RUSTUP_HOME=/usr/local/rustup \
-    CARGO_HOME=/usr/local/cargo \
-    PATH=/usr/local/cargo/bin:$PATH
-
-RUN set -eux; \
-    apt-get update && apt-get install -y --no-install-recommends \
-        build-essential \
-        pkg-config \
-        curl \
-        ca-certificates \
-    && rm -rf /var/lib/apt/lists/* \
-    # Install latest stable Rust (supports Cargo.lock v4)
-    && curl -sSf https://sh.rustup.rs | sh -s -- -y --profile minimal --default-toolchain stable \
-    && rustc --version && cargo --version
-
-COPY ./requirements/model_server.txt /tmp/requirements.txt
-RUN pip install --no-cache-dir --upgrade \
-        --retries 5 \
-        --timeout 30 \
-        -r /tmp/requirements.txt
-
-RUN apt-get remove -y --allow-remove-essential perl-base && \ 
-    apt-get autoremove -y
-
-# Pre-downloading models for setups with limited egress
-# Download tokenizers, distilbert for the Onyx model
-# Download model weights
-# Run Nomic to pull in the custom architecture and have it cached locally
-RUN python -c "from transformers import AutoTokenizer; \
-AutoTokenizer.from_pretrained('distilbert-base-uncased'); \
-AutoTokenizer.from_pretrained('mixedbread-ai/mxbai-rerank-xsmall-v1'); \
-from huggingface_hub import snapshot_download; \
-snapshot_download(repo_id='onyx-dot-app/hybrid-intent-token-classifier'); \
-snapshot_download(repo_id='onyx-dot-app/information-content-model'); \
-snapshot_download('nomic-ai/nomic-embed-text-v1'); \
-snapshot_download('mixedbread-ai/mxbai-rerank-xsmall-v1'); \
-from sentence_transformers import SentenceTransformer; \
-SentenceTransformer(model_name_or_path='nomic-ai/nomic-embed-text-v1', trust_remote_code=True);"
-
 # In case the user has volumes mounted to /app/.cache/huggingface that they've downloaded while
-# running Onyx, move the current contents of the cache folder to a temporary location to ensure 
+# running Onyx, move the current contents of the cache folder to a temporary location to ensure
 # it's preserved in order to combine with the user's cache contents
-RUN mv /app/.cache/huggingface /app/.cache/temp_huggingface && \
-    chown -R onyx:onyx /app
+COPY --chown=onyx:onyx --from=embedding-models /app/.cache/huggingface /app/.cache/temp_huggingface
 
 WORKDIR /app
 
@@ -86,4 +60,8 @@ COPY ./model_server /app/model_server
 
 ENV PYTHONPATH=/app
 
+# Default ONYX_VERSION, typically overriden during builds by GitHub Actions.
+ARG ONYX_VERSION=0.0.0-dev
+ENV ONYX_VERSION=${ONYX_VERSION}
+
 CMD ["uvicorn", "model_server.main:app", "--host", "0.0.0.0", "--port", "9000"]

backend/alembic/README.md

@@ -7,8 +7,12 @@ Onyx migrations use a generic single-database configuration with an async dbapi.
 
 ## To generate new migrations:
 
-run from onyx/backend:
-`alembic revision --autogenerate -m <DESCRIPTION_OF_MIGRATION>`
+From onyx/backend, run:
+`alembic revision -m <DESCRIPTION_OF_MIGRATION>`
+
+Note: you cannot use the `--autogenerate` flag as the automatic schema parsing does not work.
+
+Manually populate the upgrade and downgrade in your new migration.
 
 More info can be found here: https://alembic.sqlalchemy.org/en/latest/autogenerate.html
 

backend/alembic/env.py

@@ -39,7 +39,9 @@ config = context.config
 if config.config_file_name is not None and config.attributes.get(
     "configure_logger", True
 ):
-    fileConfig(config.config_file_name)
+    # disable_existing_loggers=False prevents breaking pytest's caplog fixture
+    # See: https://pytest-alembic.readthedocs.io/en/latest/setup.html#caplog-issues
+    fileConfig(config.config_file_name, disable_existing_loggers=False)
 
 target_metadata = [Base.metadata, ResultModelBase.metadata]
 
@@ -460,8 +462,49 @@ def run_migrations_offline() -> None:
 
 
 def run_migrations_online() -> None:
-    logger.info("run_migrations_online starting.")
-    asyncio.run(run_async_migrations())
+    """Run migrations in 'online' mode.
+
+    Supports pytest-alembic by checking for a pre-configured connection
+    in context.config.attributes["connection"]. If present, uses that
+    connection/engine directly instead of creating a new async engine.
+    """
+    # Check if pytest-alembic is providing a connection/engine
+    connectable = context.config.attributes.get("connection", None)
+
+    if connectable is not None:
+        # pytest-alembic is providing an engine - use it directly
+        logger.info("run_migrations_online starting (pytest-alembic mode).")
+
+        # For pytest-alembic, we use the default schema (public)
+        schema_name = context.config.attributes.get(
+            "schema_name", POSTGRES_DEFAULT_SCHEMA
+        )
+
+        # pytest-alembic passes an Engine, we need to get a connection from it
+        with connectable.connect() as connection:
+            # Set search path for the schema
+            connection.execute(text(f'SET search_path TO "{schema_name}"'))
+
+            context.configure(
+                connection=connection,
+                target_metadata=target_metadata,  # type: ignore
+                include_object=include_object,
+                version_table_schema=schema_name,
+                include_schemas=True,
+                compare_type=True,
+                compare_server_default=True,
+                script_location=config.get_main_option("script_location"),
+            )
+
+            with context.begin_transaction():
+                context.run_migrations()
+
+            # Commit the transaction to ensure changes are visible to next migration
+            connection.commit()
+    else:
+        # Normal operation - use async migrations
+        logger.info("run_migrations_online starting.")
+        asyncio.run(run_async_migrations())
 
 
 if context.is_offline_mode():

backend/alembic/versions/03d710ccf29c_add_permission_sync_attempt_tables.py

@@ -0,0 +1,153 @@
+"""add permission sync attempt tables
+
+Revision ID: 03d710ccf29c
+Revises: 96a5702df6aa
+Create Date: 2025-09-11 13:30:00.000000
+
+"""
+
+from alembic import op
+import sqlalchemy as sa
+
+
+# revision identifiers, used by Alembic.
+revision = "03d710ccf29c"  # Generate a new unique ID
+down_revision = "96a5702df6aa"
+branch_labels = None
+depends_on = None
+
+
+def upgrade() -> None:
+    # Create the permission sync status enum
+    permission_sync_status_enum = sa.Enum(
+        "not_started",
+        "in_progress",
+        "success",
+        "canceled",
+        "failed",
+        "completed_with_errors",
+        name="permissionsyncstatus",
+        native_enum=False,
+    )
+
+    # Create doc_permission_sync_attempt table
+    op.create_table(
+        "doc_permission_sync_attempt",
+        sa.Column("id", sa.Integer(), nullable=False),
+        sa.Column("connector_credential_pair_id", sa.Integer(), nullable=False),
+        sa.Column("status", permission_sync_status_enum, nullable=False),
+        sa.Column("total_docs_synced", sa.Integer(), nullable=True),
+        sa.Column("docs_with_permission_errors", sa.Integer(), nullable=True),
+        sa.Column("error_message", sa.Text(), nullable=True),
+        sa.Column(
+            "time_created",
+            sa.DateTime(timezone=True),
+            server_default=sa.text("now()"),
+            nullable=False,
+        ),
+        sa.Column("time_started", sa.DateTime(timezone=True), nullable=True),
+        sa.Column("time_finished", sa.DateTime(timezone=True), nullable=True),
+        sa.ForeignKeyConstraint(
+            ["connector_credential_pair_id"],
+            ["connector_credential_pair.id"],
+        ),
+        sa.PrimaryKeyConstraint("id"),
+    )
+
+    # Create indexes for doc_permission_sync_attempt
+    op.create_index(
+        "ix_doc_permission_sync_attempt_time_created",
+        "doc_permission_sync_attempt",
+        ["time_created"],
+        unique=False,
+    )
+    op.create_index(
+        "ix_permission_sync_attempt_latest_for_cc_pair",
+        "doc_permission_sync_attempt",
+        ["connector_credential_pair_id", "time_created"],
+        unique=False,
+    )
+    op.create_index(
+        "ix_permission_sync_attempt_status_time",
+        "doc_permission_sync_attempt",
+        ["status", sa.text("time_finished DESC")],
+        unique=False,
+    )
+
+    # Create external_group_permission_sync_attempt table
+    # connector_credential_pair_id is nullable - group syncs can be global (e.g., Confluence)
+    op.create_table(
+        "external_group_permission_sync_attempt",
+        sa.Column("id", sa.Integer(), nullable=False),
+        sa.Column("connector_credential_pair_id", sa.Integer(), nullable=True),
+        sa.Column("status", permission_sync_status_enum, nullable=False),
+        sa.Column("total_users_processed", sa.Integer(), nullable=True),
+        sa.Column("total_groups_processed", sa.Integer(), nullable=True),
+        sa.Column("total_group_memberships_synced", sa.Integer(), nullable=True),
+        sa.Column("error_message", sa.Text(), nullable=True),
+        sa.Column(
+            "time_created",
+            sa.DateTime(timezone=True),
+            server_default=sa.text("now()"),
+            nullable=False,
+        ),
+        sa.Column("time_started", sa.DateTime(timezone=True), nullable=True),
+        sa.Column("time_finished", sa.DateTime(timezone=True), nullable=True),
+        sa.ForeignKeyConstraint(
+            ["connector_credential_pair_id"],
+            ["connector_credential_pair.id"],
+        ),
+        sa.PrimaryKeyConstraint("id"),
+    )
+
+    # Create indexes for external_group_permission_sync_attempt
+    op.create_index(
+        "ix_external_group_permission_sync_attempt_time_created",
+        "external_group_permission_sync_attempt",
+        ["time_created"],
+        unique=False,
+    )
+    op.create_index(
+        "ix_group_sync_attempt_cc_pair_time",
+        "external_group_permission_sync_attempt",
+        ["connector_credential_pair_id", "time_created"],
+        unique=False,
+    )
+    op.create_index(
+        "ix_group_sync_attempt_status_time",
+        "external_group_permission_sync_attempt",
+        ["status", sa.text("time_finished DESC")],
+        unique=False,
+    )
+
+
+def downgrade() -> None:
+    # Drop indexes
+    op.drop_index(
+        "ix_group_sync_attempt_status_time",
+        table_name="external_group_permission_sync_attempt",
+    )
+    op.drop_index(
+        "ix_group_sync_attempt_cc_pair_time",
+        table_name="external_group_permission_sync_attempt",
+    )
+    op.drop_index(
+        "ix_external_group_permission_sync_attempt_time_created",
+        table_name="external_group_permission_sync_attempt",
+    )
+    op.drop_index(
+        "ix_permission_sync_attempt_status_time",
+        table_name="doc_permission_sync_attempt",
+    )
+    op.drop_index(
+        "ix_permission_sync_attempt_latest_for_cc_pair",
+        table_name="doc_permission_sync_attempt",
+    )
+    op.drop_index(
+        "ix_doc_permission_sync_attempt_time_created",
+        table_name="doc_permission_sync_attempt",
+    )
+
+    # Drop tables
+    op.drop_table("external_group_permission_sync_attempt")
+    op.drop_table("doc_permission_sync_attempt")

backend/alembic/versions/09995b8811eb_add_theme_preference_to_user.py

@@ -0,0 +1,33 @@
+"""add theme_preference to user
+
+Revision ID: 09995b8811eb
+Revises: 3d1cca026fe8
+Create Date: 2025-10-24 08:58:50.246949
+
+"""
+
+from alembic import op
+import sqlalchemy as sa
+from onyx.db.enums import ThemePreference
+
+
+# revision identifiers, used by Alembic.
+revision = "09995b8811eb"
+down_revision = "3d1cca026fe8"
+branch_labels = None
+depends_on = None
+
+
+def upgrade() -> None:
+    op.add_column(
+        "user",
+        sa.Column(
+            "theme_preference",
+            sa.Enum(ThemePreference, native_enum=False),
+            nullable=True,
+        ),
+    )
+
+
+def downgrade() -> None:
+    op.drop_column("user", "theme_preference")

backend/alembic/versions/0cd424f32b1d_user_file_data_preparation_and_backfill.py

@@ -0,0 +1,389 @@
+"""Migration 2: User file data preparation and backfill
+
+Revision ID: 0cd424f32b1d
+Revises: 9b66d3156fc6
+Create Date: 2025-09-22 09:44:42.727034
+
+This migration populates the new columns added in migration 1.
+It prepares data for the UUID transition and relationship migration.
+"""
+
+from alembic import op
+import sqlalchemy as sa
+from sqlalchemy import text
+import logging
+
+logger = logging.getLogger("alembic.runtime.migration")
+
+# revision identifiers, used by Alembic.
+revision = "0cd424f32b1d"
+down_revision = "9b66d3156fc6"
+branch_labels = None
+depends_on = None
+
+
+def upgrade() -> None:
+    """Populate new columns with data."""
+
+    bind = op.get_bind()
+    inspector = sa.inspect(bind)
+
+    # === Step 1: Populate user_file.new_id ===
+    user_file_columns = [col["name"] for col in inspector.get_columns("user_file")]
+    has_new_id = "new_id" in user_file_columns
+
+    if has_new_id:
+        logger.info("Populating user_file.new_id with UUIDs...")
+
+        # Count rows needing UUIDs
+        null_count = bind.execute(
+            text("SELECT COUNT(*) FROM user_file WHERE new_id IS NULL")
+        ).scalar_one()
+
+        if null_count > 0:
+            logger.info(f"Generating UUIDs for {null_count} user_file records...")
+
+            # Populate in batches to avoid long locks
+            batch_size = 10000
+            total_updated = 0
+
+            while True:
+                result = bind.execute(
+                    text(
+                        """
+                    UPDATE user_file
+                    SET new_id = gen_random_uuid()
+                    WHERE new_id IS NULL
+                    AND id IN (
+                        SELECT id FROM user_file
+                        WHERE new_id IS NULL
+                        LIMIT :batch_size
+                    )
+                """
+                    ),
+                    {"batch_size": batch_size},
+                )
+
+                updated = result.rowcount
+                total_updated += updated
+
+                if updated < batch_size:
+                    break
+
+                logger.info(f"  Updated {total_updated}/{null_count} records...")
+
+            logger.info(f"Generated UUIDs for {total_updated} user_file records")
+
+        # Verify all records have UUIDs
+        remaining_null = bind.execute(
+            text("SELECT COUNT(*) FROM user_file WHERE new_id IS NULL")
+        ).scalar_one()
+
+        if remaining_null > 0:
+            raise Exception(
+                f"Failed to populate all user_file.new_id values ({remaining_null} NULL)"
+            )
+
+        # Lock down the column
+        op.alter_column("user_file", "new_id", nullable=False)
+        op.alter_column("user_file", "new_id", server_default=None)
+        logger.info("Locked down user_file.new_id column")
+
+    # === Step 2: Populate persona__user_file.user_file_id_uuid ===
+    persona_user_file_columns = [
+        col["name"] for col in inspector.get_columns("persona__user_file")
+    ]
+
+    if has_new_id and "user_file_id_uuid" in persona_user_file_columns:
+        logger.info("Populating persona__user_file.user_file_id_uuid...")
+
+        # Count rows needing update
+        null_count = bind.execute(
+            text(
+                """
+            SELECT COUNT(*) FROM persona__user_file
+            WHERE user_file_id IS NOT NULL AND user_file_id_uuid IS NULL
+        """
+            )
+        ).scalar_one()
+
+        if null_count > 0:
+            logger.info(f"Updating {null_count} persona__user_file records...")
+
+            # Update in batches
+            batch_size = 10000
+            total_updated = 0
+
+            while True:
+                result = bind.execute(
+                    text(
+                        """
+                    UPDATE persona__user_file p
+                    SET user_file_id_uuid = uf.new_id
+                    FROM user_file uf
+                    WHERE p.user_file_id = uf.id
+                    AND p.user_file_id_uuid IS NULL
+                    AND p.persona_id IN (
+                        SELECT persona_id
+                        FROM persona__user_file
+                        WHERE user_file_id_uuid IS NULL
+                        LIMIT :batch_size
+                    )
+                """
+                    ),
+                    {"batch_size": batch_size},
+                )
+
+                updated = result.rowcount
+                total_updated += updated
+
+                if updated < batch_size:
+                    break
+
+                logger.info(f"  Updated {total_updated}/{null_count} records...")
+
+            logger.info(f"Updated {total_updated} persona__user_file records")
+
+        # Verify all records are populated
+        remaining_null = bind.execute(
+            text(
+                """
+            SELECT COUNT(*) FROM persona__user_file
+            WHERE user_file_id IS NOT NULL AND user_file_id_uuid IS NULL
+        """
+            )
+        ).scalar_one()
+
+        if remaining_null > 0:
+            raise Exception(
+                f"Failed to populate all persona__user_file.user_file_id_uuid values ({remaining_null} NULL)"
+            )
+
+        op.alter_column("persona__user_file", "user_file_id_uuid", nullable=False)
+        logger.info("Locked down persona__user_file.user_file_id_uuid column")
+
+    # === Step 3: Create user_project records from chat_folder ===
+    if "chat_folder" in inspector.get_table_names():
+        logger.info("Creating user_project records from chat_folder...")
+
+        result = bind.execute(
+            text(
+                """
+            INSERT INTO user_project (user_id, name)
+            SELECT cf.user_id, cf.name
+            FROM chat_folder cf
+            WHERE NOT EXISTS (
+                SELECT 1
+                FROM user_project up
+                WHERE up.user_id = cf.user_id AND up.name = cf.name
+            )
+        """
+            )
+        )
+
+        logger.info(f"Created {result.rowcount} user_project records from chat_folder")
+
+    # === Step 4: Populate chat_session.project_id ===
+    chat_session_columns = [
+        col["name"] for col in inspector.get_columns("chat_session")
+    ]
+
+    if "folder_id" in chat_session_columns and "project_id" in chat_session_columns:
+        logger.info("Populating chat_session.project_id...")
+
+        # Count sessions needing update
+        null_count = bind.execute(
+            text(
+                """
+            SELECT COUNT(*) FROM chat_session
+            WHERE project_id IS NULL AND folder_id IS NOT NULL
+        """
+            )
+        ).scalar_one()
+
+        if null_count > 0:
+            logger.info(f"Updating {null_count} chat_session records...")
+
+            result = bind.execute(
+                text(
+                    """
+                UPDATE chat_session cs
+                SET project_id = up.id
+                FROM chat_folder cf
+                JOIN user_project up ON up.user_id = cf.user_id AND up.name = cf.name
+                WHERE cs.folder_id = cf.id AND cs.project_id IS NULL
+            """
+                )
+            )
+
+            logger.info(f"Updated {result.rowcount} chat_session records")
+
+        # Verify all records are populated
+        remaining_null = bind.execute(
+            text(
+                """
+            SELECT COUNT(*) FROM chat_session
+            WHERE project_id IS NULL AND folder_id IS NOT NULL
+        """
+            )
+        ).scalar_one()
+
+        if remaining_null > 0:
+            logger.warning(
+                f"Warning: {remaining_null} chat_session records could not be mapped to projects"
+            )
+
+    # === Step 5: Update plaintext FileRecord IDs/display names to UUID scheme ===
+    # Prior to UUID migration, plaintext cache files were stored with file_id like 'plain_text_<int_id>'.
+    # After migration, we use 'plaintext_<uuid>' (note the name change to 'plaintext_').
+    # This step remaps existing FileRecord rows to the new naming while preserving object_key/bucket.
+    logger.info("Updating plaintext FileRecord ids and display names to UUID scheme...")
+
+    # Count legacy plaintext records that can be mapped to UUID user_file ids
+    count_query = text(
+        """
+        SELECT COUNT(*)
+        FROM file_record fr
+        JOIN user_file uf ON fr.file_id = CONCAT('plaintext_', uf.id::text)
+        WHERE LOWER(fr.file_origin::text) = 'plaintext_cache'
+        """
+    )
+    legacy_count = bind.execute(count_query).scalar_one()
+
+    if legacy_count and legacy_count > 0:
+        logger.info(f"Found {legacy_count} legacy plaintext file records to update")
+
+        # Update display_name first for readability (safe regardless of rename)
+        bind.execute(
+            text(
+                """
+                UPDATE file_record fr
+                SET display_name = CONCAT('Plaintext for user file ', uf.new_id::text)
+                FROM user_file uf
+                WHERE LOWER(fr.file_origin::text) = 'plaintext_cache'
+                    AND fr.file_id = CONCAT('plaintext_', uf.id::text)
+                """
+            )
+        )
+
+        # Remap file_id from 'plaintext_<int>' -> 'plaintext_<uuid>' using transitional new_id
+        # Use a single UPDATE ... WHERE file_id LIKE 'plain_text_%'
+        # and ensure it aligns to existing user_file ids to avoid renaming unrelated rows
+        result = bind.execute(
+            text(
+                """
+                UPDATE file_record fr
+                SET file_id = CONCAT('plaintext_', uf.new_id::text)
+                FROM user_file uf
+                WHERE LOWER(fr.file_origin::text) = 'plaintext_cache'
+                    AND fr.file_id = CONCAT('plaintext_', uf.id::text)
+                """
+            )
+        )
+        logger.info(
+            f"Updated {result.rowcount} plaintext file_record ids to UUID scheme"
+        )
+
+    # === Step 6: Ensure document_id_migrated default TRUE and backfill existing FALSE ===
+    # New records should default to migrated=True so the migration task won't run for them.
+    # Existing rows that had a legacy document_id should be marked as not migrated to be processed.
+
+    # Backfill existing records: if document_id is not null, set to FALSE
+    bind.execute(
+        text(
+            """
+            UPDATE user_file
+            SET document_id_migrated = FALSE
+            WHERE document_id IS NOT NULL
+            """
+        )
+    )
+
+    # === Step 7: Backfill user_file.status from index_attempt ===
+    logger.info("Backfilling user_file.status from index_attempt...")
+
+    # Update user_file status based on latest index attempt
+    # Using CTEs instead of temp tables for asyncpg compatibility
+    result = bind.execute(
+        text(
+            """
+        WITH latest_attempt AS (
+            SELECT DISTINCT ON (ia.connector_credential_pair_id)
+                ia.connector_credential_pair_id,
+                ia.status
+            FROM index_attempt ia
+            ORDER BY ia.connector_credential_pair_id, ia.time_updated DESC
+        ),
+        uf_to_ccp AS (
+            SELECT DISTINCT uf.id AS uf_id, ccp.id AS cc_pair_id
+            FROM user_file uf
+            JOIN document_by_connector_credential_pair dcc
+                ON dcc.id = REPLACE(uf.document_id, 'USER_FILE_CONNECTOR__', 'FILE_CONNECTOR__')
+            JOIN connector_credential_pair ccp
+                ON ccp.connector_id = dcc.connector_id
+                AND ccp.credential_id = dcc.credential_id
+        )
+        UPDATE user_file uf
+        SET status = CASE
+            WHEN la.status IN ('NOT_STARTED', 'IN_PROGRESS') THEN 'PROCESSING'
+            WHEN la.status = 'SUCCESS' THEN 'COMPLETED'
+            ELSE 'FAILED'
+        END
+        FROM uf_to_ccp ufc
+        LEFT JOIN latest_attempt la
+            ON la.connector_credential_pair_id = ufc.cc_pair_id
+        WHERE uf.id = ufc.uf_id
+        AND uf.status = 'PROCESSING'
+    """
+        )
+    )
+
+    logger.info(f"Updated status for {result.rowcount} user_file records")
+
+    logger.info("Migration 2 (data preparation) completed successfully")
+
+
+def downgrade() -> None:
+    """Reset populated data to allow clean downgrade of schema."""
+
+    bind = op.get_bind()
+    inspector = sa.inspect(bind)
+
+    logger.info("Starting downgrade of data preparation...")
+
+    # Reset user_file columns to allow nulls before data removal
+    if "user_file" in inspector.get_table_names():
+        columns = [col["name"] for col in inspector.get_columns("user_file")]
+
+        if "new_id" in columns:
+            op.alter_column(
+                "user_file",
+                "new_id",
+                nullable=True,
+                server_default=sa.text("gen_random_uuid()"),
+            )
+            # Optionally clear the data
+            # bind.execute(text("UPDATE user_file SET new_id = NULL"))
+            logger.info("Reset user_file.new_id to nullable")
+
+    # Reset persona__user_file.user_file_id_uuid
+    if "persona__user_file" in inspector.get_table_names():
+        columns = [col["name"] for col in inspector.get_columns("persona__user_file")]
+
+        if "user_file_id_uuid" in columns:
+            op.alter_column("persona__user_file", "user_file_id_uuid", nullable=True)
+            # Optionally clear the data
+            # bind.execute(text("UPDATE persona__user_file SET user_file_id_uuid = NULL"))
+            logger.info("Reset persona__user_file.user_file_id_uuid to nullable")
+
+    # Note: We don't delete user_project records or reset chat_session.project_id
+    # as these might be in use and can be handled by the schema downgrade
+
+    # Reset user_file.status to default
+    if "user_file" in inspector.get_table_names():
+        columns = [col["name"] for col in inspector.get_columns("user_file")]
+        if "status" in columns:
+            bind.execute(text("UPDATE user_file SET status = 'PROCESSING'"))
+            logger.info("Reset user_file.status to default")
+
+    logger.info("Downgrade completed successfully")

backend/alembic/versions/16c37a30adf2_user_file_relationship_migration.py

@@ -0,0 +1,261 @@
+"""Migration 3: User file relationship migration
+
+Revision ID: 16c37a30adf2
+Revises: 0cd424f32b1d
+Create Date: 2025-09-22 09:47:34.175596
+
+This migration converts folder-based relationships to project-based relationships.
+It migrates persona__user_folder to persona__user_file and populates project__user_file.
+"""
+
+from alembic import op
+import sqlalchemy as sa
+from sqlalchemy import text
+import logging
+
+logger = logging.getLogger("alembic.runtime.migration")
+
+# revision identifiers, used by Alembic.
+revision = "16c37a30adf2"
+down_revision = "0cd424f32b1d"
+branch_labels = None
+depends_on = None
+
+
+def upgrade() -> None:
+    """Migrate folder-based relationships to project-based relationships."""
+
+    bind = op.get_bind()
+    inspector = sa.inspect(bind)
+
+    # === Step 1: Migrate persona__user_folder to persona__user_file ===
+    table_names = inspector.get_table_names()
+
+    if "persona__user_folder" in table_names and "user_file" in table_names:
+        user_file_columns = [col["name"] for col in inspector.get_columns("user_file")]
+        has_new_id = "new_id" in user_file_columns
+
+        if has_new_id and "folder_id" in user_file_columns:
+            logger.info(
+                "Migrating persona__user_folder relationships to persona__user_file..."
+            )
+
+            # Count relationships to migrate (asyncpg-compatible)
+            count_query = text(
+                """
+                SELECT COUNT(*)
+                FROM (
+                    SELECT DISTINCT puf.persona_id, uf.id
+                    FROM persona__user_folder puf
+                    JOIN user_file uf ON uf.folder_id = puf.user_folder_id
+                    WHERE NOT EXISTS (
+                        SELECT 1
+                        FROM persona__user_file p2
+                        WHERE p2.persona_id = puf.persona_id
+                        AND p2.user_file_id = uf.id
+                    )
+                ) AS distinct_pairs
+            """
+            )
+            to_migrate = bind.execute(count_query).scalar_one()
+
+            if to_migrate > 0:
+                logger.info(f"Creating {to_migrate} persona-file relationships...")
+
+                # Migrate in batches to avoid memory issues
+                batch_size = 10000
+                total_inserted = 0
+
+                while True:
+                    # Insert batch directly using subquery (asyncpg compatible)
+                    result = bind.execute(
+                        text(
+                            """
+                        INSERT INTO persona__user_file (persona_id, user_file_id, user_file_id_uuid)
+                        SELECT DISTINCT puf.persona_id, uf.id as file_id, uf.new_id
+                        FROM persona__user_folder puf
+                        JOIN user_file uf ON uf.folder_id = puf.user_folder_id
+                        WHERE NOT EXISTS (
+                            SELECT 1
+                            FROM persona__user_file p2
+                            WHERE p2.persona_id = puf.persona_id
+                            AND p2.user_file_id = uf.id
+                        )
+                        LIMIT :batch_size
+                    """
+                        ),
+                        {"batch_size": batch_size},
+                    )
+
+                    inserted = result.rowcount
+                    total_inserted += inserted
+
+                    if inserted < batch_size:
+                        break
+
+                    logger.info(
+                        f"  Migrated {total_inserted}/{to_migrate} relationships..."
+                    )
+
+                logger.info(
+                    f"Created {total_inserted} persona__user_file relationships"
+                )
+
+    # === Step 2: Add foreign key for chat_session.project_id ===
+    chat_session_fks = inspector.get_foreign_keys("chat_session")
+    fk_exists = any(
+        fk["name"] == "fk_chat_session_project_id" for fk in chat_session_fks
+    )
+
+    if not fk_exists:
+        logger.info("Adding foreign key constraint for chat_session.project_id...")
+        op.create_foreign_key(
+            "fk_chat_session_project_id",
+            "chat_session",
+            "user_project",
+            ["project_id"],
+            ["id"],
+        )
+        logger.info("Added foreign key constraint")
+
+    # === Step 3: Populate project__user_file from user_file.folder_id ===
+    user_file_columns = [col["name"] for col in inspector.get_columns("user_file")]
+    has_new_id = "new_id" in user_file_columns
+
+    if has_new_id and "folder_id" in user_file_columns:
+        logger.info("Populating project__user_file from folder relationships...")
+
+        # Count relationships to create
+        count_query = text(
+            """
+            SELECT COUNT(*)
+            FROM user_file uf
+            WHERE uf.folder_id IS NOT NULL
+            AND NOT EXISTS (
+                SELECT 1
+                FROM project__user_file puf
+                WHERE puf.project_id = uf.folder_id
+                AND puf.user_file_id = uf.new_id
+            )
+        """
+        )
+        to_create = bind.execute(count_query).scalar_one()
+
+        if to_create > 0:
+            logger.info(f"Creating {to_create} project-file relationships...")
+
+            # Insert in batches
+            batch_size = 10000
+            total_inserted = 0
+
+            while True:
+                result = bind.execute(
+                    text(
+                        """
+                    INSERT INTO project__user_file (project_id, user_file_id)
+                    SELECT uf.folder_id, uf.new_id
+                    FROM user_file uf
+                    WHERE uf.folder_id IS NOT NULL
+                    AND NOT EXISTS (
+                        SELECT 1
+                        FROM project__user_file puf
+                        WHERE puf.project_id = uf.folder_id
+                        AND puf.user_file_id = uf.new_id
+                    )
+                    LIMIT :batch_size
+                    ON CONFLICT (project_id, user_file_id) DO NOTHING
+                """
+                    ),
+                    {"batch_size": batch_size},
+                )
+
+                inserted = result.rowcount
+                total_inserted += inserted
+
+                if inserted < batch_size:
+                    break
+
+                logger.info(f"  Created {total_inserted}/{to_create} relationships...")
+
+            logger.info(f"Created {total_inserted} project__user_file relationships")
+
+    # === Step 4: Create index on chat_session.project_id ===
+    try:
+        indexes = [ix.get("name") for ix in inspector.get_indexes("chat_session")]
+    except Exception:
+        indexes = []
+
+    if "ix_chat_session_project_id" not in indexes:
+        logger.info("Creating index on chat_session.project_id...")
+        op.create_index(
+            "ix_chat_session_project_id", "chat_session", ["project_id"], unique=False
+        )
+        logger.info("Created index")
+
+    logger.info("Migration 3 (relationship migration) completed successfully")
+
+
+def downgrade() -> None:
+    """Remove migrated relationships and constraints."""
+
+    bind = op.get_bind()
+    inspector = sa.inspect(bind)
+
+    logger.info("Starting downgrade of relationship migration...")
+
+    # Drop index on chat_session.project_id
+    try:
+        indexes = [ix.get("name") for ix in inspector.get_indexes("chat_session")]
+        if "ix_chat_session_project_id" in indexes:
+            op.drop_index("ix_chat_session_project_id", "chat_session")
+            logger.info("Dropped index on chat_session.project_id")
+    except Exception:
+        pass
+
+    # Drop foreign key constraint
+    try:
+        chat_session_fks = inspector.get_foreign_keys("chat_session")
+        fk_exists = any(
+            fk["name"] == "fk_chat_session_project_id" for fk in chat_session_fks
+        )
+        if fk_exists:
+            op.drop_constraint(
+                "fk_chat_session_project_id", "chat_session", type_="foreignkey"
+            )
+            logger.info("Dropped foreign key constraint on chat_session.project_id")
+    except Exception:
+        pass
+
+    # Clear project__user_file relationships (but keep the table for migration 1 to handle)
+    if "project__user_file" in inspector.get_table_names():
+        result = bind.execute(text("DELETE FROM project__user_file"))
+        logger.info(f"Cleared {result.rowcount} records from project__user_file")
+
+    # Remove migrated persona__user_file relationships
+    # Only remove those that came from folder relationships
+    if all(
+        table in inspector.get_table_names()
+        for table in ["persona__user_file", "persona__user_folder", "user_file"]
+    ):
+        user_file_columns = [col["name"] for col in inspector.get_columns("user_file")]
+        if "folder_id" in user_file_columns:
+            result = bind.execute(
+                text(
+                    """
+                DELETE FROM persona__user_file puf
+                WHERE EXISTS (
+                    SELECT 1
+                    FROM user_file uf
+                    JOIN persona__user_folder puf2
+                        ON puf2.user_folder_id = uf.folder_id
+                    WHERE puf.persona_id = puf2.persona_id
+                    AND puf.user_file_id = uf.id
+                )
+            """
+                )
+            )
+            logger.info(
+                f"Removed {result.rowcount} migrated persona__user_file relationships"
+            )
+
+    logger.info("Downgrade completed successfully")

backend/alembic/versions/18b5b2524446_add_is_clarification_to_chat_message.py

@@ -0,0 +1,29 @@
+"""add is_clarification to chat_message
+
+Revision ID: 18b5b2524446
+Revises: 87c52ec39f84
+Create Date: 2025-01-16
+
+"""
+
+from alembic import op
+import sqlalchemy as sa
+
+# revision identifiers, used by Alembic.
+revision = "18b5b2524446"
+down_revision = "87c52ec39f84"
+branch_labels = None
+depends_on = None
+
+
+def upgrade() -> None:
+    op.add_column(
+        "chat_message",
+        sa.Column(
+            "is_clarification", sa.Boolean(), nullable=False, server_default="false"
+        ),
+    )
+
+
+def downgrade() -> None:
+    op.drop_column("chat_message", "is_clarification")

backend/alembic/versions/1f2a3b4c5d6e_add_internet_search_and_content_providers.py

@@ -0,0 +1,89 @@
+"""add internet search and content provider tables
+
+Revision ID: 1f2a3b4c5d6e
+Revises: 9drpiiw74ljy
+Create Date: 2025-11-10 19:45:00.000000
+
+"""
+
+from alembic import op
+import sqlalchemy as sa
+from sqlalchemy.dialects import postgresql
+
+
+# revision identifiers, used by Alembic.
+revision = "1f2a3b4c5d6e"
+down_revision = "9drpiiw74ljy"
+branch_labels = None
+depends_on = None
+
+
+def upgrade() -> None:
+    op.create_table(
+        "internet_search_provider",
+        sa.Column("id", sa.Integer(), primary_key=True),
+        sa.Column("name", sa.String(), nullable=False, unique=True),
+        sa.Column("provider_type", sa.String(), nullable=False),
+        sa.Column("api_key", sa.LargeBinary(), nullable=True),
+        sa.Column("config", postgresql.JSONB(astext_type=sa.Text()), nullable=True),
+        sa.Column(
+            "is_active", sa.Boolean(), nullable=False, server_default=sa.text("false")
+        ),
+        sa.Column(
+            "time_created",
+            sa.DateTime(timezone=True),
+            nullable=False,
+            server_default=sa.text("now()"),
+        ),
+        sa.Column(
+            "time_updated",
+            sa.DateTime(timezone=True),
+            nullable=False,
+            server_default=sa.text("now()"),
+        ),
+    )
+    op.create_index(
+        "ix_internet_search_provider_is_active",
+        "internet_search_provider",
+        ["is_active"],
+    )
+
+    op.create_table(
+        "internet_content_provider",
+        sa.Column("id", sa.Integer(), primary_key=True),
+        sa.Column("name", sa.String(), nullable=False, unique=True),
+        sa.Column("provider_type", sa.String(), nullable=False),
+        sa.Column("api_key", sa.LargeBinary(), nullable=True),
+        sa.Column("config", postgresql.JSONB(astext_type=sa.Text()), nullable=True),
+        sa.Column(
+            "is_active", sa.Boolean(), nullable=False, server_default=sa.text("false")
+        ),
+        sa.Column(
+            "time_created",
+            sa.DateTime(timezone=True),
+            nullable=False,
+            server_default=sa.text("now()"),
+        ),
+        sa.Column(
+            "time_updated",
+            sa.DateTime(timezone=True),
+            nullable=False,
+            server_default=sa.text("now()"),
+        ),
+    )
+    op.create_index(
+        "ix_internet_content_provider_is_active",
+        "internet_content_provider",
+        ["is_active"],
+    )
+
+
+def downgrade() -> None:
+    op.drop_index(
+        "ix_internet_content_provider_is_active", table_name="internet_content_provider"
+    )
+    op.drop_table("internet_content_provider")
+    op.drop_index(
+        "ix_internet_search_provider_is_active", table_name="internet_search_provider"
+    )
+    op.drop_table("internet_search_provider")

backend/alembic/versions/23957775e5f5_remove_feedback_foreignkey_constraint.py

@@ -12,8 +12,8 @@ import sqlalchemy as sa
 # revision identifiers, used by Alembic.
 revision = "23957775e5f5"
 down_revision = "bc9771dccadf"
-branch_labels = None  # type: ignore
-depends_on = None  # type: ignore
+branch_labels = None
+depends_on = None
 
 
 def upgrade() -> None:

backend/alembic/versions/2a391f840e85_add_last_refreshed_at_mcp_server.py

@@ -0,0 +1,27 @@
+"""add last refreshed at mcp server
+
+Revision ID: 2a391f840e85
+Revises: 4cebcbc9b2ae
+Create Date: 2025-12-06 15:19:59.766066
+
+"""
+
+from alembic import op
+import sqlalchemy as sa
+
+# revision identifiers, used by Alembi.
+revision = "2a391f840e85"
+down_revision = "4cebcbc9b2ae"
+branch_labels = None
+depends_on = None
+
+
+def upgrade() -> None:
+    op.add_column(
+        "mcp_server",
+        sa.Column("last_refreshed_at", sa.DateTime(timezone=True), nullable=True),
+    )
+
+
+def downgrade() -> None:
+    op.drop_column("mcp_server", "last_refreshed_at")

backend/alembic/versions/2acdef638fc2_add_switchover_type_field.py

@@ -0,0 +1,72 @@
+"""add switchover_type field and remove background_reindex_enabled
+
+Revision ID: 2acdef638fc2
+Revises: a4f23d6b71c8
+Create Date: 2025-01-XX XX:XX:XX.XXXXXX
+
+"""
+
+from alembic import op
+import sqlalchemy as sa
+
+from onyx.db.enums import SwitchoverType
+
+
+# revision identifiers, used by Alembic.
+revision = "2acdef638fc2"
+down_revision = "a4f23d6b71c8"
+branch_labels = None
+depends_on = None
+
+
+def upgrade() -> None:
+    # Add switchover_type column with default value of REINDEX
+    op.add_column(
+        "search_settings",
+        sa.Column(
+            "switchover_type",
+            sa.Enum(SwitchoverType, native_enum=False),
+            nullable=False,
+            server_default=SwitchoverType.REINDEX.value,
+        ),
+    )
+
+    # Migrate existing data: set switchover_type based on background_reindex_enabled
+    # REINDEX where background_reindex_enabled=True, INSTANT where False
+    op.execute(
+        """
+        UPDATE search_settings
+        SET switchover_type = CASE
+            WHEN background_reindex_enabled = true THEN 'REINDEX'
+            ELSE 'INSTANT'
+        END
+        """
+    )
+
+    # Remove the background_reindex_enabled column (replaced by switchover_type)
+    op.drop_column("search_settings", "background_reindex_enabled")
+
+
+def downgrade() -> None:
+    # Re-add the background_reindex_enabled column with default value of True
+    op.add_column(
+        "search_settings",
+        sa.Column(
+            "background_reindex_enabled",
+            sa.Boolean(),
+            nullable=False,
+            server_default="true",
+        ),
+    )
+    # Set background_reindex_enabled based on switchover_type
+    op.execute(
+        """
+        UPDATE search_settings
+        SET background_reindex_enabled = CASE
+            WHEN switchover_type = 'INSTANT' THEN false
+            ELSE true
+        END
+        """
+    )
+    # Remove the switchover_type column
+    op.drop_column("search_settings", "switchover_type")

backend/alembic/versions/2b75d0a8ffcb_user_file_schema_cleanup.py

@@ -0,0 +1,228 @@
+"""Migration 6: User file schema cleanup
+
+Revision ID: 2b75d0a8ffcb
+Revises: 3a78dba1080a
+Create Date: 2025-09-22 10:09:26.375377
+
+This migration removes legacy columns and tables after data migration is complete.
+It should only be run after verifying all data has been successfully migrated.
+"""
+
+from alembic import op
+import sqlalchemy as sa
+from sqlalchemy import text
+import logging
+import fastapi_users_db_sqlalchemy
+
+logger = logging.getLogger("alembic.runtime.migration")
+
+# revision identifiers, used by Alembic.
+revision = "2b75d0a8ffcb"
+down_revision = "3a78dba1080a"
+branch_labels = None
+depends_on = None
+
+
+def upgrade() -> None:
+    """Remove legacy columns and tables."""
+
+    bind = op.get_bind()
+    inspector = sa.inspect(bind)
+
+    logger.info("Starting schema cleanup...")
+
+    # === Step 1: Verify data migration is complete ===
+    logger.info("Verifying data migration completion...")
+
+    # Check if any chat sessions still have folder_id references
+    chat_session_columns = [
+        col["name"] for col in inspector.get_columns("chat_session")
+    ]
+    if "folder_id" in chat_session_columns:
+        orphaned_count = bind.execute(
+            text(
+                """
+            SELECT COUNT(*) FROM chat_session
+            WHERE folder_id IS NOT NULL AND project_id IS NULL
+        """
+            )
+        ).scalar_one()
+
+        if orphaned_count > 0:
+            logger.warning(
+                f"WARNING: {orphaned_count} chat_session records still have "
+                f"folder_id without project_id. Proceeding anyway."
+            )
+
+    # === Step 2: Drop chat_session.folder_id ===
+    if "folder_id" in chat_session_columns:
+        logger.info("Dropping chat_session.folder_id...")
+
+        # Drop foreign key constraint first
+        op.execute(
+            "ALTER TABLE chat_session DROP CONSTRAINT IF EXISTS chat_session_chat_folder_fk"
+        )
+        op.execute(
+            "ALTER TABLE chat_session DROP CONSTRAINT IF EXISTS chat_session_folder_fk"
+        )
+
+        # Drop the column
+        op.drop_column("chat_session", "folder_id")
+        logger.info("Dropped chat_session.folder_id")
+
+    # === Step 3: Drop persona__user_folder table ===
+    if "persona__user_folder" in inspector.get_table_names():
+        logger.info("Dropping persona__user_folder table...")
+
+        # Check for any remaining data
+        remaining = bind.execute(
+            text("SELECT COUNT(*) FROM persona__user_folder")
+        ).scalar_one()
+
+        if remaining > 0:
+            logger.warning(
+                f"WARNING: Dropping persona__user_folder with {remaining} records"
+            )
+
+        op.drop_table("persona__user_folder")
+        logger.info("Dropped persona__user_folder table")
+
+    # === Step 4: Drop chat_folder table ===
+    if "chat_folder" in inspector.get_table_names():
+        logger.info("Dropping chat_folder table...")
+
+        # Check for any remaining data
+        remaining = bind.execute(text("SELECT COUNT(*) FROM chat_folder")).scalar_one()
+
+        if remaining > 0:
+            logger.warning(f"WARNING: Dropping chat_folder with {remaining} records")
+
+        op.drop_table("chat_folder")
+        logger.info("Dropped chat_folder table")
+
+    # === Step 5: Drop user_file legacy columns ===
+    user_file_columns = [col["name"] for col in inspector.get_columns("user_file")]
+
+    # Drop folder_id
+    if "folder_id" in user_file_columns:
+        logger.info("Dropping user_file.folder_id...")
+        op.drop_column("user_file", "folder_id")
+        logger.info("Dropped user_file.folder_id")
+
+    # Drop cc_pair_id (already handled in migration 5, but be sure)
+    if "cc_pair_id" in user_file_columns:
+        logger.info("Dropping user_file.cc_pair_id...")
+
+        # Drop any remaining foreign key constraints
+        bind.execute(
+            text(
+                """
+            DO $$
+            DECLARE r RECORD;
+            BEGIN
+              FOR r IN (
+                SELECT conname
+                FROM pg_constraint c
+                JOIN pg_class t ON c.conrelid = t.oid
+                WHERE c.contype = 'f'
+                  AND t.relname = 'user_file'
+                  AND EXISTS (
+                    SELECT 1 FROM pg_attribute a
+                    WHERE a.attrelid = t.oid
+                    AND a.attname = 'cc_pair_id'
+                  )
+              ) LOOP
+                EXECUTE format('ALTER TABLE user_file DROP CONSTRAINT IF EXISTS %I', r.conname);
+              END LOOP;
+            END$$;
+        """
+            )
+        )
+
+        op.drop_column("user_file", "cc_pair_id")
+        logger.info("Dropped user_file.cc_pair_id")
+
+    # === Step 6: Clean up any remaining constraints ===
+    logger.info("Cleaning up remaining constraints...")
+
+    # Drop any unique constraints on removed columns
+    op.execute(
+        "ALTER TABLE user_file DROP CONSTRAINT IF EXISTS user_file_cc_pair_id_key"
+    )
+
+    logger.info("Migration 6 (schema cleanup) completed successfully")
+    logger.info("Legacy schema has been fully removed")
+
+
+def downgrade() -> None:
+    """Recreate dropped columns and tables (structure only, no data)."""
+
+    bind = op.get_bind()
+    inspector = sa.inspect(bind)
+
+    logger.warning("Downgrading schema cleanup - recreating structure only, no data!")
+
+    # Recreate user_file columns
+    if "user_file" in inspector.get_table_names():
+        columns = [col["name"] for col in inspector.get_columns("user_file")]
+
+        if "cc_pair_id" not in columns:
+            op.add_column(
+                "user_file", sa.Column("cc_pair_id", sa.Integer(), nullable=True)
+            )
+
+        if "folder_id" not in columns:
+            op.add_column(
+                "user_file", sa.Column("folder_id", sa.Integer(), nullable=True)
+            )
+
+    # Recreate persona__user_folder table
+    if "persona__user_folder" not in inspector.get_table_names():
+        op.create_table(
+            "persona__user_folder",
+            sa.Column("persona_id", sa.Integer(), nullable=False),
+            sa.Column("user_folder_id", sa.Integer(), nullable=False),
+            sa.PrimaryKeyConstraint("persona_id", "user_folder_id"),
+            sa.ForeignKeyConstraint(["persona_id"], ["persona.id"]),
+            sa.ForeignKeyConstraint(["user_folder_id"], ["user_project.id"]),
+        )
+
+    # Recreate chat_folder table and related structures
+    if "chat_folder" not in inspector.get_table_names():
+        op.create_table(
+            "chat_folder",
+            sa.Column("id", sa.Integer(), nullable=False),
+            sa.Column(
+                "user_id",
+                fastapi_users_db_sqlalchemy.generics.GUID(),
+                nullable=True,
+            ),
+            sa.Column("name", sa.String(), nullable=True),
+            sa.Column("display_priority", sa.Integer(), nullable=False),
+            sa.ForeignKeyConstraint(
+                ["user_id"],
+                ["user.id"],
+                name="chat_folder_user_id_fkey",
+            ),
+            sa.PrimaryKeyConstraint("id"),
+        )
+
+    # Add folder_id back to chat_session
+    if "chat_session" in inspector.get_table_names():
+        columns = [col["name"] for col in inspector.get_columns("chat_session")]
+        if "folder_id" not in columns:
+            op.add_column(
+                "chat_session", sa.Column("folder_id", sa.Integer(), nullable=True)
+            )
+
+            # Add foreign key if chat_folder exists
+            if "chat_folder" in inspector.get_table_names():
+                op.create_foreign_key(
+                    "chat_session_chat_folder_fk",
+                    "chat_session",
+                    "chat_folder",
+                    ["folder_id"],
+                    ["id"],
+                )
+
+    logger.info("Downgrade completed - structure recreated but data is lost")

backend/alembic/versions/2b90f3af54b8_usage_limits.py

@@ -0,0 +1,46 @@
+"""usage_limits
+
+Revision ID: 2b90f3af54b8
+Revises: 9a0296d7421e
+Create Date: 2026-01-03 16:55:30.449692
+
+"""
+
+from alembic import op
+import sqlalchemy as sa
+
+
+# revision identifiers, used by Alembic.
+revision = "2b90f3af54b8"
+down_revision = "9a0296d7421e"
+branch_labels = None
+depends_on = None
+
+
+def upgrade() -> None:
+    op.create_table(
+        "tenant_usage",
+        sa.Column("id", sa.Integer(), nullable=False),
+        sa.Column(
+            "window_start", sa.DateTime(timezone=True), nullable=False, index=True
+        ),
+        sa.Column("llm_cost_cents", sa.Float(), nullable=False, server_default="0.0"),
+        sa.Column("chunks_indexed", sa.Integer(), nullable=False, server_default="0"),
+        sa.Column("api_calls", sa.Integer(), nullable=False, server_default="0"),
+        sa.Column(
+            "non_streaming_api_calls", sa.Integer(), nullable=False, server_default="0"
+        ),
+        sa.Column(
+            "updated_at",
+            sa.DateTime(timezone=True),
+            server_default=sa.func.now(),
+            nullable=True,
+        ),
+        sa.PrimaryKeyConstraint("id"),
+        sa.UniqueConstraint("window_start", name="uq_tenant_usage_window"),
+    )
+
+
+def downgrade() -> None:
+    op.drop_index("ix_tenant_usage_window_start", table_name="tenant_usage")
+    op.drop_table("tenant_usage")

backend/alembic/versions/3a78dba1080a_user_file_legacy_data_cleanup.py

@@ -0,0 +1,298 @@
+"""Migration 5: User file legacy data cleanup
+
+Revision ID: 3a78dba1080a
+Revises: 7cc3fcc116c1
+Create Date: 2025-09-22 10:04:27.986294
+
+This migration removes legacy user-file documents and connector_credential_pairs.
+It performs bulk deletions of obsolete data after the UUID migration.
+"""
+
+from alembic import op
+import sqlalchemy as sa
+from sqlalchemy.dialects import postgresql as psql
+from sqlalchemy import text
+import logging
+from typing import List
+import uuid
+
+logger = logging.getLogger("alembic.runtime.migration")
+
+# revision identifiers, used by Alembic.
+revision = "3a78dba1080a"
+down_revision = "7cc3fcc116c1"
+branch_labels = None
+depends_on = None
+
+
+def batch_delete(
+    bind: sa.engine.Connection,
+    table_name: str,
+    id_column: str,
+    ids: List[str | int | uuid.UUID],
+    batch_size: int = 1000,
+    id_type: str = "int",
+) -> int:
+    """Delete records in batches to avoid memory issues and timeouts."""
+    total_count = len(ids)
+    if total_count == 0:
+        return 0
+
+    logger.info(
+        f"Starting batch deletion of {total_count} records from {table_name}..."
+    )
+
+    # Determine appropriate ARRAY type
+    if id_type == "uuid":
+        array_type = psql.ARRAY(psql.UUID(as_uuid=True))
+    elif id_type == "int":
+        array_type = psql.ARRAY(sa.Integer())
+    else:
+        array_type = psql.ARRAY(sa.String())
+
+    total_deleted = 0
+    failed_batches = []
+
+    for i in range(0, total_count, batch_size):
+        batch_ids = ids[i : i + batch_size]
+        try:
+            stmt = text(
+                f"DELETE FROM {table_name} WHERE {id_column} = ANY(:ids)"
+            ).bindparams(sa.bindparam("ids", value=batch_ids, type_=array_type))
+            result = bind.execute(stmt)
+            total_deleted += result.rowcount
+
+            # Log progress every 10 batches or at completion
+            batch_num = (i // batch_size) + 1
+            if batch_num % 10 == 0 or i + batch_size >= total_count:
+                logger.info(
+                    f"  Deleted {min(i + batch_size, total_count)}/{total_count} records "
+                    f"({total_deleted} actual) from {table_name}"
+                )
+        except Exception as e:
+            logger.error(f"Failed to delete batch {(i // batch_size) + 1}: {e}")
+            failed_batches.append((i, min(i + batch_size, total_count)))
+
+    if failed_batches:
+        logger.warning(
+            f"Failed to delete {len(failed_batches)} batches from {table_name}. "
+            f"Total deleted: {total_deleted}/{total_count}"
+        )
+        # Fail the migration to avoid silently succeeding on partial cleanup
+        raise RuntimeError(
+            f"Batch deletion failed for {table_name}: "
+            f"{len(failed_batches)} failed batches out of "
+            f"{(total_count + batch_size - 1) // batch_size}."
+        )
+
+    return total_deleted
+
+
+def upgrade() -> None:
+    """Remove legacy user-file documents and connector_credential_pairs."""
+
+    bind = op.get_bind()
+    inspector = sa.inspect(bind)
+
+    logger.info("Starting legacy data cleanup...")
+
+    # === Step 1: Identify and delete user-file documents ===
+    logger.info("Identifying user-file documents to delete...")
+
+    # Get document IDs to delete
+    doc_rows = bind.execute(
+        text(
+            """
+        SELECT DISTINCT dcc.id AS document_id
+        FROM document_by_connector_credential_pair dcc
+        JOIN connector_credential_pair u
+          ON u.connector_id = dcc.connector_id
+         AND u.credential_id = dcc.credential_id
+        WHERE u.is_user_file IS TRUE
+    """
+        )
+    ).fetchall()
+
+    doc_ids = [r[0] for r in doc_rows]
+
+    if doc_ids:
+        logger.info(f"Found {len(doc_ids)} user-file documents to delete")
+
+        # Delete dependent rows first
+        tables_to_clean = [
+            ("document_retrieval_feedback", "document_id"),
+            ("document__tag", "document_id"),
+            ("chunk_stats", "document_id"),
+        ]
+
+        for table_name, column_name in tables_to_clean:
+            if table_name in inspector.get_table_names():
+                # document_id is a string in these tables
+                deleted = batch_delete(
+                    bind, table_name, column_name, doc_ids, id_type="str"
+                )
+                logger.info(f"Deleted {deleted} records from {table_name}")
+
+        # Delete document_by_connector_credential_pair entries
+        deleted = batch_delete(
+            bind, "document_by_connector_credential_pair", "id", doc_ids, id_type="str"
+        )
+        logger.info(f"Deleted {deleted} document_by_connector_credential_pair records")
+
+        # Delete documents themselves
+        deleted = batch_delete(bind, "document", "id", doc_ids, id_type="str")
+        logger.info(f"Deleted {deleted} document records")
+    else:
+        logger.info("No user-file documents found to delete")
+
+    # === Step 2: Clean up user-file connector_credential_pairs ===
+    logger.info("Cleaning up user-file connector_credential_pairs...")
+
+    # Get cc_pair IDs
+    cc_pair_rows = bind.execute(
+        text(
+            """
+        SELECT id AS cc_pair_id
+        FROM connector_credential_pair
+        WHERE is_user_file IS TRUE
+    """
+        )
+    ).fetchall()
+
+    cc_pair_ids = [r[0] for r in cc_pair_rows]
+
+    if cc_pair_ids:
+        logger.info(
+            f"Found {len(cc_pair_ids)} user-file connector_credential_pairs to clean up"
+        )
+
+        # Delete related records
+        # Clean child tables first to satisfy foreign key constraints,
+        # then the parent tables
+        tables_to_clean = [
+            ("index_attempt_errors", "connector_credential_pair_id"),
+            ("index_attempt", "connector_credential_pair_id"),
+            ("background_error", "cc_pair_id"),
+            ("document_set__connector_credential_pair", "connector_credential_pair_id"),
+            ("user_group__connector_credential_pair", "cc_pair_id"),
+        ]
+
+        for table_name, column_name in tables_to_clean:
+            if table_name in inspector.get_table_names():
+                deleted = batch_delete(
+                    bind, table_name, column_name, cc_pair_ids, id_type="int"
+                )
+                logger.info(f"Deleted {deleted} records from {table_name}")
+
+    # === Step 3: Identify connectors and credentials to delete ===
+    logger.info("Identifying orphaned connectors and credentials...")
+
+    # Get connectors used only by user-file cc_pairs
+    connector_rows = bind.execute(
+        text(
+            """
+        SELECT DISTINCT ccp.connector_id
+        FROM connector_credential_pair ccp
+        WHERE ccp.is_user_file IS TRUE
+          AND ccp.connector_id != 0  -- Exclude system default
+          AND NOT EXISTS (
+            SELECT 1
+            FROM connector_credential_pair c2
+            WHERE c2.connector_id = ccp.connector_id
+              AND c2.is_user_file IS NOT TRUE
+          )
+    """
+        )
+    ).fetchall()
+
+    userfile_only_connector_ids = [r[0] for r in connector_rows]
+
+    # Get credentials used only by user-file cc_pairs
+    credential_rows = bind.execute(
+        text(
+            """
+        SELECT DISTINCT ccp.credential_id
+        FROM connector_credential_pair ccp
+        WHERE ccp.is_user_file IS TRUE
+          AND ccp.credential_id != 0  -- Exclude public/default
+          AND NOT EXISTS (
+            SELECT 1
+            FROM connector_credential_pair c2
+            WHERE c2.credential_id = ccp.credential_id
+              AND c2.is_user_file IS NOT TRUE
+          )
+    """
+        )
+    ).fetchall()
+
+    userfile_only_credential_ids = [r[0] for r in credential_rows]
+
+    # === Step 4: Delete the cc_pairs themselves ===
+    if cc_pair_ids:
+        # Remove FK dependency from user_file first
+        bind.execute(
+            text(
+                """
+            DO $$
+            DECLARE r RECORD;
+            BEGIN
+              FOR r IN (
+                SELECT conname
+                FROM pg_constraint c
+                JOIN pg_class t ON c.conrelid = t.oid
+                JOIN pg_class ft ON c.confrelid = ft.oid
+                WHERE c.contype = 'f'
+                  AND t.relname = 'user_file'
+                  AND ft.relname = 'connector_credential_pair'
+              ) LOOP
+                EXECUTE format('ALTER TABLE user_file DROP CONSTRAINT IF EXISTS %I', r.conname);
+              END LOOP;
+            END$$;
+        """
+            )
+        )
+
+        # Delete cc_pairs
+        deleted = batch_delete(
+            bind, "connector_credential_pair", "id", cc_pair_ids, id_type="int"
+        )
+        logger.info(f"Deleted {deleted} connector_credential_pair records")
+
+    # === Step 5: Delete orphaned connectors ===
+    if userfile_only_connector_ids:
+        deleted = batch_delete(
+            bind, "connector", "id", userfile_only_connector_ids, id_type="int"
+        )
+        logger.info(f"Deleted {deleted} orphaned connector records")
+
+    # === Step 6: Delete orphaned credentials ===
+    if userfile_only_credential_ids:
+        # Clean up credential__user_group mappings first
+        deleted = batch_delete(
+            bind,
+            "credential__user_group",
+            "credential_id",
+            userfile_only_credential_ids,
+            id_type="int",
+        )
+        logger.info(f"Deleted {deleted} credential__user_group records")
+
+        # Delete credentials
+        deleted = batch_delete(
+            bind, "credential", "id", userfile_only_credential_ids, id_type="int"
+        )
+        logger.info(f"Deleted {deleted} orphaned credential records")
+
+    logger.info("Migration 5 (legacy data cleanup) completed successfully")
+
+
+def downgrade() -> None:
+    """Cannot restore deleted data - requires backup restoration."""
+
+    logger.error("CRITICAL: Downgrading data cleanup cannot restore deleted data!")
+    logger.error("Data restoration requires backup files or database backup.")
+
+    # raise NotImplementedError(
+    #     "Downgrade of legacy data cleanup is not supported. "
+    #     "Deleted data must be restored from backups."
+    # )

backend/alembic/versions/3c9a65f1207f_seed_exa_provider_from_env.py

@@ -0,0 +1,89 @@
+"""seed_exa_provider_from_env
+
+Revision ID: 3c9a65f1207f
+Revises: 1f2a3b4c5d6e
+Create Date: 2025-11-20 19:18:00.000000
+
+"""
+
+from __future__ import annotations
+
+import os
+
+from alembic import op
+import sqlalchemy as sa
+from sqlalchemy.dialects import postgresql
+from dotenv import load_dotenv, find_dotenv
+
+from onyx.utils.encryption import encrypt_string_to_bytes
+
+revision = "3c9a65f1207f"
+down_revision = "1f2a3b4c5d6e"
+branch_labels = None
+depends_on = None
+
+
+EXA_PROVIDER_NAME = "Exa"
+
+
+def _get_internet_search_table(metadata: sa.MetaData) -> sa.Table:
+    return sa.Table(
+        "internet_search_provider",
+        metadata,
+        sa.Column("id", sa.Integer, primary_key=True),
+        sa.Column("name", sa.String),
+        sa.Column("provider_type", sa.String),
+        sa.Column("api_key", sa.LargeBinary),
+        sa.Column("config", postgresql.JSONB),
+        sa.Column("is_active", sa.Boolean),
+        sa.Column(
+            "time_created",
+            sa.DateTime(timezone=True),
+            nullable=False,
+            server_default=sa.text("now()"),
+        ),
+        sa.Column(
+            "time_updated",
+            sa.DateTime(timezone=True),
+            nullable=False,
+            server_default=sa.text("now()"),
+        ),
+    )
+
+
+def upgrade() -> None:
+    load_dotenv(find_dotenv())
+
+    exa_api_key = os.environ.get("EXA_API_KEY")
+    if not exa_api_key:
+        return
+
+    bind = op.get_bind()
+    metadata = sa.MetaData()
+    table = _get_internet_search_table(metadata)
+
+    existing = bind.execute(
+        sa.select(table.c.id).where(table.c.name == EXA_PROVIDER_NAME)
+    ).first()
+    if existing:
+        return
+
+    encrypted_key = encrypt_string_to_bytes(exa_api_key)
+
+    has_active_provider = bind.execute(
+        sa.select(table.c.id).where(table.c.is_active.is_(True))
+    ).first()
+
+    bind.execute(
+        table.insert().values(
+            name=EXA_PROVIDER_NAME,
+            provider_type="exa",
+            api_key=encrypted_key,
+            config=None,
+            is_active=not bool(has_active_provider),
+        )
+    )
+
+
+def downgrade() -> None:
+    return

backend/alembic/versions/3d1cca026fe8_add_oauth_config_and_user_tokens.py

@@ -0,0 +1,121 @@
+"""add_oauth_config_and_user_tokens
+
+Revision ID: 3d1cca026fe8
+Revises: c8a93a2af083
+Create Date: 2025-10-21 13:27:34.274721
+
+"""
+
+from alembic import op
+import fastapi_users_db_sqlalchemy
+import sqlalchemy as sa
+from sqlalchemy.dialects import postgresql
+
+# revision identifiers, used by Alembic.
+revision = "3d1cca026fe8"
+down_revision = "c8a93a2af083"
+branch_labels = None
+depends_on = None
+
+
+def upgrade() -> None:
+    # Create oauth_config table
+    op.create_table(
+        "oauth_config",
+        sa.Column("id", sa.Integer(), nullable=False),
+        sa.Column("name", sa.String(), nullable=False),
+        sa.Column("authorization_url", sa.Text(), nullable=False),
+        sa.Column("token_url", sa.Text(), nullable=False),
+        sa.Column("client_id", sa.LargeBinary(), nullable=False),
+        sa.Column("client_secret", sa.LargeBinary(), nullable=False),
+        sa.Column("scopes", postgresql.JSONB(astext_type=sa.Text()), nullable=True),
+        sa.Column(
+            "additional_params",
+            postgresql.JSONB(astext_type=sa.Text()),
+            nullable=True,
+        ),
+        sa.Column(
+            "created_at",
+            sa.DateTime(timezone=True),
+            server_default=sa.text("now()"),
+            nullable=False,
+        ),
+        sa.Column(
+            "updated_at",
+            sa.DateTime(timezone=True),
+            server_default=sa.text("now()"),
+            nullable=False,
+        ),
+        sa.PrimaryKeyConstraint("id"),
+        sa.UniqueConstraint("name"),
+    )
+
+    # Create oauth_user_token table
+    op.create_table(
+        "oauth_user_token",
+        sa.Column("id", sa.Integer(), nullable=False),
+        sa.Column("oauth_config_id", sa.Integer(), nullable=False),
+        sa.Column(
+            "user_id",
+            fastapi_users_db_sqlalchemy.generics.GUID(),
+            nullable=False,
+        ),
+        sa.Column("token_data", sa.LargeBinary(), nullable=False),
+        sa.Column(
+            "created_at",
+            sa.DateTime(timezone=True),
+            server_default=sa.text("now()"),
+            nullable=False,
+        ),
+        sa.Column(
+            "updated_at",
+            sa.DateTime(timezone=True),
+            server_default=sa.text("now()"),
+            nullable=False,
+        ),
+        sa.ForeignKeyConstraint(
+            ["oauth_config_id"], ["oauth_config.id"], ondelete="CASCADE"
+        ),
+        sa.ForeignKeyConstraint(["user_id"], ["user.id"], ondelete="CASCADE"),
+        sa.PrimaryKeyConstraint("id"),
+        sa.UniqueConstraint("oauth_config_id", "user_id", name="uq_oauth_user_token"),
+    )
+
+    # Create index on user_id for efficient user-based token lookups
+    # Note: unique constraint on (oauth_config_id, user_id) already creates
+    # an index for config-based lookups
+    op.create_index(
+        "ix_oauth_user_token_user_id",
+        "oauth_user_token",
+        ["user_id"],
+    )
+
+    # Add oauth_config_id column to tool table
+    op.add_column("tool", sa.Column("oauth_config_id", sa.Integer(), nullable=True))
+
+    # Create foreign key from tool to oauth_config
+    op.create_foreign_key(
+        "tool_oauth_config_fk",
+        "tool",
+        "oauth_config",
+        ["oauth_config_id"],
+        ["id"],
+        ondelete="SET NULL",
+    )
+
+
+def downgrade() -> None:
+    # Drop foreign key from tool to oauth_config
+    op.drop_constraint("tool_oauth_config_fk", "tool", type_="foreignkey")
+
+    # Drop oauth_config_id column from tool table
+    op.drop_column("tool", "oauth_config_id")
+
+    # Drop index on user_id
+    op.drop_index("ix_oauth_user_token_user_id", table_name="oauth_user_token")
+
+    # Drop oauth_user_token table (will cascade delete tokens)
+    op.drop_table("oauth_user_token")
+
+    # Drop oauth_config table
+    op.drop_table("oauth_config")

backend/alembic/versions/40926a4dab77_reset_userfile_document_id_migrated_.py

@@ -0,0 +1,28 @@
+"""reset userfile document_id_migrated field
+
+Revision ID: 40926a4dab77
+Revises: 64bd5677aeb6
+Create Date: 2025-10-06 16:10:32.898668
+
+"""
+
+from alembic import op
+
+# revision identifiers, used by Alembic.
+revision = "40926a4dab77"
+down_revision = "64bd5677aeb6"
+branch_labels = None
+depends_on = None
+
+
+def upgrade() -> None:
+    # Set all existing records to not migrated
+    op.execute(
+        "UPDATE user_file SET document_id_migrated = FALSE "
+        "WHERE document_id_migrated IS DISTINCT FROM FALSE;"
+    )
+
+
+def downgrade() -> None:
+    # No-op
+    pass

backend/alembic/versions/47a07e1a38f1_fix_invalid_model_configurations_state.py

@@ -11,7 +11,7 @@ from pydantic import BaseModel, ConfigDict
 import sqlalchemy as sa
 from sqlalchemy.dialects import postgresql
 
-from onyx.llm.llm_provider_options import (
+from onyx.llm.well_known_providers.llm_provider_options import (
     fetch_model_names_for_provider_as_set,
     fetch_visible_model_names_for_provider_as_set,
 )

backend/alembic/versions/4cebcbc9b2ae_add_tab_index_to_tool_call.py

@@ -0,0 +1,27 @@
+"""add tab_index to tool_call
+
+Revision ID: 4cebcbc9b2ae
+Revises: a1b2c3d4e5f6
+Create Date: 2025-12-16
+
+"""
+
+from alembic import op
+import sqlalchemy as sa
+
+# revision identifiers, used by Alembic.
+revision = "4cebcbc9b2ae"
+down_revision = "a1b2c3d4e5f6"
+branch_labels: None = None
+depends_on: None = None
+
+
+def upgrade() -> None:
+    op.add_column(
+        "tool_call",
+        sa.Column("tab_index", sa.Integer(), nullable=False, server_default="0"),
+    )
+
+
+def downgrade() -> None:
+    op.drop_column("tool_call", "tab_index")

backend/alembic/versions/4ea2c93919c1_add_type_to_credentials.py

@@ -62,6 +62,11 @@ def upgrade() -> None:
     )
     """
     )
+
+    # Drop the temporary table to avoid conflicts if migration runs again
+    # (e.g., during upgrade -> downgrade -> upgrade cycles in tests)
+    op.execute("DROP TABLE IF EXISTS temp_connector_credential")
+
     # If no exception was raised, alter the column
     op.alter_column("credential", "source", nullable=True)  # TODO modify
     # # ### end Alembic commands ###

backend/alembic/versions/4f8a2b3c1d9e_add_open_url_tool.py

@@ -0,0 +1,104 @@
+"""add_open_url_tool
+
+Revision ID: 4f8a2b3c1d9e
+Revises: a852cbe15577
+Create Date: 2025-11-24 12:00:00.000000
+
+"""
+
+from alembic import op
+import sqlalchemy as sa
+
+
+# revision identifiers, used by Alembic.
+revision = "4f8a2b3c1d9e"
+down_revision = "a852cbe15577"
+branch_labels = None
+depends_on = None
+
+
+OPEN_URL_TOOL = {
+    "name": "OpenURLTool",
+    "display_name": "Open URL",
+    "description": (
+        "The Open URL Action allows the agent to fetch and read contents of web pages."
+    ),
+    "in_code_tool_id": "OpenURLTool",
+    "enabled": True,
+}
+
+
+def upgrade() -> None:
+    conn = op.get_bind()
+
+    # Check if tool already exists
+    existing = conn.execute(
+        sa.text("SELECT id FROM tool WHERE in_code_tool_id = :in_code_tool_id"),
+        {"in_code_tool_id": OPEN_URL_TOOL["in_code_tool_id"]},
+    ).fetchone()
+
+    if existing:
+        tool_id = existing[0]
+        # Update existing tool
+        conn.execute(
+            sa.text(
+                """
+                UPDATE tool
+                SET name = :name,
+                    display_name = :display_name,
+                    description = :description
+                WHERE in_code_tool_id = :in_code_tool_id
+                """
+            ),
+            OPEN_URL_TOOL,
+        )
+    else:
+        # Insert new tool
+        conn.execute(
+            sa.text(
+                """
+                INSERT INTO tool (name, display_name, description, in_code_tool_id, enabled)
+                VALUES (:name, :display_name, :description, :in_code_tool_id, :enabled)
+                """
+            ),
+            OPEN_URL_TOOL,
+        )
+        # Get the newly inserted tool's id
+        result = conn.execute(
+            sa.text("SELECT id FROM tool WHERE in_code_tool_id = :in_code_tool_id"),
+            {"in_code_tool_id": OPEN_URL_TOOL["in_code_tool_id"]},
+        ).fetchone()
+        tool_id = result[0]  # type: ignore
+
+    # Associate the tool with all existing personas
+    # Get all persona IDs
+    persona_ids = conn.execute(sa.text("SELECT id FROM persona")).fetchall()
+
+    for (persona_id,) in persona_ids:
+        # Check if association already exists
+        exists = conn.execute(
+            sa.text(
+                """
+                SELECT 1 FROM persona__tool
+                WHERE persona_id = :persona_id AND tool_id = :tool_id
+                """
+            ),
+            {"persona_id": persona_id, "tool_id": tool_id},
+        ).fetchone()
+
+        if not exists:
+            conn.execute(
+                sa.text(
+                    """
+                    INSERT INTO persona__tool (persona_id, tool_id)
+                    VALUES (:persona_id, :tool_id)
+                    """
+                ),
+                {"persona_id": persona_id, "tool_id": tool_id},
+            )
+
+
+def downgrade() -> None:
+    # We don't remove the tool on downgrade since it's fine to have it around.
+    # If we upgrade again, it will be a no-op.
+    pass

backend/alembic/versions/5c3dca366b35_backend_driven_notification_details.py

@@ -0,0 +1,35 @@
+"""backend driven notification details
+
+Revision ID: 5c3dca366b35
+Revises: 9087b548dd69
+Create Date: 2026-01-06 16:03:11.413724
+
+"""
+
+from alembic import op
+import sqlalchemy as sa
+
+
+# revision identifiers, used by Alembic.
+revision = "5c3dca366b35"
+down_revision = "9087b548dd69"
+branch_labels = None
+depends_on = None
+
+
+def upgrade() -> None:
+    op.add_column(
+        "notification",
+        sa.Column(
+            "title", sa.String(), nullable=False, server_default="New Notification"
+        ),
+    )
+    op.add_column(
+        "notification",
+        sa.Column("description", sa.String(), nullable=True, server_default=""),
+    )
+
+
+def downgrade() -> None:
+    op.drop_column("notification", "title")
+    op.drop_column("notification", "description")

backend/alembic/versions/5e1c073d48a3_add_personal_access_token_table.py

@@ -0,0 +1,88 @@
+"""add_personal_access_token_table
+
+Revision ID: 5e1c073d48a3
+Revises: 09995b8811eb
+Create Date: 2025-10-30 17:30:24.308521
+
+"""
+
+from alembic import op
+import sqlalchemy as sa
+from sqlalchemy.dialects import postgresql
+
+
+# revision identifiers, used by Alembic.
+revision = "5e1c073d48a3"
+down_revision = "09995b8811eb"
+branch_labels = None
+depends_on = None
+
+
+def upgrade() -> None:
+    # Create personal_access_token table
+    op.create_table(
+        "personal_access_token",
+        sa.Column("id", sa.Integer(), nullable=False),
+        sa.Column("name", sa.String(), nullable=False),
+        sa.Column("hashed_token", sa.String(length=64), nullable=False),
+        sa.Column("token_display", sa.String(), nullable=False),
+        sa.Column(
+            "user_id",
+            postgresql.UUID(as_uuid=True),
+            nullable=False,
+        ),
+        sa.Column(
+            "expires_at",
+            sa.DateTime(timezone=True),
+            nullable=True,
+        ),
+        sa.Column(
+            "created_at",
+            sa.DateTime(timezone=True),
+            server_default=sa.text("now()"),
+            nullable=False,
+        ),
+        sa.Column(
+            "last_used_at",
+            sa.DateTime(timezone=True),
+            nullable=True,
+        ),
+        sa.Column(
+            "is_revoked",
+            sa.Boolean(),
+            server_default=sa.text("false"),
+            nullable=False,
+        ),
+        sa.ForeignKeyConstraint(
+            ["user_id"],
+            ["user.id"],
+            ondelete="CASCADE",
+        ),
+        sa.PrimaryKeyConstraint("id"),
+        sa.UniqueConstraint("hashed_token"),
+    )
+
+    # Create indexes
+    op.create_index(
+        "ix_personal_access_token_expires_at",
+        "personal_access_token",
+        ["expires_at"],
+        unique=False,
+    )
+    op.create_index(
+        "ix_pat_user_created",
+        "personal_access_token",
+        ["user_id", sa.text("created_at DESC")],
+        unique=False,
+    )
+
+
+def downgrade() -> None:
+    # Drop indexes first
+    op.drop_index("ix_pat_user_created", table_name="personal_access_token")
+    op.drop_index(
+        "ix_personal_access_token_expires_at", table_name="personal_access_token"
+    )
+
+    # Drop table
+    op.drop_table("personal_access_token")

backend/alembic/versions/5e6f7a8b9c0d_update_default_persona_prompt.py

@@ -0,0 +1,55 @@
+"""update_default_persona_prompt
+
+Revision ID: 5e6f7a8b9c0d
+Revises: 4f8a2b3c1d9e
+Create Date: 2025-11-30 12:00:00.000000
+
+"""
+
+from alembic import op
+import sqlalchemy as sa
+
+
+# revision identifiers, used by Alembic.
+revision = "5e6f7a8b9c0d"
+down_revision = "4f8a2b3c1d9e"
+branch_labels = None
+depends_on = None
+
+
+DEFAULT_PERSONA_ID = 0
+
+# ruff: noqa: E501, W605 start
+DEFAULT_SYSTEM_PROMPT = """
+You are a highly capable, thoughtful, and precise assistant. Your goal is to deeply understand the user's intent, ask clarifying questions when needed, think step-by-step through complex problems, provide clear and accurate answers, and proactively anticipate helpful follow-up information. Always prioritize being truthful, nuanced, insightful, and efficient.
+
+The current date is [[CURRENT_DATETIME]].{citation_reminder_or_empty}
+
+# Response Style
+You use different text styles, bolding, emojis (sparingly), block quotes, and other formatting to make your responses more readable and engaging.
+You use proper Markdown and LaTeX to format your responses for math, scientific, and chemical formulas, symbols, etc.: '$$\\n[expression]\\n$$' for standalone cases and '\\( [expression] \\)' when inline.
+For code you prefer to use Markdown and specify the language.
+You can use horizontal rules (---) to separate sections of your responses.
+You can use Markdown tables to format your responses for data, lists, and other structured information.
+""".lstrip()
+# ruff: noqa: E501, W605 end
+
+
+def upgrade() -> None:
+    conn = op.get_bind()
+    conn.execute(
+        sa.text(
+            """
+            UPDATE persona
+            SET system_prompt = :system_prompt
+            WHERE id = :persona_id
+            """
+        ),
+        {"system_prompt": DEFAULT_SYSTEM_PROMPT, "persona_id": DEFAULT_PERSONA_ID},
+    )
+
+
+def downgrade() -> None:
+    # We don't revert the system prompt on downgrade since we don't know
+    # what the previous value was. The new prompt is a reasonable default.
+    pass

backend/alembic/versions/6436661d5b65_add_created_at_in_project_userfile.py

@@ -0,0 +1,44 @@
+"""add_created_at_in_project_userfile
+
+Revision ID: 6436661d5b65
+Revises: c7e9f4a3b2d1
+Create Date: 2025-11-24 11:50:24.536052
+
+"""
+
+from alembic import op
+import sqlalchemy as sa
+
+# revision identifiers, used by Alembic.
+revision = "6436661d5b65"
+down_revision = "c7e9f4a3b2d1"
+branch_labels = None
+depends_on = None
+
+
+def upgrade() -> None:
+    # Add created_at column to project__user_file table
+    op.add_column(
+        "project__user_file",
+        sa.Column(
+            "created_at",
+            sa.DateTime(timezone=True),
+            server_default=sa.text("now()"),
+            nullable=False,
+        ),
+    )
+    # Add composite index on (project_id, created_at DESC)
+    op.create_index(
+        "ix_project__user_file_project_id_created_at",
+        "project__user_file",
+        ["project_id", sa.text("created_at DESC")],
+    )
+
+
+def downgrade() -> None:
+    # Remove composite index on (project_id, created_at)
+    op.drop_index(
+        "ix_project__user_file_project_id_created_at", table_name="project__user_file"
+    )
+    # Remove created_at column from project__user_file table
+    op.drop_column("project__user_file", "created_at")

backend/alembic/versions/64bd5677aeb6_add_image_input_support_to_model_config.py

@@ -0,0 +1,37 @@
+"""Add image input support to model config
+
+Revision ID: 64bd5677aeb6
+Revises: b30353be4eec
+Create Date: 2025-09-28 15:48:12.003612
+
+"""
+
+from alembic import op
+import sqlalchemy as sa
+
+
+# revision identifiers, used by Alembic.
+revision = "64bd5677aeb6"
+down_revision = "b30353be4eec"
+branch_labels = None
+depends_on = None
+
+
+def upgrade() -> None:
+    op.add_column(
+        "model_configuration",
+        sa.Column("supports_image_input", sa.Boolean(), nullable=True),
+    )
+
+    # Seems to be left over from when model visibility was introduced and a nullable field.
+    # Set any null is_visible values to False
+    connection = op.get_bind()
+    connection.execute(
+        sa.text(
+            "UPDATE model_configuration SET is_visible = false WHERE is_visible IS NULL"
+        )
+    )
+
+
+def downgrade() -> None:
+    op.drop_column("model_configuration", "supports_image_input")

backend/alembic/versions/699221885109_nullify_default_task_prompt.py

@@ -0,0 +1,75 @@
+"""nullify_default_task_prompt
+
+Revision ID: 699221885109
+Revises: 7e490836d179
+Create Date: 2025-12-30 10:00:00.000000
+
+"""
+
+from alembic import op
+import sqlalchemy as sa
+
+
+# revision identifiers, used by Alembic.
+revision = "699221885109"
+down_revision = "7e490836d179"
+branch_labels = None
+depends_on = None
+
+DEFAULT_PERSONA_ID = 0
+
+
+def upgrade() -> None:
+    # Make task_prompt column nullable
+    # Note: The model had nullable=True but the DB column was NOT NULL until this point
+    op.alter_column(
+        "persona",
+        "task_prompt",
+        nullable=True,
+    )
+
+    # Set task_prompt to NULL for the default persona
+    conn = op.get_bind()
+    conn.execute(
+        sa.text(
+            """
+            UPDATE persona
+            SET task_prompt = NULL
+            WHERE id = :persona_id
+            """
+        ),
+        {"persona_id": DEFAULT_PERSONA_ID},
+    )
+
+
+def downgrade() -> None:
+    # Restore task_prompt to empty string for the default persona
+    conn = op.get_bind()
+    conn.execute(
+        sa.text(
+            """
+            UPDATE persona
+            SET task_prompt = ''
+            WHERE id = :persona_id AND task_prompt IS NULL
+            """
+        ),
+        {"persona_id": DEFAULT_PERSONA_ID},
+    )
+
+    # Set any remaining NULL task_prompts to empty string before making non-nullable
+    conn.execute(
+        sa.text(
+            """
+            UPDATE persona
+            SET task_prompt = ''
+            WHERE task_prompt IS NULL
+            """
+        )
+    )
+
+    # Revert task_prompt column to not nullable
+    op.alter_column(
+        "persona",
+        "task_prompt",
+        nullable=False,
+    )

backend/alembic/versions/6f4f86aef280_add_queries_and_is_web_fetch_to_.py

@@ -0,0 +1,37 @@
+"""add queries and is web fetch to iteration answer
+
+Revision ID: 6f4f86aef280
+Revises: 03d710ccf29c
+Create Date: 2025-10-14 18:08:30.920123
+
+"""
+
+from alembic import op
+import sqlalchemy as sa
+from sqlalchemy.dialects import postgresql
+
+
+# revision identifiers, used by Alembic.
+revision = "6f4f86aef280"
+down_revision = "03d710ccf29c"
+branch_labels = None
+depends_on = None
+
+
+def upgrade() -> None:
+    # Add is_web_fetch column
+    op.add_column(
+        "research_agent_iteration_sub_step",
+        sa.Column("is_web_fetch", sa.Boolean(), nullable=True),
+    )
+
+    # Add queries column
+    op.add_column(
+        "research_agent_iteration_sub_step",
+        sa.Column("queries", postgresql.JSONB(), nullable=True),
+    )
+
+
+def downgrade() -> None:
+    op.drop_column("research_agent_iteration_sub_step", "queries")
+    op.drop_column("research_agent_iteration_sub_step", "is_web_fetch")

backend/alembic/versions/7206234e012a_add_image_generation_config_table.py

@@ -0,0 +1,54 @@
+"""add image generation config table
+
+Revision ID: 7206234e012a
+Revises: 699221885109
+Create Date: 2025-12-21 00:00:00.000000
+
+"""
+
+from alembic import op
+import sqlalchemy as sa
+
+
+# revision identifiers, used by Alembic.
+revision = "7206234e012a"
+down_revision = "699221885109"
+branch_labels = None
+depends_on = None
+
+
+def upgrade() -> None:
+    op.create_table(
+        "image_generation_config",
+        sa.Column("image_provider_id", sa.String(), primary_key=True),
+        sa.Column("model_configuration_id", sa.Integer(), nullable=False),
+        sa.Column("is_default", sa.Boolean(), nullable=False),
+        sa.ForeignKeyConstraint(
+            ["model_configuration_id"],
+            ["model_configuration.id"],
+            ondelete="CASCADE",
+        ),
+    )
+    op.create_index(
+        "ix_image_generation_config_is_default",
+        "image_generation_config",
+        ["is_default"],
+        unique=False,
+    )
+    op.create_index(
+        "ix_image_generation_config_model_configuration_id",
+        "image_generation_config",
+        ["model_configuration_id"],
+        unique=False,
+    )
+
+
+def downgrade() -> None:
+    op.drop_index(
+        "ix_image_generation_config_model_configuration_id",
+        table_name="image_generation_config",
+    )
+    op.drop_index(
+        "ix_image_generation_config_is_default", table_name="image_generation_config"
+    )
+    op.drop_table("image_generation_config")

backend/alembic/versions/7547d982db8f_chat_folders.py

@@ -45,8 +45,23 @@ def upgrade() -> None:
 
 
 def downgrade() -> None:
-    op.drop_constraint(
-        "chat_session_chat_folder_fk", "chat_session", type_="foreignkey"
-    )
-    op.drop_column("chat_session", "folder_id")
-    op.drop_table("chat_folder")
+    bind = op.get_bind()
+    inspector = sa.inspect(bind)
+
+    if "chat_session" in inspector.get_table_names():
+        chat_session_fks = {
+            fk.get("name") for fk in inspector.get_foreign_keys("chat_session")
+        }
+        if "chat_session_chat_folder_fk" in chat_session_fks:
+            op.drop_constraint(
+                "chat_session_chat_folder_fk", "chat_session", type_="foreignkey"
+            )
+
+        chat_session_columns = {
+            col["name"] for col in inspector.get_columns("chat_session")
+        }
+        if "folder_id" in chat_session_columns:
+            op.drop_column("chat_session", "folder_id")
+
+    if "chat_folder" in inspector.get_table_names():
+        op.drop_table("chat_folder")

backend/alembic/versions/7a70b7664e37_add_model_configuration_table.py

@@ -10,7 +10,7 @@ from alembic import op
 import sqlalchemy as sa
 from sqlalchemy.dialects import postgresql
 
-from onyx.llm.llm_provider_options import (
+from onyx.llm.well_known_providers.llm_provider_options import (
     fetch_model_names_for_provider_as_set,
     fetch_visible_model_names_for_provider_as_set,
 )

backend/alembic/versions/7bd55f264e1b_add_display_name_to_model_configuration.py

@@ -0,0 +1,27 @@
+"""Add display_name to model_configuration
+
+Revision ID: 7bd55f264e1b
+Revises: e8f0d2a38171
+Create Date: 2025-12-04
+
+"""
+
+from alembic import op
+import sqlalchemy as sa
+
+# revision identifiers, used by Alembic.
+revision = "7bd55f264e1b"
+down_revision = "e8f0d2a38171"
+branch_labels = None
+depends_on = None
+
+
+def upgrade() -> None:
+    op.add_column(
+        "model_configuration",
+        sa.Column("display_name", sa.String(), nullable=True),
+    )
+
+
+def downgrade() -> None:
+    op.drop_column("model_configuration", "display_name")

backend/alembic/versions/7cc3fcc116c1_user_file_uuid_primary_key_swap.py

@@ -0,0 +1,341 @@
+"""Migration 4: User file UUID primary key swap
+
+Revision ID: 7cc3fcc116c1
+Revises: 16c37a30adf2
+Create Date: 2025-09-22 09:54:38.292952
+
+This migration performs the critical UUID primary key swap on user_file table.
+It updates all foreign key references to use UUIDs instead of integers.
+"""
+
+from alembic import op
+import sqlalchemy as sa
+from sqlalchemy.dialects import postgresql as psql
+import logging
+
+logger = logging.getLogger("alembic.runtime.migration")
+
+# revision identifiers, used by Alembic.
+revision = "7cc3fcc116c1"
+down_revision = "16c37a30adf2"
+branch_labels = None
+depends_on = None
+
+
+def upgrade() -> None:
+    """Swap user_file primary key from integer to UUID."""
+
+    bind = op.get_bind()
+    inspector = sa.inspect(bind)
+
+    # Verify we're in the expected state
+    user_file_columns = [col["name"] for col in inspector.get_columns("user_file")]
+    if "new_id" not in user_file_columns:
+        logger.warning(
+            "user_file.new_id not found - migration may have already been applied"
+        )
+        return
+
+    logger.info("Starting UUID primary key swap...")
+
+    # === Step 1: Update persona__user_file foreign key to UUID ===
+    logger.info("Updating persona__user_file foreign key...")
+
+    # Drop existing foreign key constraints
+    op.execute(
+        "ALTER TABLE persona__user_file DROP CONSTRAINT IF EXISTS persona__user_file_user_file_id_uuid_fkey"
+    )
+    op.execute(
+        "ALTER TABLE persona__user_file DROP CONSTRAINT IF EXISTS persona__user_file_user_file_id_fkey"
+    )
+
+    # Create new foreign key to user_file.new_id
+    op.create_foreign_key(
+        "persona__user_file_user_file_id_fkey",
+        "persona__user_file",
+        "user_file",
+        local_cols=["user_file_id_uuid"],
+        remote_cols=["new_id"],
+    )
+
+    # Drop the old integer column and rename UUID column
+    op.execute("ALTER TABLE persona__user_file DROP COLUMN IF EXISTS user_file_id")
+    op.alter_column(
+        "persona__user_file",
+        "user_file_id_uuid",
+        new_column_name="user_file_id",
+        existing_type=psql.UUID(as_uuid=True),
+        nullable=False,
+    )
+
+    # Recreate composite primary key
+    op.execute(
+        "ALTER TABLE persona__user_file DROP CONSTRAINT IF EXISTS persona__user_file_pkey"
+    )
+    op.execute(
+        "ALTER TABLE persona__user_file ADD PRIMARY KEY (persona_id, user_file_id)"
+    )
+
+    logger.info("Updated persona__user_file to use UUID foreign key")
+
+    # === Step 2: Perform the primary key swap on user_file ===
+    logger.info("Swapping user_file primary key to UUID...")
+
+    # Drop the primary key constraint
+    op.execute("ALTER TABLE user_file DROP CONSTRAINT IF EXISTS user_file_pkey")
+
+    # Drop the old id column and rename new_id to id
+    op.execute("ALTER TABLE user_file DROP COLUMN IF EXISTS id")
+    op.alter_column(
+        "user_file",
+        "new_id",
+        new_column_name="id",
+        existing_type=psql.UUID(as_uuid=True),
+        nullable=False,
+    )
+
+    # Set default for new inserts
+    op.alter_column(
+        "user_file",
+        "id",
+        existing_type=psql.UUID(as_uuid=True),
+        server_default=sa.text("gen_random_uuid()"),
+    )
+
+    # Create new primary key
+    op.execute("ALTER TABLE user_file ADD PRIMARY KEY (id)")
+
+    logger.info("Swapped user_file primary key to UUID")
+
+    # === Step 3: Update foreign key constraints ===
+    logger.info("Updating foreign key constraints...")
+
+    # Recreate persona__user_file foreign key to point to user_file.id
+    # Drop existing FK first to break dependency on the unique constraint
+    op.execute(
+        "ALTER TABLE persona__user_file DROP CONSTRAINT IF EXISTS persona__user_file_user_file_id_fkey"
+    )
+    # Drop the unique constraint on (formerly) new_id BEFORE recreating the FK,
+    # so the FK will bind to the primary key instead of the unique index.
+    op.execute("ALTER TABLE user_file DROP CONSTRAINT IF EXISTS uq_user_file_new_id")
+    # Now recreate FK to the primary key column
+    op.create_foreign_key(
+        "persona__user_file_user_file_id_fkey",
+        "persona__user_file",
+        "user_file",
+        local_cols=["user_file_id"],
+        remote_cols=["id"],
+    )
+
+    # Add foreign keys for project__user_file
+    existing_fks = inspector.get_foreign_keys("project__user_file")
+
+    has_user_file_fk = any(
+        fk.get("referred_table") == "user_file"
+        and fk.get("constrained_columns") == ["user_file_id"]
+        for fk in existing_fks
+    )
+
+    if not has_user_file_fk:
+        op.create_foreign_key(
+            "fk_project__user_file_user_file_id",
+            "project__user_file",
+            "user_file",
+            ["user_file_id"],
+            ["id"],
+        )
+        logger.info("Added project__user_file -> user_file foreign key")
+
+    has_project_fk = any(
+        fk.get("referred_table") == "user_project"
+        and fk.get("constrained_columns") == ["project_id"]
+        for fk in existing_fks
+    )
+
+    if not has_project_fk:
+        op.create_foreign_key(
+            "fk_project__user_file_project_id",
+            "project__user_file",
+            "user_project",
+            ["project_id"],
+            ["id"],
+        )
+        logger.info("Added project__user_file -> user_project foreign key")
+
+    # === Step 4: Mark files for document_id migration ===
+    logger.info("Marking files for background document_id migration...")
+
+    logger.info("Migration 4 (UUID primary key swap) completed successfully")
+    logger.info(
+        "NOTE: Background task will update document IDs in Vespa and search_doc"
+    )
+
+
+def downgrade() -> None:
+    """Revert UUID primary key back to integer (data destructive!)."""
+
+    logger.error("CRITICAL: Downgrading UUID primary key swap is data destructive!")
+    logger.error(
+        "This will break all UUID-based references created after the migration."
+    )
+    logger.error("Only proceed if absolutely necessary and have backups.")
+
+    bind = op.get_bind()
+    inspector = sa.inspect(bind)
+
+    # Capture existing primary key definitions so we can restore them after swaps
+    persona_pk = inspector.get_pk_constraint("persona__user_file") or {}
+    persona_pk_name = persona_pk.get("name")
+    persona_pk_cols = persona_pk.get("constrained_columns") or []
+
+    project_pk = inspector.get_pk_constraint("project__user_file") or {}
+    project_pk_name = project_pk.get("name")
+    project_pk_cols = project_pk.get("constrained_columns") or []
+
+    # Drop foreign keys that reference the UUID primary key
+    op.drop_constraint(
+        "persona__user_file_user_file_id_fkey",
+        "persona__user_file",
+        type_="foreignkey",
+    )
+    op.drop_constraint(
+        "fk_project__user_file_user_file_id",
+        "project__user_file",
+        type_="foreignkey",
+    )
+
+    # Drop primary keys that rely on the UUID column so we can replace it
+    if persona_pk_name:
+        op.drop_constraint(persona_pk_name, "persona__user_file", type_="primary")
+    if project_pk_name:
+        op.drop_constraint(project_pk_name, "project__user_file", type_="primary")
+
+    # Rebuild integer IDs on user_file using a sequence-backed column
+    op.execute("CREATE SEQUENCE IF NOT EXISTS user_file_id_seq")
+    op.add_column(
+        "user_file",
+        sa.Column(
+            "id_int",
+            sa.Integer(),
+            server_default=sa.text("nextval('user_file_id_seq')"),
+            nullable=False,
+        ),
+    )
+    op.execute("ALTER SEQUENCE user_file_id_seq OWNED BY user_file.id_int")
+
+    # Prepare integer foreign key columns on referencing tables
+    op.add_column(
+        "persona__user_file",
+        sa.Column("user_file_id_int", sa.Integer(), nullable=True),
+    )
+    op.add_column(
+        "project__user_file",
+        sa.Column("user_file_id_int", sa.Integer(), nullable=True),
+    )
+
+    # Populate the new integer foreign key columns by mapping from the UUID IDs
+    op.execute(
+        """
+        UPDATE persona__user_file AS p
+        SET user_file_id_int = uf.id_int
+        FROM user_file AS uf
+        WHERE p.user_file_id = uf.id
+        """
+    )
+    op.execute(
+        """
+        UPDATE project__user_file AS p
+        SET user_file_id_int = uf.id_int
+        FROM user_file AS uf
+        WHERE p.user_file_id = uf.id
+        """
+    )
+
+    op.alter_column(
+        "persona__user_file",
+        "user_file_id_int",
+        existing_type=sa.Integer(),
+        nullable=False,
+    )
+    op.alter_column(
+        "project__user_file",
+        "user_file_id_int",
+        existing_type=sa.Integer(),
+        nullable=False,
+    )
+
+    # Remove the UUID foreign key columns and rename the integer replacements
+    op.drop_column("persona__user_file", "user_file_id")
+    op.alter_column(
+        "persona__user_file",
+        "user_file_id_int",
+        new_column_name="user_file_id",
+        existing_type=sa.Integer(),
+        nullable=False,
+    )
+
+    op.drop_column("project__user_file", "user_file_id")
+    op.alter_column(
+        "project__user_file",
+        "user_file_id_int",
+        new_column_name="user_file_id",
+        existing_type=sa.Integer(),
+        nullable=False,
+    )
+
+    # Swap the user_file primary key back to the integer column
+    op.drop_constraint("user_file_pkey", "user_file", type_="primary")
+    op.drop_column("user_file", "id")
+    op.alter_column(
+        "user_file",
+        "id_int",
+        new_column_name="id",
+        existing_type=sa.Integer(),
+    )
+    op.alter_column(
+        "user_file",
+        "id",
+        existing_type=sa.Integer(),
+        nullable=False,
+        server_default=sa.text("nextval('user_file_id_seq')"),
+    )
+    op.execute("ALTER SEQUENCE user_file_id_seq OWNED BY user_file.id")
+    op.execute(
+        """
+        SELECT setval(
+            'user_file_id_seq',
+            GREATEST(COALESCE(MAX(id), 1), 1),
+            MAX(id) IS NOT NULL
+        )
+        FROM user_file
+        """
+    )
+    op.create_primary_key("user_file_pkey", "user_file", ["id"])
+
+    # Restore primary keys on referencing tables
+    if persona_pk_cols:
+        op.create_primary_key(
+            "persona__user_file_pkey", "persona__user_file", persona_pk_cols
+        )
+    if project_pk_cols:
+        op.create_primary_key(
+            "project__user_file_pkey",
+            "project__user_file",
+            project_pk_cols,
+        )
+
+    # Recreate foreign keys pointing at the integer primary key
+    op.create_foreign_key(
+        "persona__user_file_user_file_id_fkey",
+        "persona__user_file",
+        "user_file",
+        ["user_file_id"],
+        ["id"],
+    )
+    op.create_foreign_key(
+        "fk_project__user_file_user_file_id",
+        "project__user_file",
+        "user_file",
+        ["user_file_id"],
+        ["id"],
+    )

backend/alembic/versions/7e490836d179_nullify_default_system_prompt.py

@@ -0,0 +1,80 @@
+"""nullify_default_system_prompt
+
+Revision ID: 7e490836d179
+Revises: c1d2e3f4a5b6
+Create Date: 2025-12-29 16:54:36.635574
+
+"""
+
+from alembic import op
+import sqlalchemy as sa
+
+
+# revision identifiers, used by Alembic.
+revision = "7e490836d179"
+down_revision = "c1d2e3f4a5b6"
+branch_labels = None
+depends_on = None
+
+
+# This is the default system prompt from the previous migration (87c52ec39f84)
+# ruff: noqa: E501, W605 start
+PREVIOUS_DEFAULT_SYSTEM_PROMPT = """
+You are a highly capable, thoughtful, and precise assistant. Your goal is to deeply understand the user's intent, ask clarifying questions when needed, think step-by-step through complex problems, provide clear and accurate answers, and proactively anticipate helpful follow-up information. Always prioritize being truthful, nuanced, insightful, and efficient.
+
+The current date is [[CURRENT_DATETIME]].[[CITATION_GUIDANCE]]
+
+# Response Style
+You use different text styles, bolding, emojis (sparingly), block quotes, and other formatting to make your responses more readable and engaging.
+You use proper Markdown and LaTeX to format your responses for math, scientific, and chemical formulas, symbols, etc.: '$$\\n[expression]\\n$$' for standalone cases and '\\( [expression] \\)' when inline.
+For code you prefer to use Markdown and specify the language.
+You can use horizontal rules (---) to separate sections of your responses.
+You can use Markdown tables to format your responses for data, lists, and other structured information.
+""".lstrip()
+# ruff: noqa: E501, W605 end
+
+
+def upgrade() -> None:
+    # Make system_prompt column nullable (model already has nullable=True but DB doesn't)
+    op.alter_column(
+        "persona",
+        "system_prompt",
+        nullable=True,
+    )
+
+    # Set system_prompt to NULL where it matches the previous default
+    conn = op.get_bind()
+    conn.execute(
+        sa.text(
+            """
+            UPDATE persona
+            SET system_prompt = NULL
+            WHERE system_prompt = :previous_default
+            """
+        ),
+        {"previous_default": PREVIOUS_DEFAULT_SYSTEM_PROMPT},
+    )
+
+
+def downgrade() -> None:
+    # Restore the default system prompt for personas that have NULL
+    # Note: This may restore the prompt to personas that originally had NULL
+    # before this migration, but there's no way to distinguish them
+    conn = op.get_bind()
+    conn.execute(
+        sa.text(
+            """
+            UPDATE persona
+            SET system_prompt = :previous_default
+            WHERE system_prompt IS NULL
+            """
+        ),
+        {"previous_default": PREVIOUS_DEFAULT_SYSTEM_PROMPT},
+    )
+
+    # Revert system_prompt column to not nullable
+    op.alter_column(
+        "persona",
+        "system_prompt",
+        nullable=False,
+    )

backend/alembic/versions/7ed603b64d5a_add_mcp_server_and_connection_config_.py

@@ -42,13 +42,13 @@ def upgrade() -> None:
         sa.Column(
             "created_at",
             sa.DateTime(timezone=True),
-            server_default=sa.text("now()"),  # type: ignore
+            server_default=sa.text("now()"),
             nullable=False,
         ),
         sa.Column(
             "updated_at",
             sa.DateTime(timezone=True),
-            server_default=sa.text("now()"),  # type: ignore
+            server_default=sa.text("now()"),
             nullable=False,
         ),
     )
@@ -63,13 +63,13 @@ def upgrade() -> None:
         sa.Column(
             "created_at",
             sa.DateTime(timezone=True),
-            server_default=sa.text("now()"),  # type: ignore
+            server_default=sa.text("now()"),
             nullable=False,
         ),
         sa.Column(
             "updated_at",
             sa.DateTime(timezone=True),
-            server_default=sa.text("now()"),  # type: ignore
+            server_default=sa.text("now()"),
             nullable=False,
         ),
         sa.ForeignKeyConstraint(

backend/alembic/versions/87c52ec39f84_update_default_system_prompt.py

@@ -0,0 +1,55 @@
+"""update_default_system_prompt
+
+Revision ID: 87c52ec39f84
+Revises: 7bd55f264e1b
+Create Date: 2025-12-05 15:54:06.002452
+
+"""
+
+from alembic import op
+import sqlalchemy as sa
+
+
+# revision identifiers, used by Alembic.
+revision = "87c52ec39f84"
+down_revision = "7bd55f264e1b"
+branch_labels = None
+depends_on = None
+
+
+DEFAULT_PERSONA_ID = 0
+
+# ruff: noqa: E501, W605 start
+DEFAULT_SYSTEM_PROMPT = """
+You are a highly capable, thoughtful, and precise assistant. Your goal is to deeply understand the user's intent, ask clarifying questions when needed, think step-by-step through complex problems, provide clear and accurate answers, and proactively anticipate helpful follow-up information. Always prioritize being truthful, nuanced, insightful, and efficient.
+
+The current date is [[CURRENT_DATETIME]].[[CITATION_GUIDANCE]]
+
+# Response Style
+You use different text styles, bolding, emojis (sparingly), block quotes, and other formatting to make your responses more readable and engaging.
+You use proper Markdown and LaTeX to format your responses for math, scientific, and chemical formulas, symbols, etc.: '$$\\n[expression]\\n$$' for standalone cases and '\\( [expression] \\)' when inline.
+For code you prefer to use Markdown and specify the language.
+You can use horizontal rules (---) to separate sections of your responses.
+You can use Markdown tables to format your responses for data, lists, and other structured information.
+""".lstrip()
+# ruff: noqa: E501, W605 end
+
+
+def upgrade() -> None:
+    conn = op.get_bind()
+    conn.execute(
+        sa.text(
+            """
+            UPDATE persona
+            SET system_prompt = :system_prompt
+            WHERE id = :persona_id
+            """
+        ),
+        {"system_prompt": DEFAULT_SYSTEM_PROMPT, "persona_id": DEFAULT_PERSONA_ID},
+    )
+
+
+def downgrade() -> None:
+    # We don't revert the system prompt on downgrade since we don't know
+    # what the previous value was. The new prompt is a reasonable default.
+    pass

backend/alembic/versions/9087b548dd69_seed_default_image_gen_config.py

@@ -0,0 +1,136 @@
+"""seed_default_image_gen_config
+
+Revision ID: 9087b548dd69
+Revises: 2b90f3af54b8
+Create Date: 2026-01-05 00:00:00.000000
+
+"""
+
+from alembic import op
+import sqlalchemy as sa
+
+
+# revision identifiers, used by Alembic.
+revision = "9087b548dd69"
+down_revision = "2b90f3af54b8"
+branch_labels = None
+depends_on = None
+
+# Constants for default image generation config
+# Source: web/src/app/admin/configuration/image-generation/constants.ts
+IMAGE_PROVIDER_ID = "openai_gpt_image_1"
+MODEL_NAME = "gpt-image-1"
+PROVIDER_NAME = "openai"
+
+
+def upgrade() -> None:
+    conn = op.get_bind()
+
+    # Check if image_generation_config table already has records
+    existing_configs = (
+        conn.execute(sa.text("SELECT COUNT(*) FROM image_generation_config")).scalar()
+        or 0
+    )
+
+    if existing_configs > 0:
+        # Skip if configs already exist - user may have configured manually
+        return
+
+    # Find the first OpenAI LLM provider
+    openai_provider = conn.execute(
+        sa.text(
+            """
+            SELECT id, api_key
+            FROM llm_provider
+            WHERE provider = :provider
+            ORDER BY id
+            LIMIT 1
+            """
+        ),
+        {"provider": PROVIDER_NAME},
+    ).fetchone()
+
+    if not openai_provider:
+        # No OpenAI provider found - nothing to do
+        return
+
+    source_provider_id, api_key = openai_provider
+
+    # Create new LLM provider for image generation (clone only api_key)
+    result = conn.execute(
+        sa.text(
+            """
+            INSERT INTO llm_provider (
+                name, provider, api_key, api_base, api_version,
+                deployment_name, default_model_name, is_public,
+                is_default_provider, is_default_vision_provider, is_auto_mode
+            )
+            VALUES (
+                :name, :provider, :api_key, NULL, NULL,
+                NULL, :default_model_name, :is_public,
+                NULL, NULL, :is_auto_mode
+            )
+            RETURNING id
+            """
+        ),
+        {
+            "name": f"Image Gen - {IMAGE_PROVIDER_ID}",
+            "provider": PROVIDER_NAME,
+            "api_key": api_key,
+            "default_model_name": MODEL_NAME,
+            "is_public": True,
+            "is_auto_mode": False,
+        },
+    )
+    new_provider_id = result.scalar()
+
+    # Create model configuration
+    result = conn.execute(
+        sa.text(
+            """
+            INSERT INTO model_configuration (
+                llm_provider_id, name, is_visible, max_input_tokens,
+                supports_image_input, display_name
+            )
+            VALUES (
+                :llm_provider_id, :name, :is_visible, :max_input_tokens,
+                :supports_image_input, :display_name
+            )
+            RETURNING id
+            """
+        ),
+        {
+            "llm_provider_id": new_provider_id,
+            "name": MODEL_NAME,
+            "is_visible": True,
+            "max_input_tokens": None,
+            "supports_image_input": False,
+            "display_name": None,
+        },
+    )
+    model_config_id = result.scalar()
+
+    # Create image generation config
+    conn.execute(
+        sa.text(
+            """
+            INSERT INTO image_generation_config (
+                image_provider_id, model_configuration_id, is_default
+            )
+            VALUES (
+                :image_provider_id, :model_configuration_id, :is_default
+            )
+            """
+        ),
+        {
+            "image_provider_id": IMAGE_PROVIDER_ID,
+            "model_configuration_id": model_config_id,
+            "is_default": True,
+        },
+    )
+
+
+def downgrade() -> None:
+    # We don't remove the config on downgrade since it's safe to keep around
+    # If we upgrade again, it will be a no-op due to the existing records check
+    pass

backend/alembic/versions/96a5702df6aa_mcp_tool_enabled.py

@@ -0,0 +1,45 @@
+"""mcp_tool_enabled
+
+Revision ID: 96a5702df6aa
+Revises: 40926a4dab77
+Create Date: 2025-10-09 12:10:21.733097
+
+"""
+
+from alembic import op
+import sqlalchemy as sa
+
+
+# revision identifiers, used by Alembic.
+revision = "96a5702df6aa"
+down_revision = "40926a4dab77"
+branch_labels = None
+depends_on = None
+
+
+DELETE_DISABLED_TOOLS_SQL = "DELETE FROM tool WHERE enabled = false"
+
+
+def upgrade() -> None:
+    op.add_column(
+        "tool",
+        sa.Column(
+            "enabled",
+            sa.Boolean(),
+            nullable=False,
+            server_default=sa.true(),
+        ),
+    )
+    op.create_index(
+        "ix_tool_mcp_server_enabled",
+        "tool",
+        ["mcp_server_id", "enabled"],
+    )
+    # Remove the server default so application controls defaulting
+    op.alter_column("tool", "enabled", server_default=None)
+
+
+def downgrade() -> None:
+    op.execute(DELETE_DISABLED_TOOLS_SQL)
+    op.drop_index("ix_tool_mcp_server_enabled", table_name="tool")
+    op.drop_column("tool", "enabled")

backend/alembic/versions/9a0296d7421e_add_is_auto_mode_to_llm_provider.py

@@ -0,0 +1,33 @@
+"""add_is_auto_mode_to_llm_provider
+
+Revision ID: 9a0296d7421e
+Revises: 7206234e012a
+Create Date: 2025-12-17 18:14:29.620981
+
+"""
+
+from alembic import op
+import sqlalchemy as sa
+
+
+# revision identifiers, used by Alembic.
+revision = "9a0296d7421e"
+down_revision = "7206234e012a"
+branch_labels = None
+depends_on = None
+
+
+def upgrade() -> None:
+    op.add_column(
+        "llm_provider",
+        sa.Column(
+            "is_auto_mode",
+            sa.Boolean(),
+            nullable=False,
+            server_default="false",
+        ),
+    )
+
+
+def downgrade() -> None:
+    op.drop_column("llm_provider", "is_auto_mode")

backend/alembic/versions/9b66d3156fc6_user_file_schema_additions.py

@@ -0,0 +1,268 @@
+"""Migration 1: User file schema additions
+
+Revision ID: 9b66d3156fc6
+Revises: b4ef3ae0bf6e
+Create Date: 2025-09-22 09:42:06.086732
+
+This migration adds new columns and tables without modifying existing data.
+It is safe to run and can be easily rolled back.
+"""
+
+from alembic import op
+import sqlalchemy as sa
+from sqlalchemy.dialects import postgresql as psql
+import logging
+
+logger = logging.getLogger("alembic.runtime.migration")
+# revision identifiers, used by Alembic.
+revision = "9b66d3156fc6"
+down_revision = "b4ef3ae0bf6e"
+branch_labels = None
+depends_on = None
+
+
+def upgrade() -> None:
+    """Add new columns and tables without modifying existing data."""
+
+    # Enable pgcrypto for UUID generation
+    op.execute("CREATE EXTENSION IF NOT EXISTS pgcrypto")
+
+    bind = op.get_bind()
+    inspector = sa.inspect(bind)
+
+    # === USER_FILE: Add new columns ===
+    logger.info("Adding new columns to user_file table...")
+
+    user_file_columns = [col["name"] for col in inspector.get_columns("user_file")]
+
+    # Check if ID is already UUID (in case of re-run after partial migration)
+    id_is_uuid = any(
+        col["name"] == "id" and "uuid" in str(col["type"]).lower()
+        for col in inspector.get_columns("user_file")
+    )
+
+    # Add transitional UUID column only if ID is not already UUID
+    if "new_id" not in user_file_columns and not id_is_uuid:
+        op.add_column(
+            "user_file",
+            sa.Column(
+                "new_id",
+                psql.UUID(as_uuid=True),
+                nullable=True,
+                server_default=sa.text("gen_random_uuid()"),
+            ),
+        )
+        op.create_unique_constraint("uq_user_file_new_id", "user_file", ["new_id"])
+        logger.info("Added new_id column to user_file")
+
+    # Add status column
+    if "status" not in user_file_columns:
+        op.add_column(
+            "user_file",
+            sa.Column(
+                "status",
+                sa.Enum(
+                    "PROCESSING",
+                    "COMPLETED",
+                    "FAILED",
+                    "CANCELED",
+                    name="userfilestatus",
+                    native_enum=False,
+                ),
+                nullable=False,
+                server_default="PROCESSING",
+            ),
+        )
+        logger.info("Added status column to user_file")
+
+    # Add other tracking columns
+    if "chunk_count" not in user_file_columns:
+        op.add_column(
+            "user_file", sa.Column("chunk_count", sa.Integer(), nullable=True)
+        )
+        logger.info("Added chunk_count column to user_file")
+
+    if "last_accessed_at" not in user_file_columns:
+        op.add_column(
+            "user_file",
+            sa.Column("last_accessed_at", sa.DateTime(timezone=True), nullable=True),
+        )
+        logger.info("Added last_accessed_at column to user_file")
+
+    if "needs_project_sync" not in user_file_columns:
+        op.add_column(
+            "user_file",
+            sa.Column(
+                "needs_project_sync",
+                sa.Boolean(),
+                nullable=False,
+                server_default=sa.text("false"),
+            ),
+        )
+        logger.info("Added needs_project_sync column to user_file")
+
+    if "last_project_sync_at" not in user_file_columns:
+        op.add_column(
+            "user_file",
+            sa.Column(
+                "last_project_sync_at", sa.DateTime(timezone=True), nullable=True
+            ),
+        )
+        logger.info("Added last_project_sync_at column to user_file")
+
+    if "document_id_migrated" not in user_file_columns:
+        op.add_column(
+            "user_file",
+            sa.Column(
+                "document_id_migrated",
+                sa.Boolean(),
+                nullable=False,
+                server_default=sa.text("true"),
+            ),
+        )
+        logger.info("Added document_id_migrated column to user_file")
+
+    # === USER_FOLDER -> USER_PROJECT rename ===
+    table_names = set(inspector.get_table_names())
+
+    if "user_folder" in table_names:
+        logger.info("Updating user_folder table...")
+        # Make description nullable first
+        op.alter_column("user_folder", "description", nullable=True)
+
+        # Rename table if user_project doesn't exist
+        if "user_project" not in table_names:
+            op.execute("ALTER TABLE user_folder RENAME TO user_project")
+            logger.info("Renamed user_folder to user_project")
+    elif "user_project" in table_names:
+        # If already renamed, ensure column nullability
+        project_cols = [col["name"] for col in inspector.get_columns("user_project")]
+        if "description" in project_cols:
+            op.alter_column("user_project", "description", nullable=True)
+
+    # Add instructions column to user_project
+    inspector = sa.inspect(bind)  # Refresh after rename
+    if "user_project" in inspector.get_table_names():
+        project_columns = [col["name"] for col in inspector.get_columns("user_project")]
+        if "instructions" not in project_columns:
+            op.add_column(
+                "user_project",
+                sa.Column("instructions", sa.String(), nullable=True),
+            )
+            logger.info("Added instructions column to user_project")
+
+    # === CHAT_SESSION: Add project_id ===
+    chat_session_columns = [
+        col["name"] for col in inspector.get_columns("chat_session")
+    ]
+    if "project_id" not in chat_session_columns:
+        op.add_column(
+            "chat_session",
+            sa.Column("project_id", sa.Integer(), nullable=True),
+        )
+        logger.info("Added project_id column to chat_session")
+
+    # === PERSONA__USER_FILE: Add UUID column ===
+    persona_user_file_columns = [
+        col["name"] for col in inspector.get_columns("persona__user_file")
+    ]
+    if "user_file_id_uuid" not in persona_user_file_columns:
+        op.add_column(
+            "persona__user_file",
+            sa.Column("user_file_id_uuid", psql.UUID(as_uuid=True), nullable=True),
+        )
+        logger.info("Added user_file_id_uuid column to persona__user_file")
+
+    # === PROJECT__USER_FILE: Create new table ===
+    if "project__user_file" not in inspector.get_table_names():
+        op.create_table(
+            "project__user_file",
+            sa.Column("project_id", sa.Integer(), nullable=False),
+            sa.Column("user_file_id", psql.UUID(as_uuid=True), nullable=False),
+            sa.PrimaryKeyConstraint("project_id", "user_file_id"),
+        )
+        logger.info("Created project__user_file table")
+
+    # Only create the index if it doesn't exist
+    existing_indexes = [
+        ix["name"] for ix in inspector.get_indexes("project__user_file")
+    ]
+    if "idx_project__user_file_user_file_id" not in existing_indexes:
+        op.create_index(
+            "idx_project__user_file_user_file_id",
+            "project__user_file",
+            ["user_file_id"],
+        )
+        logger.info(
+            "Created index idx_project__user_file_user_file_id on project__user_file"
+        )
+
+    logger.info("Migration 1 (schema additions) completed successfully")
+
+
+def downgrade() -> None:
+    """Remove added columns and tables."""
+
+    bind = op.get_bind()
+    inspector = sa.inspect(bind)
+
+    logger.info("Starting downgrade of schema additions...")
+
+    # Drop project__user_file table
+    if "project__user_file" in inspector.get_table_names():
+        # op.drop_index("idx_project__user_file_user_file_id", "project__user_file")
+        op.drop_table("project__user_file")
+        logger.info("Dropped project__user_file table")
+
+    # Remove columns from persona__user_file
+    if "persona__user_file" in inspector.get_table_names():
+        columns = [col["name"] for col in inspector.get_columns("persona__user_file")]
+        if "user_file_id_uuid" in columns:
+            op.drop_column("persona__user_file", "user_file_id_uuid")
+            logger.info("Dropped user_file_id_uuid from persona__user_file")
+
+    # Remove columns from chat_session
+    if "chat_session" in inspector.get_table_names():
+        columns = [col["name"] for col in inspector.get_columns("chat_session")]
+        if "project_id" in columns:
+            op.drop_column("chat_session", "project_id")
+            logger.info("Dropped project_id from chat_session")
+
+    # Rename user_project back to user_folder and remove instructions
+    if "user_project" in inspector.get_table_names():
+        columns = [col["name"] for col in inspector.get_columns("user_project")]
+        if "instructions" in columns:
+            op.drop_column("user_project", "instructions")
+        op.execute("ALTER TABLE user_project RENAME TO user_folder")
+        # Update NULL descriptions to empty string before setting NOT NULL constraint
+        op.execute("UPDATE user_folder SET description = '' WHERE description IS NULL")
+        op.alter_column("user_folder", "description", nullable=False)
+        logger.info("Renamed user_project back to user_folder")
+
+    # Remove columns from user_file
+    if "user_file" in inspector.get_table_names():
+        columns = [col["name"] for col in inspector.get_columns("user_file")]
+
+        columns_to_drop = [
+            "document_id_migrated",
+            "last_project_sync_at",
+            "needs_project_sync",
+            "last_accessed_at",
+            "chunk_count",
+            "status",
+        ]
+
+        for col in columns_to_drop:
+            if col in columns:
+                op.drop_column("user_file", col)
+                logger.info(f"Dropped {col} from user_file")
+
+        if "new_id" in columns:
+            op.drop_constraint("uq_user_file_new_id", "user_file", type_="unique")
+            op.drop_column("user_file", "new_id")
+            logger.info("Dropped new_id from user_file")
+
+    # Drop enum type if no columns use it
+    bind.execute(sa.text("DROP TYPE IF EXISTS userfilestatus"))
+
+    logger.info("Downgrade completed successfully")

backend/alembic/versions/9drpiiw74ljy_add_config_to_federated_connector.py

@@ -0,0 +1,97 @@
+"""add config to federated_connector
+
+Revision ID: 9drpiiw74ljy
+Revises: 2acdef638fc2
+Create Date: 2025-11-03 12:00:00.000000
+
+"""
+
+from alembic import op
+import sqlalchemy as sa
+from sqlalchemy.dialects import postgresql
+
+# revision identifiers, used by Alembic.
+revision = "9drpiiw74ljy"
+down_revision = "2acdef638fc2"
+branch_labels = None
+depends_on = None
+
+
+def upgrade() -> None:
+    connection = op.get_bind()
+
+    # Check if column already exists in current schema
+    result = connection.execute(
+        sa.text(
+            """
+            SELECT column_name
+            FROM information_schema.columns
+            WHERE table_schema = current_schema()
+            AND table_name = 'federated_connector'
+            AND column_name = 'config'
+            """
+        )
+    )
+    column_exists = result.fetchone() is not None
+
+    # Add config column with default empty object (only if it doesn't exist)
+    if not column_exists:
+        op.add_column(
+            "federated_connector",
+            sa.Column(
+                "config", postgresql.JSONB(), nullable=False, server_default="{}"
+            ),
+        )
+
+    # Data migration: Single bulk update for all Slack connectors
+    connection.execute(
+        sa.text(
+            """
+            WITH connector_configs AS (
+                SELECT
+                    fc.id as connector_id,
+                    CASE
+                        WHEN fcds.entities->'channels' IS NOT NULL
+                            AND jsonb_typeof(fcds.entities->'channels') = 'array'
+                            AND jsonb_array_length(fcds.entities->'channels') > 0
+                        THEN
+                            jsonb_build_object(
+                                'channels', fcds.entities->'channels',
+                                'search_all_channels', false
+                            ) ||
+                            CASE
+                                WHEN fcds.entities->'include_dm' IS NOT NULL
+                                THEN jsonb_build_object('include_dm', fcds.entities->'include_dm')
+                                ELSE '{}'::jsonb
+                            END
+                        ELSE
+                            jsonb_build_object('search_all_channels', true) ||
+                            CASE
+                                WHEN fcds.entities->'include_dm' IS NOT NULL
+                                THEN jsonb_build_object('include_dm', fcds.entities->'include_dm')
+                                ELSE '{}'::jsonb
+                            END
+                    END as config
+                FROM federated_connector fc
+                LEFT JOIN LATERAL (
+                    SELECT entities
+                    FROM federated_connector__document_set
+                    WHERE federated_connector_id = fc.id
+                    AND entities IS NOT NULL
+                    ORDER BY id
+                    LIMIT 1
+                ) fcds ON true
+                WHERE fc.source = 'FEDERATED_SLACK'
+                AND fcds.entities IS NOT NULL
+            )
+            UPDATE federated_connector fc
+            SET config = cc.config
+            FROM connector_configs cc
+            WHERE fc.id = cc.connector_id
+            """
+        )
+    )
+
+
+def downgrade() -> None:
+    op.drop_column("federated_connector", "config")

backend/alembic/versions/a01bf2971c5d_update_default_tool_descriptions.py

@@ -0,0 +1,62 @@
+"""update_default_tool_descriptions
+
+Revision ID: a01bf2971c5d
+Revises: 87c52ec39f84
+Create Date: 2025-12-16 15:21:25.656375
+
+"""
+
+from alembic import op
+import sqlalchemy as sa
+
+
+# revision identifiers, used by Alembic.
+revision = "a01bf2971c5d"
+down_revision = "18b5b2524446"
+branch_labels = None
+depends_on = None
+
+# new tool descriptions (12/2025)
+TOOL_DESCRIPTIONS = {
+    "SearchTool": "The Search Action allows the agent to search through connected knowledge to help build an answer.",
+    "ImageGenerationTool": (
+        "The Image Generation Action allows the agent to use DALL-E 3 or GPT-IMAGE-1 to generate images. "
+        "The action will be used when the user asks the agent to generate an image."
+    ),
+    "WebSearchTool": (
+        "The Web Search Action allows the agent "
+        "to perform internet searches for up-to-date information."
+    ),
+    "KnowledgeGraphTool": (
+        "The Knowledge Graph Search Action allows the agent to search the "
+        "Knowledge Graph for information. This tool can (for now) only be active in the KG Beta Agent, "
+        "and it requires the Knowledge Graph to be enabled."
+    ),
+    "OktaProfileTool": (
+        "The Okta Profile Action allows the agent to fetch the current user's information from Okta. "
+        "This may include the user's name, email, phone number, address, and other details such as their "
+        "manager and direct reports."
+    ),
+}
+
+
+def upgrade() -> None:
+    conn = op.get_bind()
+    conn.execute(sa.text("BEGIN"))
+
+    try:
+        for tool_id, description in TOOL_DESCRIPTIONS.items():
+            conn.execute(
+                sa.text(
+                    "UPDATE tool SET description = :description WHERE in_code_tool_id = :tool_id"
+                ),
+                {"description": description, "tool_id": tool_id},
+            )
+        conn.execute(sa.text("COMMIT"))
+    except Exception as e:
+        conn.execute(sa.text("ROLLBACK"))
+        raise e
+
+
+def downgrade() -> None:
+    pass

backend/alembic/versions/a1b2c3d4e5f6_add_license_table.py

@@ -0,0 +1,49 @@
+"""add license table
+
+Revision ID: a1b2c3d4e5f6
+Revises: a01bf2971c5d
+Create Date: 2025-12-04 10:00:00.000000
+
+"""
+
+from alembic import op
+import sqlalchemy as sa
+
+# revision identifiers, used by Alembic.
+revision = "a1b2c3d4e5f6"
+down_revision = "a01bf2971c5d"
+branch_labels = None
+depends_on = None
+
+
+def upgrade() -> None:
+    op.create_table(
+        "license",
+        sa.Column("id", sa.Integer(), primary_key=True),
+        sa.Column("license_data", sa.Text(), nullable=False),
+        sa.Column(
+            "created_at",
+            sa.DateTime(timezone=True),
+            server_default=sa.func.now(),
+            nullable=False,
+        ),
+        sa.Column(
+            "updated_at",
+            sa.DateTime(timezone=True),
+            server_default=sa.func.now(),
+            nullable=False,
+        ),
+    )
+
+    # Singleton pattern - only ever one row in this table
+    op.create_index(
+        "idx_license_singleton",
+        "license",
+        [sa.text("(true)")],
+        unique=True,
+    )
+
+
+def downgrade() -> None:
+    op.drop_index("idx_license_singleton", table_name="license")
+    op.drop_table("license")

backend/alembic/versions/a2b3c4d5e6f7_remove_fast_default_model_name.py

@@ -0,0 +1,27 @@
+"""Remove fast_default_model_name from llm_provider
+
+Revision ID: a2b3c4d5e6f7
+Revises: 2a391f840e85
+Create Date: 2024-12-17
+
+"""
+
+from alembic import op
+import sqlalchemy as sa
+
+# revision identifiers, used by Alembic.
+revision = "a2b3c4d5e6f7"
+down_revision = "2a391f840e85"
+branch_labels: None = None
+depends_on: None = None
+
+
+def upgrade() -> None:
+    op.drop_column("llm_provider", "fast_default_model_name")
+
+
+def downgrade() -> None:
+    op.add_column(
+        "llm_provider",
+        sa.Column("fast_default_model_name", sa.String(), nullable=True),
+    )

backend/alembic/versions/a3c1a7904cd0_remove_userfile_related_deprecated_.py

@@ -0,0 +1,39 @@
+"""remove userfile related deprecated fields
+
+Revision ID: a3c1a7904cd0
+Revises: 5c3dca366b35
+Create Date: 2026-01-06 13:00:30.634396
+
+"""
+
+from alembic import op
+import sqlalchemy as sa
+
+# revision identifiers, used by Alembic.
+revision = "a3c1a7904cd0"
+down_revision = "5c3dca366b35"
+branch_labels = None
+depends_on = None
+
+
+def upgrade() -> None:
+    op.drop_column("user_file", "document_id")
+    op.drop_column("user_file", "document_id_migrated")
+    op.drop_column("connector_credential_pair", "is_user_file")
+
+
+def downgrade() -> None:
+    op.add_column(
+        "connector_credential_pair",
+        sa.Column("is_user_file", sa.Boolean(), nullable=False, server_default="false"),
+    )
+    op.add_column(
+        "user_file",
+        sa.Column("document_id", sa.String(), nullable=True),
+    )
+    op.add_column(
+        "user_file",
+        sa.Column(
+            "document_id_migrated", sa.Boolean(), nullable=False, server_default="true"
+        ),
+    )