debug

Co-authored-by: Claude Opus 4.5 <noreply@anthropic.com>

.github/CODEOWNERS

@@ -6,5 +6,5 @@
 /web/STANDARDS.md @raunakab @Weves
 
 # Agent context files
-/CLAUDE.md.template @Weves
-/AGENTS.md.template @Weves
+/CLAUDE.md @Weves
+/AGENTS.md @Weves

.github/pull_request_template.md

@@ -8,4 +8,5 @@
 
 ## Additional Options
 
+- [ ] [Required] I have considered whether this PR needs to be cherry-picked to the latest beta branch.
 - [ ] [Optional] Override Linear Check

.github/workflows/deployment.yml

@@ -26,12 +26,14 @@ jobs:
       build-web: ${{ steps.check.outputs.build-web }}
       build-web-cloud: ${{ steps.check.outputs.build-web-cloud }}
       build-backend: ${{ steps.check.outputs.build-backend }}
+      build-backend-craft: ${{ steps.check.outputs.build-backend-craft }}
       build-model-server: ${{ steps.check.outputs.build-model-server }}
       is-cloud-tag: ${{ steps.check.outputs.is-cloud-tag }}
       is-stable: ${{ steps.check.outputs.is-stable }}
       is-beta: ${{ steps.check.outputs.is-beta }}
       is-stable-standalone: ${{ steps.check.outputs.is-stable-standalone }}
       is-beta-standalone: ${{ steps.check.outputs.is-beta-standalone }}
+      is-craft-latest: ${{ steps.check.outputs.is-craft-latest }}
       is-test-run: ${{ steps.check.outputs.is-test-run }}
       sanitized-tag: ${{ steps.check.outputs.sanitized-tag }}
       short-sha: ${{ steps.check.outputs.short-sha }}
@@ -54,15 +56,20 @@ jobs:
           IS_BETA=false
           IS_STABLE_STANDALONE=false
           IS_BETA_STANDALONE=false
+          IS_CRAFT_LATEST=false
           IS_PROD_TAG=false
           IS_TEST_RUN=false
           BUILD_DESKTOP=false
           BUILD_WEB=false
           BUILD_WEB_CLOUD=false
           BUILD_BACKEND=true
+          BUILD_BACKEND_CRAFT=false
           BUILD_MODEL_SERVER=true
 
           # Determine tag type based on pattern matching (do regex checks once)
+          if [[ "$TAG" == craft-* ]]; then
+            IS_CRAFT_LATEST=true
+          fi
           if [[ "$TAG" == *cloud* ]]; then
             IS_CLOUD=true
           fi
@@ -75,7 +82,7 @@ jobs:
           if [[ "$TAG" =~ ^v[0-9]+\.[0-9]+\.[0-9]+$ ]]; then
             IS_STABLE=true
           fi
-          if [[ "$TAG" =~ ^v[0-9]+\.[0-9]+\.[0-9]+-beta\.[0-9]+$ ]]; then
+          if [[ "$TAG" =~ ^v[0-9]+\.[0-9]+\.[0-9]+-beta(\.[0-9]+)?$ ]]; then
             IS_BETA=true
           fi
 
@@ -90,6 +97,12 @@ jobs:
             fi
           fi
 
+          # Craft-latest builds backend with Craft enabled
+          if [[ "$IS_CRAFT_LATEST" == "true" ]]; then
+            BUILD_BACKEND_CRAFT=true
+            BUILD_BACKEND=false
+          fi
+
           # Standalone version checks (for backend/model-server - version excluding cloud tags)
           if [[ "$IS_STABLE" == "true" ]] && [[ "$IS_CLOUD" != "true" ]]; then
             IS_STABLE_STANDALONE=true
@@ -113,12 +126,14 @@ jobs:
             echo "build-web=$BUILD_WEB"
             echo "build-web-cloud=$BUILD_WEB_CLOUD"
             echo "build-backend=$BUILD_BACKEND"
+            echo "build-backend-craft=$BUILD_BACKEND_CRAFT"
             echo "build-model-server=$BUILD_MODEL_SERVER"
             echo "is-cloud-tag=$IS_CLOUD"
             echo "is-stable=$IS_STABLE"
             echo "is-beta=$IS_BETA"
             echo "is-stable-standalone=$IS_STABLE_STANDALONE"
             echo "is-beta-standalone=$IS_BETA_STANDALONE"
+            echo "is-craft-latest=$IS_CRAFT_LATEST"
             echo "is-test-run=$IS_TEST_RUN"
             echo "sanitized-tag=$SANITIZED_TAG"
             echo "short-sha=$SHORT_SHA"
@@ -130,13 +145,13 @@ jobs:
     if: ${{ !startsWith(github.ref_name, 'nightly-latest') && github.event_name != 'workflow_dispatch' }}
     steps:
       - name: Checkout
-        uses: actions/checkout@8e8c483db84b4bee98b60c0593521ed34d9990e8 # ratchet:actions/checkout@v6
+        uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # ratchet:actions/checkout@v6
         with:
           persist-credentials: false
           fetch-depth: 0
 
       - name: Setup uv
-        uses: astral-sh/setup-uv@ed21f2f24f8dd64503750218de024bcf64c7250a # ratchet:astral-sh/setup-uv@v7
+        uses: astral-sh/setup-uv@61cb8a9741eeb8a550a1b8544337180c0fc8476b # ratchet:astral-sh/setup-uv@v7
         with:
           version: "0.9.9"
           # NOTE: This isn't caching much and zizmor suggests this could be poisoned, so disable.
@@ -155,27 +170,14 @@ jobs:
     environment: release
     steps:
       - name: Checkout
-        uses: actions/checkout@8e8c483db84b4bee98b60c0593521ed34d9990e8 # ratchet:actions/checkout@v6
+        uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # ratchet:actions/checkout@v6
         with:
           persist-credentials: false
 
-      - name: Configure AWS credentials
-        uses: aws-actions/configure-aws-credentials@61815dcd50bd041e203e49132bacad1fd04d2708
-        with:
-          role-to-assume: ${{ secrets.AWS_OIDC_ROLE_ARN }}
-          aws-region: us-east-2
-
-      - name: Get AWS Secrets
-        uses: aws-actions/aws-secretsmanager-get-secrets@a9a7eb4e2f2871d30dc5b892576fde60a2ecc802
-        with:
-          secret-ids: |
-            MONITOR_DEPLOYMENTS_WEBHOOK, deploy/monitor-deployments-webhook
-          parse-json-secrets: true
-
       - name: Send Slack notification
         uses: ./.github/actions/slack-notify
         with:
-          webhook-url: ${{ env.MONITOR_DEPLOYMENTS_WEBHOOK }}
+          webhook-url: ${{ secrets.MONITOR_DEPLOYMENTS_WEBHOOK }}
           failed-jobs: "• check-version-tag"
           title: "🚨 Version Tag Check Failed"
           ref-name: ${{ github.ref_name }}
@@ -204,7 +206,7 @@ jobs:
     timeout-minutes: 90
     environment: release
     steps:
-      - uses: actions/checkout@8e8c483db84b4bee98b60c0593521ed34d9990e8 # ratchet:actions/checkout@v6.0.1
+      - uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # ratchet:actions/checkout@v6.0.2
         with:
           # NOTE: persist-credentials is needed for tauri-action to create GitHub releases.
           persist-credentials: true # zizmor: ignore[artipacked]
@@ -377,7 +379,7 @@ jobs:
       - uses: runs-on/action@cd2b598b0515d39d78c38a02d529db87d2196d1e # ratchet:runs-on/action@v2
 
       - name: Checkout
-        uses: actions/checkout@8e8c483db84b4bee98b60c0593521ed34d9990e8 # ratchet:actions/checkout@v6
+        uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # ratchet:actions/checkout@v6
         with:
           persist-credentials: false
 
@@ -450,7 +452,7 @@ jobs:
       - uses: runs-on/action@cd2b598b0515d39d78c38a02d529db87d2196d1e # ratchet:runs-on/action@v2
 
       - name: Checkout
-        uses: actions/checkout@8e8c483db84b4bee98b60c0593521ed34d9990e8 # ratchet:actions/checkout@v6
+        uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # ratchet:actions/checkout@v6
         with:
           persist-credentials: false
 
@@ -588,7 +590,7 @@ jobs:
       - uses: runs-on/action@cd2b598b0515d39d78c38a02d529db87d2196d1e # ratchet:runs-on/action@v2
 
       - name: Checkout
-        uses: actions/checkout@8e8c483db84b4bee98b60c0593521ed34d9990e8 # ratchet:actions/checkout@v6
+        uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # ratchet:actions/checkout@v6
         with:
           persist-credentials: false
 
@@ -669,7 +671,7 @@ jobs:
       - uses: runs-on/action@cd2b598b0515d39d78c38a02d529db87d2196d1e # ratchet:runs-on/action@v2
 
       - name: Checkout
-        uses: actions/checkout@8e8c483db84b4bee98b60c0593521ed34d9990e8 # ratchet:actions/checkout@v6
+        uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # ratchet:actions/checkout@v6
         with:
           persist-credentials: false
 
@@ -812,7 +814,7 @@ jobs:
       - uses: runs-on/action@cd2b598b0515d39d78c38a02d529db87d2196d1e # ratchet:runs-on/action@v2
 
       - name: Checkout code
-        uses: actions/checkout@8e8c483db84b4bee98b60c0593521ed34d9990e8 # ratchet:actions/checkout@v6
+        uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # ratchet:actions/checkout@v6
         with:
           persist-credentials: false
 
@@ -884,7 +886,7 @@ jobs:
       - uses: runs-on/action@cd2b598b0515d39d78c38a02d529db87d2196d1e # ratchet:runs-on/action@v2
 
       - name: Checkout code
-        uses: actions/checkout@8e8c483db84b4bee98b60c0593521ed34d9990e8 # ratchet:actions/checkout@v6
+        uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # ratchet:actions/checkout@v6
         with:
           persist-credentials: false
 
@@ -1003,6 +1005,217 @@ jobs:
             $(printf '%s\n' "${META_TAGS}" | xargs -I {} echo -t {}) \
             $IMAGES
 
+  build-backend-craft-amd64:
+    needs: determine-builds
+    if: needs.determine-builds.outputs.build-backend-craft == 'true'
+    runs-on:
+      - runs-on
+      - runner=2cpu-linux-x64
+      - run-id=${{ github.run_id }}-backend-craft-amd64
+      - extras=ecr-cache
+    timeout-minutes: 90
+    environment: release
+    outputs:
+      digest: ${{ steps.build.outputs.digest }}
+    env:
+      REGISTRY_IMAGE: onyxdotapp/onyx-backend
+    steps:
+      - uses: runs-on/action@cd2b598b0515d39d78c38a02d529db87d2196d1e # ratchet:runs-on/action@v2
+
+      - name: Checkout code
+        uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # ratchet:actions/checkout@v6
+        with:
+          persist-credentials: false
+
+      - name: Configure AWS credentials
+        uses: aws-actions/configure-aws-credentials@61815dcd50bd041e203e49132bacad1fd04d2708
+        with:
+          role-to-assume: ${{ secrets.AWS_OIDC_ROLE_ARN }}
+          aws-region: us-east-2
+
+      - name: Get AWS Secrets
+        uses: aws-actions/aws-secretsmanager-get-secrets@a9a7eb4e2f2871d30dc5b892576fde60a2ecc802
+        with:
+          secret-ids: |
+            DOCKER_USERNAME, deploy/docker-username
+            DOCKER_TOKEN, deploy/docker-token
+          parse-json-secrets: true
+
+      - name: Docker meta
+        id: meta
+        uses: docker/metadata-action@c299e40c65443455700f0fdfc63efafe5b349051 # ratchet:docker/metadata-action@v5
+        with:
+          images: ${{ env.REGISTRY_IMAGE }}
+          flavor: |
+            latest=false
+
+      - name: Set up Docker Buildx
+        uses: docker/setup-buildx-action@8d2750c68a42422c14e847fe6c8ac0403b4cbd6f # ratchet:docker/setup-buildx-action@v3
+
+      - name: Login to Docker Hub
+        uses: docker/login-action@5e57cd118135c172c3672efd75eb46360885c0ef # ratchet:docker/login-action@v3
+        with:
+          username: ${{ env.DOCKER_USERNAME }}
+          password: ${{ env.DOCKER_TOKEN }}
+
+      - name: Build and push AMD64
+        id: build
+        uses: docker/build-push-action@263435318d21b8e681c14492fe198d362a7d2c83 # ratchet:docker/build-push-action@v6
+        with:
+          context: ./backend
+          file: ./backend/Dockerfile
+          platforms: linux/amd64
+          labels: ${{ steps.meta.outputs.labels }}
+          build-args: |
+            ONYX_VERSION=${{ github.ref_name }}
+            ENABLE_CRAFT=true
+          cache-from: |
+            type=registry,ref=${{ env.REGISTRY_IMAGE }}:latest
+            type=registry,ref=${{ env.RUNS_ON_ECR_CACHE }}:backend-craft-cache-amd64
+          cache-to: |
+            type=inline
+            type=registry,ref=${{ env.RUNS_ON_ECR_CACHE }}:backend-craft-cache-amd64,mode=max
+          outputs: type=image,name=${{ env.REGISTRY_IMAGE }},push-by-digest=true,name-canonical=true,push=true
+          no-cache: ${{ vars.DOCKER_NO_CACHE == 'true' }}
+
+  build-backend-craft-arm64:
+    needs: determine-builds
+    if: needs.determine-builds.outputs.build-backend-craft == 'true'
+    runs-on:
+      - runs-on
+      - runner=2cpu-linux-arm64
+      - run-id=${{ github.run_id }}-backend-craft-arm64
+      - extras=ecr-cache
+    timeout-minutes: 90
+    environment: release
+    outputs:
+      digest: ${{ steps.build.outputs.digest }}
+    env:
+      REGISTRY_IMAGE: onyxdotapp/onyx-backend
+    steps:
+      - uses: runs-on/action@cd2b598b0515d39d78c38a02d529db87d2196d1e # ratchet:runs-on/action@v2
+
+      - name: Checkout code
+        uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # ratchet:actions/checkout@v6
+        with:
+          persist-credentials: false
+
+      - name: Configure AWS credentials
+        uses: aws-actions/configure-aws-credentials@61815dcd50bd041e203e49132bacad1fd04d2708
+        with:
+          role-to-assume: ${{ secrets.AWS_OIDC_ROLE_ARN }}
+          aws-region: us-east-2
+
+      - name: Get AWS Secrets
+        uses: aws-actions/aws-secretsmanager-get-secrets@a9a7eb4e2f2871d30dc5b892576fde60a2ecc802
+        with:
+          secret-ids: |
+            DOCKER_USERNAME, deploy/docker-username
+            DOCKER_TOKEN, deploy/docker-token
+          parse-json-secrets: true
+
+      - name: Docker meta
+        id: meta
+        uses: docker/metadata-action@c299e40c65443455700f0fdfc63efafe5b349051 # ratchet:docker/metadata-action@v5
+        with:
+          images: ${{ env.REGISTRY_IMAGE }}
+          flavor: |
+            latest=false
+
+      - name: Set up Docker Buildx
+        uses: docker/setup-buildx-action@8d2750c68a42422c14e847fe6c8ac0403b4cbd6f # ratchet:docker/setup-buildx-action@v3
+
+      - name: Login to Docker Hub
+        uses: docker/login-action@5e57cd118135c172c3672efd75eb46360885c0ef # ratchet:docker/login-action@v3
+        with:
+          username: ${{ env.DOCKER_USERNAME }}
+          password: ${{ env.DOCKER_TOKEN }}
+
+      - name: Build and push ARM64
+        id: build
+        uses: docker/build-push-action@263435318d21b8e681c14492fe198d362a7d2c83 # ratchet:docker/build-push-action@v6
+        with:
+          context: ./backend
+          file: ./backend/Dockerfile
+          platforms: linux/arm64
+          labels: ${{ steps.meta.outputs.labels }}
+          build-args: |
+            ONYX_VERSION=${{ github.ref_name }}
+            ENABLE_CRAFT=true
+          cache-from: |
+            type=registry,ref=${{ env.REGISTRY_IMAGE }}:latest
+            type=registry,ref=${{ env.RUNS_ON_ECR_CACHE }}:backend-craft-cache-arm64
+          cache-to: |
+            type=inline
+            type=registry,ref=${{ env.RUNS_ON_ECR_CACHE }}:backend-craft-cache-arm64,mode=max
+          outputs: type=image,name=${{ env.REGISTRY_IMAGE }},push-by-digest=true,name-canonical=true,push=true
+          no-cache: ${{ vars.DOCKER_NO_CACHE == 'true' }}
+
+  merge-backend-craft:
+    needs:
+      - determine-builds
+      - build-backend-craft-amd64
+      - build-backend-craft-arm64
+    if: needs.determine-builds.outputs.build-backend-craft == 'true'
+    runs-on:
+      - runs-on
+      - runner=2cpu-linux-x64
+      - run-id=${{ github.run_id }}-merge-backend-craft
+      - extras=ecr-cache
+    timeout-minutes: 90
+    environment: release
+    env:
+      REGISTRY_IMAGE: onyxdotapp/onyx-backend
+    steps:
+      - uses: runs-on/action@cd2b598b0515d39d78c38a02d529db87d2196d1e # ratchet:runs-on/action@v2
+
+      - name: Configure AWS credentials
+        uses: aws-actions/configure-aws-credentials@61815dcd50bd041e203e49132bacad1fd04d2708
+        with:
+          role-to-assume: ${{ secrets.AWS_OIDC_ROLE_ARN }}
+          aws-region: us-east-2
+
+      - name: Get AWS Secrets
+        uses: aws-actions/aws-secretsmanager-get-secrets@a9a7eb4e2f2871d30dc5b892576fde60a2ecc802
+        with:
+          secret-ids: |
+            DOCKER_USERNAME, deploy/docker-username
+            DOCKER_TOKEN, deploy/docker-token
+          parse-json-secrets: true
+
+      - name: Set up Docker Buildx
+        uses: docker/setup-buildx-action@8d2750c68a42422c14e847fe6c8ac0403b4cbd6f # ratchet:docker/setup-buildx-action@v3
+
+      - name: Login to Docker Hub
+        uses: docker/login-action@5e57cd118135c172c3672efd75eb46360885c0ef # ratchet:docker/login-action@v3
+        with:
+          username: ${{ env.DOCKER_USERNAME }}
+          password: ${{ env.DOCKER_TOKEN }}
+
+      - name: Docker meta
+        id: meta
+        uses: docker/metadata-action@c299e40c65443455700f0fdfc63efafe5b349051 # ratchet:docker/metadata-action@v5
+        with:
+          images: ${{ env.REGISTRY_IMAGE }}
+          flavor: |
+            latest=false
+          tags: |
+            type=raw,value=craft-latest
+            # TODO: Consider aligning craft-latest tags with regular backend builds (e.g., latest, edge, beta)
+            # to keep tagging strategy consistent across all backend images
+
+      - name: Create and push manifest
+        env:
+          IMAGE_REPO: ${{ env.REGISTRY_IMAGE }}
+          AMD64_DIGEST: ${{ needs.build-backend-craft-amd64.outputs.digest }}
+          ARM64_DIGEST: ${{ needs.build-backend-craft-arm64.outputs.digest }}
+          META_TAGS: ${{ steps.meta.outputs.tags }}
+        run: |
+          IMAGES="${IMAGE_REPO}@${AMD64_DIGEST} ${IMAGE_REPO}@${ARM64_DIGEST}"
+          docker buildx imagetools create \
+            $(printf '%s\n' "${META_TAGS}" | xargs -I {} echo -t {}) \
+            $IMAGES
+
   build-model-server-amd64:
     needs: determine-builds
     if: needs.determine-builds.outputs.build-model-server == 'true'
@@ -1022,7 +1235,7 @@ jobs:
       - uses: runs-on/action@cd2b598b0515d39d78c38a02d529db87d2196d1e # ratchet:runs-on/action@v2
 
       - name: Checkout code
-        uses: actions/checkout@8e8c483db84b4bee98b60c0593521ed34d9990e8 # ratchet:actions/checkout@v6
+        uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # ratchet:actions/checkout@v6
         with:
           persist-credentials: false
 
@@ -1101,7 +1314,7 @@ jobs:
       - uses: runs-on/action@cd2b598b0515d39d78c38a02d529db87d2196d1e # ratchet:runs-on/action@v2
 
       - name: Checkout code
-        uses: actions/checkout@8e8c483db84b4bee98b60c0593521ed34d9990e8 # ratchet:actions/checkout@v6
+        uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # ratchet:actions/checkout@v6
         with:
           persist-credentials: false
 
@@ -1354,7 +1567,7 @@ jobs:
       - uses: runs-on/action@cd2b598b0515d39d78c38a02d529db87d2196d1e # ratchet:runs-on/action@v2
 
       - name: Checkout
-        uses: actions/checkout@8e8c483db84b4bee98b60c0593521ed34d9990e8 # ratchet:actions/checkout@v6
+        uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # ratchet:actions/checkout@v6
         with:
           persist-credentials: false
 
@@ -1466,33 +1679,23 @@ jobs:
       - build-backend-amd64
       - build-backend-arm64
       - merge-backend
+      - build-backend-craft-amd64
+      - build-backend-craft-arm64
+      - merge-backend-craft
       - build-model-server-amd64
       - build-model-server-arm64
       - merge-model-server
-    if: always() && (needs.build-desktop.result == 'failure' || needs.build-web-amd64.result == 'failure' || needs.build-web-arm64.result == 'failure' || needs.merge-web.result == 'failure' || needs.build-web-cloud-amd64.result == 'failure' || needs.build-web-cloud-arm64.result == 'failure' || needs.merge-web-cloud.result == 'failure' || needs.build-backend-amd64.result == 'failure' || needs.build-backend-arm64.result == 'failure' || needs.merge-backend.result == 'failure' || needs.build-model-server-amd64.result == 'failure' || needs.build-model-server-arm64.result == 'failure' || needs.merge-model-server.result == 'failure') && needs.determine-builds.outputs.is-test-run != 'true'
+    if: always() && (needs.build-desktop.result == 'failure' || needs.build-web-amd64.result == 'failure' || needs.build-web-arm64.result == 'failure' || needs.merge-web.result == 'failure' || needs.build-web-cloud-amd64.result == 'failure' || needs.build-web-cloud-arm64.result == 'failure' || needs.merge-web-cloud.result == 'failure' || needs.build-backend-amd64.result == 'failure' || needs.build-backend-arm64.result == 'failure' || needs.merge-backend.result == 'failure' || (needs.determine-builds.outputs.build-backend-craft == 'true' && (needs.build-backend-craft-amd64.result == 'failure' || needs.build-backend-craft-arm64.result == 'failure' || needs.merge-backend-craft.result == 'failure')) || needs.build-model-server-amd64.result == 'failure' || needs.build-model-server-arm64.result == 'failure' || needs.merge-model-server.result == 'failure') && needs.determine-builds.outputs.is-test-run != 'true'
     # NOTE: Github-hosted runners have about 20s faster queue times and are preferred here.
     runs-on: ubuntu-slim
     timeout-minutes: 90
     environment: release
     steps:
       - name: Checkout
-        uses: actions/checkout@8e8c483db84b4bee98b60c0593521ed34d9990e8 # ratchet:actions/checkout@v6
+        uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # ratchet:actions/checkout@v6
         with:
           persist-credentials: false
 
-      - name: Configure AWS credentials
-        uses: aws-actions/configure-aws-credentials@61815dcd50bd041e203e49132bacad1fd04d2708
-        with:
-          role-to-assume: ${{ secrets.AWS_OIDC_ROLE_ARN }}
-          aws-region: us-east-2
-
-      - name: Get AWS Secrets
-        uses: aws-actions/aws-secretsmanager-get-secrets@a9a7eb4e2f2871d30dc5b892576fde60a2ecc802
-        with:
-          secret-ids: |
-            MONITOR_DEPLOYMENTS_WEBHOOK, deploy/monitor-deployments-webhook
-          parse-json-secrets: true
-
       - name: Determine failed jobs
         id: failed-jobs
         shell: bash
@@ -1558,7 +1761,7 @@ jobs:
       - name: Send Slack notification
         uses: ./.github/actions/slack-notify
         with:
-          webhook-url: ${{ env.MONITOR_DEPLOYMENTS_WEBHOOK }}
+          webhook-url: ${{ secrets.MONITOR_DEPLOYMENTS_WEBHOOK }}
           failed-jobs: ${{ steps.failed-jobs.outputs.jobs }}
           title: "🚨 Deployment Workflow Failed"
           ref-name: ${{ github.ref_name }}

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

@@ -15,7 +15,7 @@ jobs:
     timeout-minutes: 45
     steps:
       - name: Checkout
-        uses: actions/checkout@8e8c483db84b4bee98b60c0593521ed34d9990e8 # ratchet:actions/checkout@v6
+        uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # ratchet:actions/checkout@v6
         with:
           fetch-depth: 0
           persist-credentials: false

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

@@ -28,7 +28,7 @@ jobs:
 
     steps:
       - name: Checkout code
-        uses: actions/checkout@8e8c483db84b4bee98b60c0593521ed34d9990e8 # ratchet:actions/checkout@v6
+        uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # ratchet:actions/checkout@v6
         with:
           persist-credentials: false
 

.github/workflows/pr-beta-cherrypick-check.yml

@@ -0,0 +1,28 @@
+name: Require beta cherry-pick consideration
+concurrency:
+  group: Require-Beta-Cherrypick-Consideration-${{ github.workflow }}-${{ 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:
+  beta-cherrypick-check:
+    runs-on: ubuntu-latest
+    timeout-minutes: 45
+    steps:
+      - name: Check PR body for beta cherry-pick consideration
+        env:
+          PR_BODY: ${{ github.event.pull_request.body }}
+        run: |
+          if echo "$PR_BODY" | grep -qiE "\\[x\\][[:space:]]*\\[Required\\][[:space:]]*I have considered whether this PR needs to be cherry[- ]picked to the latest beta branch"; then
+            echo "Cherry-pick consideration box is checked. Check passed."
+            exit 0
+          fi
+
+          echo "::error::Please check the 'I have considered whether this PR needs to be cherry-picked to the latest beta branch' box in the PR description."
+          exit 1

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

@@ -27,7 +27,7 @@ jobs:
       - uses: runs-on/action@cd2b598b0515d39d78c38a02d529db87d2196d1e # ratchet:runs-on/action@v2
 
       - name: Checkout code
-        uses: actions/checkout@8e8c483db84b4bee98b60c0593521ed34d9990e8 # ratchet:actions/checkout@v6
+        uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # ratchet:actions/checkout@v6
         with:
           persist-credentials: false
 

.github/workflows/pr-desktop-build.yml

@@ -0,0 +1,114 @@
+name: Build Desktop App
+concurrency:
+  group: Build-Desktop-App-${{ github.workflow }}-${{ github.head_ref || github.event.workflow_run.head_branch || github.run_id }}
+  cancel-in-progress: true
+
+on:
+  merge_group:
+  pull_request:
+    branches:
+      - main
+      - "release/**"
+    paths:
+      - "desktop/**"
+      - ".github/workflows/pr-desktop-build.yml"
+  push:
+    tags:
+      - "v*.*.*"
+
+permissions:
+  contents: read
+
+jobs:
+  build-desktop:
+    name: Build Desktop (${{ matrix.platform }})
+    runs-on: ${{ matrix.os }}
+    timeout-minutes: 60
+    strategy:
+      fail-fast: false
+      matrix:
+        include:
+          - platform: linux
+            os: ubuntu-latest
+            target: x86_64-unknown-linux-gnu
+            args: "--bundles deb,rpm"
+          # TODO: Fix and enable the macOS build.
+          #- platform: macos
+          #  os: macos-latest
+          #  target: universal-apple-darwin
+          #  args: "--target universal-apple-darwin"
+          # TODO: Fix and enable the Windows build.
+          #- platform: windows
+          #  os: windows-latest
+          #  target: x86_64-pc-windows-msvc
+          #  args: ""
+
+    steps:
+      - name: Checkout code
+        uses: actions/checkout@8e8c483db84b4bee98b60c0593521ed34d9990e8
+        with:
+          persist-credentials: false
+
+      - name: Setup node
+        uses: actions/setup-node@49933ea5288caeca8642d1e84afbd3f7d6820020
+        with:
+          node-version: 24
+          cache: "npm" # zizmor: ignore[cache-poisoning]
+          cache-dependency-path: ./desktop/package-lock.json
+
+      - name: Setup Rust
+        uses: dtolnay/rust-toolchain@4be9e76fd7c4901c61fb841f559994984270fce7
+        with:
+          toolchain: stable
+          targets: ${{ matrix.target }}
+
+      - name: Cache Cargo registry and build
+        uses: actions/cache@0057852bfaa89a56745cba8c7296529d2fc39830 # zizmor: ignore[cache-poisoning]
+        with:
+          path: |
+            ~/.cargo/bin/
+            ~/.cargo/registry/index/
+            ~/.cargo/registry/cache/
+            ~/.cargo/git/db/
+            desktop/src-tauri/target/
+          key: ${{ runner.os }}-cargo-${{ hashFiles('desktop/src-tauri/Cargo.lock') }}
+          restore-keys: |
+            ${{ runner.os }}-cargo-
+
+      - name: Install Linux dependencies
+        if: matrix.platform == 'linux'
+        run: |
+          sudo apt-get update
+          sudo apt-get install -y \
+            build-essential \
+            libglib2.0-dev \
+            libgirepository1.0-dev \
+            libgtk-3-dev \
+            libjavascriptcoregtk-4.1-dev \
+            libwebkit2gtk-4.1-dev \
+            libayatana-appindicator3-dev \
+            gobject-introspection \
+            pkg-config \
+            curl \
+            xdg-utils
+
+      - name: Install npm dependencies
+        working-directory: ./desktop
+        run: npm ci
+
+      - name: Build desktop app
+        working-directory: ./desktop
+        run: npx tauri build ${{ matrix.args }}
+        env:
+          TAURI_SIGNING_PRIVATE_KEY: ""
+          TAURI_SIGNING_PRIVATE_KEY_PASSWORD: ""
+
+      - name: Upload build artifacts
+        if: always()
+        uses: actions/upload-artifact@ea165f8d65b6e75b540449e92b4886f43607fa02
+        with:
+          name: desktop-build-${{ matrix.platform }}-${{ github.run_id }}
+          path: |
+            desktop/src-tauri/target/release/bundle/
+          retention-days: 7
+          if-no-files-found: ignore

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

@@ -57,7 +57,7 @@ jobs:
       test-dirs: ${{ steps.set-matrix.outputs.test-dirs }}
     steps:
       - name: Checkout code
-        uses: actions/checkout@8e8c483db84b4bee98b60c0593521ed34d9990e8 # ratchet:actions/checkout@v6
+        uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # ratchet:actions/checkout@v6
         with:
           persist-credentials: false
 
@@ -91,7 +91,7 @@ jobs:
       - uses: runs-on/action@cd2b598b0515d39d78c38a02d529db87d2196d1e # ratchet:runs-on/action@v2
 
       - name: Checkout code
-        uses: actions/checkout@8e8c483db84b4bee98b60c0593521ed34d9990e8 # ratchet:actions/checkout@v6
+        uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # ratchet:actions/checkout@v6
         with:
           persist-credentials: false
 

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

@@ -30,7 +30,7 @@ jobs:
     # fetch-depth 0 is required for helm/chart-testing-action
     steps:
       - name: Checkout code
-        uses: actions/checkout@8e8c483db84b4bee98b60c0593521ed34d9990e8 # ratchet:actions/checkout@v6
+        uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # ratchet:actions/checkout@v6
         with:
           fetch-depth: 0
           persist-credentials: false
@@ -197,7 +197,6 @@ jobs:
               --set=auth.opensearch.enabled=true \
               --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 \

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

@@ -48,7 +48,7 @@ jobs:
       test-dirs: ${{ steps.set-matrix.outputs.test-dirs }}
     steps:
       - name: Checkout code
-        uses: actions/checkout@8e8c483db84b4bee98b60c0593521ed34d9990e8 # ratchet:actions/checkout@v6
+        uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # ratchet:actions/checkout@v6
         with:
           persist-credentials: false
 
@@ -84,7 +84,7 @@ jobs:
     steps:
       - uses: runs-on/action@cd2b598b0515d39d78c38a02d529db87d2196d1e # ratchet:runs-on/action@v2
       - name: Checkout code
-        uses: actions/checkout@8e8c483db84b4bee98b60c0593521ed34d9990e8 # ratchet:actions/checkout@v6
+        uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # ratchet:actions/checkout@v6
         with:
           persist-credentials: false
 
@@ -144,7 +144,7 @@ jobs:
     steps:
       - uses: runs-on/action@cd2b598b0515d39d78c38a02d529db87d2196d1e # ratchet:runs-on/action@v2
       - name: Checkout code
-        uses: actions/checkout@8e8c483db84b4bee98b60c0593521ed34d9990e8 # ratchet:actions/checkout@v6
+        uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # ratchet:actions/checkout@v6
         with:
           persist-credentials: false
 
@@ -203,7 +203,7 @@ jobs:
     steps:
       - uses: runs-on/action@cd2b598b0515d39d78c38a02d529db87d2196d1e # ratchet:runs-on/action@v2
       - name: Checkout code
-        uses: actions/checkout@8e8c483db84b4bee98b60c0593521ed34d9990e8 # ratchet:actions/checkout@v6
+        uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # ratchet:actions/checkout@v6
         with:
           persist-credentials: false
 
@@ -279,7 +279,7 @@ jobs:
     steps:
       - uses: runs-on/action@cd2b598b0515d39d78c38a02d529db87d2196d1e # ratchet:runs-on/action@v2
       - name: Checkout code
-        uses: actions/checkout@8e8c483db84b4bee98b60c0593521ed34d9990e8 # ratchet:actions/checkout@v6
+        uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # ratchet:actions/checkout@v6
         with:
           persist-credentials: false
 
@@ -460,7 +460,7 @@ jobs:
     steps:
       - uses: runs-on/action@cd2b598b0515d39d78c38a02d529db87d2196d1e # ratchet:runs-on/action@v2
       - name: Checkout code
-        uses: actions/checkout@8e8c483db84b4bee98b60c0593521ed34d9990e8 # ratchet:actions/checkout@v6
+        uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # ratchet:actions/checkout@v6
         with:
           persist-credentials: false
 

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

@@ -23,7 +23,7 @@ jobs:
     timeout-minutes: 45
     steps:
       - name: Checkout code
-        uses: actions/checkout@8e8c483db84b4bee98b60c0593521ed34d9990e8 # ratchet:actions/checkout@v6
+        uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # ratchet:actions/checkout@v6
         with:
           persist-credentials: false
 

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

@@ -40,7 +40,7 @@ jobs:
       test-dirs: ${{ steps.set-matrix.outputs.test-dirs }}
     steps:
       - name: Checkout code
-        uses: actions/checkout@8e8c483db84b4bee98b60c0593521ed34d9990e8 # ratchet:actions/checkout@v6
+        uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # ratchet:actions/checkout@v6
         with:
           persist-credentials: false
 
@@ -76,7 +76,7 @@ jobs:
     steps:
       - uses: runs-on/action@cd2b598b0515d39d78c38a02d529db87d2196d1e # ratchet:runs-on/action@v2
       - name: Checkout code
-        uses: actions/checkout@8e8c483db84b4bee98b60c0593521ed34d9990e8 # ratchet:actions/checkout@v6
+        uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # ratchet:actions/checkout@v6
         with:
           persist-credentials: false
 
@@ -136,7 +136,7 @@ jobs:
     steps:
       - uses: runs-on/action@cd2b598b0515d39d78c38a02d529db87d2196d1e # ratchet:runs-on/action@v2
       - name: Checkout code
-        uses: actions/checkout@8e8c483db84b4bee98b60c0593521ed34d9990e8 # ratchet:actions/checkout@v6
+        uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # ratchet:actions/checkout@v6
         with:
           persist-credentials: false
 
@@ -195,7 +195,7 @@ jobs:
     steps:
       - uses: runs-on/action@cd2b598b0515d39d78c38a02d529db87d2196d1e # ratchet:runs-on/action@v2
       - name: Checkout code
-        uses: actions/checkout@8e8c483db84b4bee98b60c0593521ed34d9990e8 # ratchet:actions/checkout@v6
+        uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # ratchet:actions/checkout@v6
         with:
           persist-credentials: false
 
@@ -271,7 +271,7 @@ jobs:
     steps:
       - uses: runs-on/action@cd2b598b0515d39d78c38a02d529db87d2196d1e # ratchet:runs-on/action@v2
       - name: Checkout code
-        uses: actions/checkout@8e8c483db84b4bee98b60c0593521ed34d9990e8 # ratchet:actions/checkout@v6
+        uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # ratchet:actions/checkout@v6
         with:
           persist-credentials: false
 

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

@@ -66,7 +66,7 @@ jobs:
       - uses: runs-on/action@cd2b598b0515d39d78c38a02d529db87d2196d1e # ratchet:runs-on/action@v2
 
       - name: Checkout code
-        uses: actions/checkout@8e8c483db84b4bee98b60c0593521ed34d9990e8 # ratchet:actions/checkout@v6
+        uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # ratchet:actions/checkout@v6
         with:
           persist-credentials: false
 
@@ -127,7 +127,7 @@ jobs:
       - uses: runs-on/action@cd2b598b0515d39d78c38a02d529db87d2196d1e # ratchet:runs-on/action@v2
 
       - name: Checkout code
-        uses: actions/checkout@8e8c483db84b4bee98b60c0593521ed34d9990e8 # ratchet:actions/checkout@v6
+        uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # ratchet:actions/checkout@v6
         with:
           persist-credentials: false
 
@@ -188,7 +188,7 @@ jobs:
       - uses: runs-on/action@cd2b598b0515d39d78c38a02d529db87d2196d1e # ratchet:runs-on/action@v2
 
       - name: Checkout code
-        uses: actions/checkout@8e8c483db84b4bee98b60c0593521ed34d9990e8 # ratchet:actions/checkout@v6
+        uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # ratchet:actions/checkout@v6
         with:
           persist-credentials: false
 
@@ -249,12 +249,12 @@ jobs:
     strategy:
       fail-fast: false
       matrix:
-        project: [admin, no-auth, exclusive]
+        project: [admin, exclusive]
     steps:
       - uses: runs-on/action@cd2b598b0515d39d78c38a02d529db87d2196d1e # ratchet:runs-on/action@v2
 
       - name: Checkout code
-        uses: actions/checkout@8e8c483db84b4bee98b60c0593521ed34d9990e8 # ratchet:actions/checkout@v6
+        uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # ratchet:actions/checkout@v6
         with:
           persist-credentials: false
 
@@ -299,9 +299,6 @@ jobs:
           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
@@ -430,9 +427,6 @@ jobs:
         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@b7c566a772e6b6bfb58ed0dc250532a479d7789f
@@ -488,7 +482,7 @@ jobs:
 #     ]
 #   steps:
 #     - name: Checkout code
-#       uses: actions/checkout@8e8c483db84b4bee98b60c0593521ed34d9990e8 # ratchet:actions/checkout@v6
+#       uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # ratchet:actions/checkout@v6
 #       with:
 #         fetch-depth: 0
 

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

@@ -27,7 +27,7 @@ jobs:
     steps:
       - uses: runs-on/action@cd2b598b0515d39d78c38a02d529db87d2196d1e # ratchet:runs-on/action@v2
       - name: Checkout code
-        uses: actions/checkout@8e8c483db84b4bee98b60c0593521ed34d9990e8 # ratchet:actions/checkout@v6
+        uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # ratchet:actions/checkout@v6
         with:
           persist-credentials: false
 
@@ -50,8 +50,9 @@ jobs:
         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') }}
+          key: mypy-${{ runner.os }}-${{ github.base_ref || github.event.merge_group.base_ref || 'main' }}-${{ hashFiles('**/*.py', '**/*.pyi', 'backend/pyproject.toml') }}
           restore-keys: |
+            mypy-${{ runner.os }}-${{ github.base_ref || github.event.merge_group.base_ref || 'main' }}-
             mypy-${{ runner.os }}-
 
       - name: Run MyPy

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

@@ -65,7 +65,7 @@ env:
   ZENDESK_TOKEN: ${{ secrets.ZENDESK_TOKEN }}
 
   # Salesforce
-  SF_USERNAME: ${{ secrets.SF_USERNAME }}
+  SF_USERNAME: ${{ vars.SF_USERNAME }}
   SF_PASSWORD: ${{ secrets.SF_PASSWORD }}
   SF_SECURITY_TOKEN: ${{ secrets.SF_SECURITY_TOKEN }}
 
@@ -110,6 +110,9 @@ env:
   # Slack
   SLACK_BOT_TOKEN: ${{ secrets.SLACK_BOT_TOKEN }}
 
+  # Discord
+  DISCORD_CONNECTOR_BOT_TOKEN: ${{ secrets.DISCORD_CONNECTOR_BOT_TOKEN }}
+
   # Teams
   TEAMS_APPLICATION_ID: ${{ secrets.TEAMS_APPLICATION_ID }}
   TEAMS_DIRECTORY_ID: ${{ secrets.TEAMS_DIRECTORY_ID }}
@@ -139,7 +142,7 @@ jobs:
       - uses: runs-on/action@cd2b598b0515d39d78c38a02d529db87d2196d1e # ratchet:runs-on/action@v2
 
       - name: Checkout code
-        uses: actions/checkout@8e8c483db84b4bee98b60c0593521ed34d9990e8 # ratchet:actions/checkout@v6
+        uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # ratchet:actions/checkout@v6
         with:
           persist-credentials: false
 

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

@@ -38,7 +38,7 @@ jobs:
 
     steps:
       - name: Checkout code
-        uses: actions/checkout@8e8c483db84b4bee98b60c0593521ed34d9990e8 # ratchet:actions/checkout@v6
+        uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # ratchet:actions/checkout@v6
         with:
           persist-credentials: false
 

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

@@ -32,7 +32,7 @@ jobs:
     - uses: runs-on/action@cd2b598b0515d39d78c38a02d529db87d2196d1e # ratchet:runs-on/action@v2
 
     - name: Checkout code
-      uses: actions/checkout@8e8c483db84b4bee98b60c0593521ed34d9990e8 # ratchet:actions/checkout@v6
+      uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # ratchet:actions/checkout@v6
       with:
         persist-credentials: false
 

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

@@ -20,7 +20,7 @@ jobs:
     runs-on: ubuntu-latest
     timeout-minutes: 45
     steps:
-      - uses: actions/checkout@8e8c483db84b4bee98b60c0593521ed34d9990e8 # ratchet:actions/checkout@v6
+      - uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # ratchet:actions/checkout@v6
         with:
           fetch-depth: 0
           persist-credentials: false
@@ -38,7 +38,7 @@ jobs:
       - name: Install node dependencies
         working-directory: ./web
         run: npm ci
-      - uses: j178/prek-action@91fd7d7cf70ae1dee9f4f44e7dfa5d1073fe6623 # ratchet:j178/prek-action@v1
+      - uses: j178/prek-action@9d6a3097e0c1865ecce00cfb89fe80f2ee91b547 # ratchet:j178/prek-action@v1
         with:
           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' || '' }}

.github/workflows/release-devtools.yml

@@ -24,11 +24,11 @@ jobs:
           - { goos: "darwin", goarch: "arm64" }
           - { goos: "", goarch: "" }
     steps:
-      - uses: actions/checkout@8e8c483db84b4bee98b60c0593521ed34d9990e8 # ratchet:actions/checkout@v6
+      - uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # ratchet:actions/checkout@v6
         with:
           persist-credentials: false
           fetch-depth: 0
-      - uses: astral-sh/setup-uv@ed21f2f24f8dd64503750218de024bcf64c7250a # ratchet:astral-sh/setup-uv@v7
+      - uses: astral-sh/setup-uv@61cb8a9741eeb8a550a1b8544337180c0fc8476b # ratchet:astral-sh/setup-uv@v7
         with:
           enable-cache: false
           version: "0.9.9"

.github/workflows/sync_foss.yml

@@ -14,7 +14,7 @@ jobs:
       contents: read
     steps:
       - name: Checkout main Onyx repo
-        uses: actions/checkout@8e8c483db84b4bee98b60c0593521ed34d9990e8 # ratchet:actions/checkout@v6
+        uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # ratchet:actions/checkout@v6
         with:
           fetch-depth: 0
           persist-credentials: false

.github/workflows/tag-nightly.yml

@@ -18,7 +18,7 @@ jobs:
       # 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@8e8c483db84b4bee98b60c0593521ed34d9990e8 # ratchet:actions/checkout@v6
+        uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # ratchet:actions/checkout@v6
         with:
           ssh-key: "${{ secrets.DEPLOY_KEY }}"
           persist-credentials: true

.github/workflows/zizmor.yml

@@ -17,7 +17,7 @@ jobs:
       security-events: write # needed for SARIF uploads
     steps:
       - name: Checkout repository
-        uses: actions/checkout@8e8c483db84b4bee98b60c0593521ed34d9990e8 # ratchet:actions/checkout@v6.0.1
+        uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # ratchet:actions/checkout@v6.0.2
         with:
           persist-credentials: false
 
@@ -31,7 +31,7 @@ jobs:
 
       - 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
+        uses: astral-sh/setup-uv@61cb8a9741eeb8a550a1b8544337180c0fc8476b # ratchet:astral-sh/setup-uv@v7
         with:
           enable-cache: false
           version: "0.9.9"

.gitignore

@@ -1,6 +1,7 @@
 # editors
-.vscode
+.vscode/*
 !/.vscode/env_template.txt
+!/.vscode/env.web_template.txt
 !/.vscode/launch.json
 !/.vscode/tasks.template.jsonc
 .zed
@@ -39,10 +40,6 @@ settings.json
 /backend/tests/regression/answer_quality/search_test_config.yaml
 *.egg-info
 
-# Claude
-AGENTS.md
-CLAUDE.md
-
 # Local .terraform directories
 **/.terraform/*
 

.pre-commit-config.yaml

@@ -66,7 +66,8 @@ repos:
       - id: uv-run
         name: Check lazy imports
         args: ["--active", "--with=onyx-devtools", "ods", "check-lazy-imports"]
-        files: ^backend/(?!\.venv/).*\.py$
+        pass_filenames: true
+        files: ^backend/(?!\.venv/|scripts/).*\.py$
       # NOTE: This takes ~6s on a single, large module which is prohibitively slow.
       # - id: uv-run
       #   name: mypy

.vscode/env.web_template.txt

@@ -0,0 +1,16 @@
+# Copy this file to .env.web in the .vscode folder.
+# Fill in the <REPLACE THIS> values as needed
+# Web Server specific environment variables
+# Minimal set needed for Next.js dev server
+
+# Auth
+AUTH_TYPE=basic
+DEV_MODE=true
+
+# 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
+
+# Enable Onyx Craft
+ENABLE_CRAFT=true

.vscode/env_template.txt

@@ -6,13 +6,13 @@
 # processes.
 
 
-# For local dev, often user Authentication is not needed.
-AUTH_TYPE=disabled
+AUTH_TYPE=basic
+DEV_MODE=true
 
 
 # Always keep these on for Dev.
 # Logs model prompts, reasoning, and answer to stdout.
-LOG_ONYX_MODEL_INTERACTIONS=True
+LOG_ONYX_MODEL_INTERACTIONS=False
 # More verbose logging
 LOG_LEVEL=debug
 
@@ -35,7 +35,6 @@ 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.
 GEN_AI_MODEL_VERSION=gpt-4o
-FAST_GEN_AI_MODEL_VERSION=gpt-4o
 
 
 # Python stuff

.vscode/launch.json

@@ -25,6 +25,7 @@
         "Celery heavy",
         "Celery docfetching",
         "Celery docprocessing",
+        "Celery user_file_processing",
         "Celery beat"
       ],
       "presentation": {
@@ -86,7 +87,7 @@
       "request": "launch",
       "cwd": "${workspaceRoot}/web",
       "runtimeExecutable": "npm",
-      "envFile": "${workspaceFolder}/.vscode/.env",
+      "envFile": "${workspaceFolder}/.vscode/.env.web",
       "runtimeArgs": ["run", "dev"],
       "presentation": {
         "group": "2"
@@ -121,7 +122,6 @@
       "cwd": "${workspaceFolder}/backend",
       "envFile": "${workspaceFolder}/.vscode/.env",
       "env": {
-        "LOG_ONYX_MODEL_INTERACTIONS": "True",
         "LOG_LEVEL": "DEBUG",
         "PYTHONUNBUFFERED": "1"
       },
@@ -149,6 +149,24 @@
       },
       "consoleTitle": "Slack Bot Console"
     },
+    {
+      "name": "Discord Bot",
+      "consoleName": "Discord Bot",
+      "type": "debugpy",
+      "request": "launch",
+      "program": "onyx/onyxbot/discord/client.py",
+      "cwd": "${workspaceFolder}/backend",
+      "envFile": "${workspaceFolder}/.vscode/.env",
+      "env": {
+        "LOG_LEVEL": "DEBUG",
+        "PYTHONUNBUFFERED": "1",
+        "PYTHONPATH": "."
+      },
+      "presentation": {
+        "group": "2"
+      },
+      "consoleTitle": "Discord Bot Console"
+    },
     {
       "name": "MCP Server",
       "consoleName": "MCP Server",
@@ -397,7 +415,6 @@
         "onyx.background.celery.versioned_apps.docfetching",
         "worker",
         "--pool=threads",
-        "--concurrency=1",
         "--prefetch-multiplier=1",
         "--loglevel=INFO",
         "--hostname=docfetching@%n",
@@ -428,7 +445,6 @@
         "onyx.background.celery.versioned_apps.docprocessing",
         "worker",
         "--pool=threads",
-        "--concurrency=6",
         "--prefetch-multiplier=1",
         "--loglevel=INFO",
         "--hostname=docprocessing@%n",
@@ -556,7 +572,6 @@
       "cwd": "${workspaceFolder}/backend",
       "envFile": "${workspaceFolder}/.vscode/.env",
       "env": {
-        "LOG_ONYX_MODEL_INTERACTIONS": "True",
         "LOG_LEVEL": "DEBUG",
         "PYTHONUNBUFFERED": "1",
         "PYTHONPATH": "."
@@ -577,6 +592,23 @@
         "group": "3"
       }
     },
+    {
+      "name": "Build Sandbox Templates",
+      "type": "debugpy",
+      "request": "launch",
+      "module": "onyx.server.features.build.sandbox.build_templates",
+      "cwd": "${workspaceFolder}/backend",
+      "envFile": "${workspaceFolder}/.vscode/.env",
+      "env": {
+        "PYTHONUNBUFFERED": "1",
+        "PYTHONPATH": "."
+      },
+      "console": "integratedTerminal",
+      "presentation": {
+        "group": "3"
+      },
+      "consoleTitle": "Build Sandbox Templates"
+    },
     {
       // Dummy entry used to label the group
       "name": "--- Database ---",
@@ -587,6 +619,27 @@
         "order": 0
       }
     },
+    {
+      "name": "Restore seeded database dump",
+      "type": "node",
+      "request": "launch",
+      "runtimeExecutable": "uv",
+      "runtimeArgs": [
+        "run",
+        "--with",
+        "onyx-devtools",
+        "ods",
+        "db",
+        "restore",
+        "--fetch-seeded",
+        "--yes"
+      ],
+      "cwd": "${workspaceFolder}",
+      "console": "integratedTerminal",
+      "presentation": {
+        "group": "4"
+      }
+    },
     {
       "name": "Clean restore seeded database dump (destructive)",
       "type": "node",

AGENTS.md

@@ -1,26 +1,25 @@
-# CLAUDE.md
+# PROJECT KNOWLEDGE BASE
 
-This file provides guidance to Claude Code (claude.ai/code) 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 .venv/bin/activate` \
-to assume the python venv.
+  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`.
+  `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.
+  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.
+  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:
@@ -92,6 +91,7 @@ Onyx uses Celery for asynchronous task processing with multiple specialized work
 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)
@@ -105,12 +105,14 @@ Onyx supports two deployment modes for background workers, controlled by the `US
 - 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
@@ -119,18 +121,18 @@ The deployment mode affects:
 
 - **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.
+  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**: 
+**Defining Tasks**:
+
 - Always use `@shared_task` rather than `@celery_app`
 - Put tasks under `background/celery/tasks/` or `ee/background/celery/tasks`
 
@@ -143,6 +145,7 @@ If you make any updates to a celery worker and you want to test these changes, y
 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
@@ -154,6 +157,7 @@ NOTE: Always make sure everything is strictly typed (both in Python and Typescri
 ## 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
@@ -435,6 +439,7 @@ function ContactForm() {
 **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`
@@ -467,6 +472,7 @@ function ContactForm() {
 ## Database & Migrations
 
 ### Running Migrations
+
 ```bash
 # Standard migrations
 alembic upgrade head
@@ -476,6 +482,7 @@ alembic -n schema_private upgrade head
 ```
 
 ### Creating Migrations
+
 ```bash
 # Create migration
 alembic revision -m "description"
@@ -488,13 +495,14 @@ Write the migration manually and place it in the file that alembic creates when
 
 ## Testing Strategy
 
-First, you must activate the virtual environment with `source .venv/bin/activate`. 
+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 
+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:
@@ -504,13 +512,14 @@ 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, 
+
+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. 
+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, 
+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).
 
@@ -523,15 +532,16 @@ python -m dotenv -f .vscode/.env run -- pytest backend/tests/external_dependency
 ```
 
 ### 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 
+
+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 
+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 
+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`.
@@ -543,8 +553,9 @@ 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.
+
+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.
 
@@ -556,13 +567,11 @@ To run them:
 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. 
-
+will be tailing their logs to this file.
 
 ## Security Considerations
 
@@ -581,6 +590,7 @@ will be tailing their logs to this file.
 - 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**
@@ -593,10 +603,10 @@ Things you come across in your research that are important to the implementation
 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 
+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*
+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.

AGENTS.md.template

@@ -1,599 +0,0 @@
-# AGENTS.md
-
-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 .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:
-  - 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
-- **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 `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
-```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
-
-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
-
-## 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

@@ -0,0 +1 @@
+AGENTS.md

backend/.dockerignore

@@ -16,3 +16,8 @@ dist/
 .coverage
 htmlcov/
 model_server/legacy/
+
+# Craft: demo_data directory should be unzipped at container startup, not copied
+**/demo_data/
+# Craft: templates/outputs/venv is created at container startup
+**/templates/outputs/venv

backend/Dockerfile

@@ -7,6 +7,10 @@ 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"
 
+# Build argument for Craft support (disabled by default)
+# Use --build-arg ENABLE_CRAFT=true to include Node.js and opencode CLI
+ARG ENABLE_CRAFT=false
+
 # DO_NOT_TRACK is used to disable telemetry for Unstructured
 ENV DANSWER_RUNNING_IN_DOCKER="true" \
     DO_NOT_TRACK="true" \
@@ -46,7 +50,23 @@ RUN apt-get update && \
     rm -rf /var/lib/apt/lists/* && \
     apt-get clean
 
+# Conditionally install Node.js 20 for Craft (required for Next.js)
+# Only installed when ENABLE_CRAFT=true
+RUN if [ "$ENABLE_CRAFT" = "true" ]; then \
+        echo "Installing Node.js 20 for Craft support..." && \
+        curl -fsSL https://deb.nodesource.com/setup_20.x | bash - && \
+        apt-get install -y nodejs && \
+        rm -rf /var/lib/apt/lists/*; \
+    fi
 
+# Conditionally install opencode CLI for Craft agent functionality
+# Only installed when ENABLE_CRAFT=true
+# TODO: download a specific, versioned release of the opencode CLI
+RUN if [ "$ENABLE_CRAFT" = "true" ]; then \
+        echo "Installing opencode CLI for Craft support..." && \
+        curl -fsSL https://opencode.ai/install | bash; \
+    fi
+ENV PATH="/root/.opencode/bin:${PATH}"
 
 # Install Python dependencies
 # Remove py which is pulled in by retry, py is not needed and is a CVE
@@ -89,6 +109,12 @@ RUN uv pip install --system --no-cache-dir --upgrade \
 RUN python -c "from tokenizers import Tokenizer; \
 Tokenizer.from_pretrained('nomic-ai/nomic-embed-text-v1')"
 
+# Pre-downloading NLTK for setups with limited egress
+RUN python -c "import nltk; \
+    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')"
@@ -113,7 +139,20 @@ COPY --chown=onyx:onyx ./static /app/static
 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
+COPY --chown=onyx:onyx ./scripts/setup_craft_templates.sh /app/scripts/setup_craft_templates.sh
+RUN chmod +x /app/scripts/supervisord_entrypoint.sh /app/scripts/setup_craft_templates.sh
+
+# Run Craft template setup at build time when ENABLE_CRAFT=true
+# This pre-bakes demo data, Python venv, and npm dependencies into the image
+RUN if [ "$ENABLE_CRAFT" = "true" ]; then \
+        echo "Running Craft template setup at build time..." && \
+        ENABLE_CRAFT=true /app/scripts/setup_craft_templates.sh; \
+    fi
+
+# Set Craft template paths to the in-image locations
+# These match the paths where setup_craft_templates.sh creates the templates
+ENV OUTPUTS_TEMPLATE_PATH=/app/onyx/server/features/build/sandbox/kubernetes/docker/templates/outputs
+ENV VENV_TEMPLATE_PATH=/app/onyx/server/features/build/sandbox/kubernetes/docker/templates/venv
 
 # Put logo in assets
 COPY --chown=onyx:onyx ./assets /app/assets

backend/Dockerfile.model_server

@@ -48,6 +48,7 @@ WORKDIR /app
 # Utils used by model server
 COPY ./onyx/utils/logger.py /app/onyx/utils/logger.py
 COPY ./onyx/utils/middleware.py /app/onyx/utils/middleware.py
+COPY ./onyx/utils/tenant.py /app/onyx/utils/tenant.py
 
 # Place to fetch version information
 COPY ./onyx/__init__.py /app/onyx/__init__.py

backend/alembic/env.py

@@ -57,7 +57,7 @@ if USE_IAM_AUTH:
 
 
 def include_object(
-    object: SchemaItem,
+    object: SchemaItem,  # noqa: ARG001
     name: str | None,
     type_: Literal[
         "schema",
@@ -67,8 +67,8 @@ def include_object(
         "unique_constraint",
         "foreign_key_constraint",
     ],
-    reflected: bool,
-    compare_to: SchemaItem | None,
+    reflected: bool,  # noqa: ARG001
+    compare_to: SchemaItem | None,  # noqa: ARG001
 ) -> bool:
     if type_ == "table" and name in EXCLUDE_TABLES:
         return False
@@ -244,7 +244,7 @@ def do_run_migrations(
 
 
 def provide_iam_token_for_alembic(
-    dialect: Any, conn_rec: Any, cargs: Any, cparams: Any
+    dialect: Any, conn_rec: Any, cargs: Any, cparams: Any  # noqa: ARG001
 ) -> None:
     if USE_IAM_AUTH:
         # Database connection settings

backend/alembic/versions/1f60f60c3401_embedding_model_search_settings.py

@@ -10,8 +10,6 @@ from alembic import op
 import sqlalchemy as sa
 from sqlalchemy.dialects import postgresql
 
-from onyx.configs.chat_configs import NUM_POSTPROCESSED_RESULTS
-
 # revision identifiers, used by Alembic.
 revision = "1f60f60c3401"
 down_revision = "f17bf3b0d9f1"
@@ -66,7 +64,7 @@ def upgrade() -> None:
             "num_rerank",
             sa.Integer(),
             nullable=False,
-            server_default=str(NUM_POSTPROCESSED_RESULTS),
+            server_default=str(20),
         ),
     )
 

backend/alembic/versions/2020d417ec84_single_onyx_craft_migration.py

@@ -0,0 +1,351 @@
+"""single onyx craft migration
+
+Consolidates all buildmode/onyx craft tables into a single migration.
+
+Tables created:
+- build_session: User build sessions with status tracking
+- sandbox: User-owned containerized environments (one per user)
+- artifact: Build output files (web apps, documents, images)
+- snapshot: Sandbox filesystem snapshots
+- build_message: Conversation messages for build sessions
+
+Existing table modified:
+- connector_credential_pair: Added processing_mode column
+
+Revision ID: 2020d417ec84
+Revises: 41fa44bef321
+Create Date: 2026-01-26 14:43:54.641405
+
+"""
+
+from alembic import op
+import sqlalchemy as sa
+from sqlalchemy.dialects import postgresql
+
+
+# revision identifiers, used by Alembic.
+revision = "2020d417ec84"
+down_revision = "41fa44bef321"
+branch_labels = None
+depends_on = None
+
+
+def upgrade() -> None:
+    # ==========================================================================
+    # ENUMS
+    # ==========================================================================
+
+    # Build session status enum
+    build_session_status_enum = sa.Enum(
+        "active",
+        "idle",
+        name="buildsessionstatus",
+        native_enum=False,
+    )
+
+    # Sandbox status enum
+    sandbox_status_enum = sa.Enum(
+        "provisioning",
+        "running",
+        "idle",
+        "sleeping",
+        "terminated",
+        "failed",
+        name="sandboxstatus",
+        native_enum=False,
+    )
+
+    # Artifact type enum
+    artifact_type_enum = sa.Enum(
+        "web_app",
+        "pptx",
+        "docx",
+        "markdown",
+        "excel",
+        "image",
+        name="artifacttype",
+        native_enum=False,
+    )
+
+    # ==========================================================================
+    # BUILD_SESSION TABLE
+    # ==========================================================================
+
+    op.create_table(
+        "build_session",
+        sa.Column("id", postgresql.UUID(as_uuid=True), primary_key=True),
+        sa.Column(
+            "user_id",
+            postgresql.UUID(as_uuid=True),
+            sa.ForeignKey("user.id", ondelete="CASCADE"),
+            nullable=True,
+        ),
+        sa.Column("name", sa.String(), nullable=True),
+        sa.Column(
+            "status",
+            build_session_status_enum,
+            nullable=False,
+            server_default="active",
+        ),
+        sa.Column(
+            "created_at",
+            sa.DateTime(timezone=True),
+            server_default=sa.text("now()"),
+            nullable=False,
+        ),
+        sa.Column(
+            "last_activity_at",
+            sa.DateTime(timezone=True),
+            server_default=sa.text("now()"),
+            nullable=False,
+        ),
+        sa.Column("nextjs_port", sa.Integer(), nullable=True),
+        sa.PrimaryKeyConstraint("id"),
+    )
+
+    op.create_index(
+        "ix_build_session_user_created",
+        "build_session",
+        ["user_id", sa.text("created_at DESC")],
+        unique=False,
+    )
+    op.create_index(
+        "ix_build_session_status",
+        "build_session",
+        ["status"],
+        unique=False,
+    )
+
+    # ==========================================================================
+    # SANDBOX TABLE (user-owned, one per user)
+    # ==========================================================================
+
+    op.create_table(
+        "sandbox",
+        sa.Column("id", postgresql.UUID(as_uuid=True), primary_key=True),
+        sa.Column(
+            "user_id",
+            postgresql.UUID(as_uuid=True),
+            sa.ForeignKey("user.id", ondelete="CASCADE"),
+            nullable=False,
+        ),
+        sa.Column("container_id", sa.String(), nullable=True),
+        sa.Column(
+            "status",
+            sandbox_status_enum,
+            nullable=False,
+            server_default="provisioning",
+        ),
+        sa.Column(
+            "created_at",
+            sa.DateTime(timezone=True),
+            server_default=sa.text("now()"),
+            nullable=False,
+        ),
+        sa.Column("last_heartbeat", sa.DateTime(timezone=True), nullable=True),
+        sa.PrimaryKeyConstraint("id"),
+        sa.UniqueConstraint("user_id", name="sandbox_user_id_key"),
+    )
+
+    op.create_index(
+        "ix_sandbox_status",
+        "sandbox",
+        ["status"],
+        unique=False,
+    )
+    op.create_index(
+        "ix_sandbox_container_id",
+        "sandbox",
+        ["container_id"],
+        unique=False,
+    )
+
+    # ==========================================================================
+    # ARTIFACT TABLE
+    # ==========================================================================
+
+    op.create_table(
+        "artifact",
+        sa.Column("id", postgresql.UUID(as_uuid=True), primary_key=True),
+        sa.Column(
+            "session_id",
+            postgresql.UUID(as_uuid=True),
+            sa.ForeignKey("build_session.id", ondelete="CASCADE"),
+            nullable=False,
+        ),
+        sa.Column("type", artifact_type_enum, nullable=False),
+        sa.Column("path", sa.String(), nullable=False),
+        sa.Column("name", sa.String(), 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.PrimaryKeyConstraint("id"),
+    )
+
+    op.create_index(
+        "ix_artifact_session_created",
+        "artifact",
+        ["session_id", sa.text("created_at DESC")],
+        unique=False,
+    )
+    op.create_index(
+        "ix_artifact_type",
+        "artifact",
+        ["type"],
+        unique=False,
+    )
+
+    # ==========================================================================
+    # SNAPSHOT TABLE
+    # ==========================================================================
+
+    op.create_table(
+        "snapshot",
+        sa.Column("id", postgresql.UUID(as_uuid=True), primary_key=True),
+        sa.Column(
+            "session_id",
+            postgresql.UUID(as_uuid=True),
+            sa.ForeignKey("build_session.id", ondelete="CASCADE"),
+            nullable=False,
+        ),
+        sa.Column("storage_path", sa.String(), nullable=False),
+        sa.Column("size_bytes", sa.BigInteger(), nullable=False, server_default="0"),
+        sa.Column(
+            "created_at",
+            sa.DateTime(timezone=True),
+            server_default=sa.text("now()"),
+            nullable=False,
+        ),
+        sa.PrimaryKeyConstraint("id"),
+    )
+
+    op.create_index(
+        "ix_snapshot_session_created",
+        "snapshot",
+        ["session_id", sa.text("created_at DESC")],
+        unique=False,
+    )
+
+    # ==========================================================================
+    # BUILD_MESSAGE TABLE
+    # ==========================================================================
+
+    op.create_table(
+        "build_message",
+        sa.Column("id", postgresql.UUID(as_uuid=True), primary_key=True),
+        sa.Column(
+            "session_id",
+            postgresql.UUID(as_uuid=True),
+            sa.ForeignKey("build_session.id", ondelete="CASCADE"),
+            nullable=False,
+        ),
+        sa.Column(
+            "turn_index",
+            sa.Integer(),
+            nullable=False,
+        ),
+        sa.Column(
+            "type",
+            sa.Enum(
+                "SYSTEM",
+                "USER",
+                "ASSISTANT",
+                "DANSWER",
+                name="messagetype",
+                create_type=False,
+                native_enum=False,
+            ),
+            nullable=False,
+        ),
+        sa.Column(
+            "message_metadata",
+            postgresql.JSONB(),
+            nullable=False,
+        ),
+        sa.Column(
+            "created_at",
+            sa.DateTime(timezone=True),
+            server_default=sa.text("now()"),
+            nullable=False,
+        ),
+        sa.PrimaryKeyConstraint("id"),
+    )
+
+    op.create_index(
+        "ix_build_message_session_turn",
+        "build_message",
+        ["session_id", "turn_index", sa.text("created_at ASC")],
+        unique=False,
+    )
+
+    # ==========================================================================
+    # CONNECTOR_CREDENTIAL_PAIR MODIFICATION
+    # ==========================================================================
+
+    op.add_column(
+        "connector_credential_pair",
+        sa.Column(
+            "processing_mode",
+            sa.String(),
+            nullable=False,
+            server_default="regular",
+        ),
+    )
+
+
+def downgrade() -> None:
+    # ==========================================================================
+    # CONNECTOR_CREDENTIAL_PAIR MODIFICATION
+    # ==========================================================================
+
+    op.drop_column("connector_credential_pair", "processing_mode")
+
+    # ==========================================================================
+    # BUILD_MESSAGE TABLE
+    # ==========================================================================
+
+    op.drop_index("ix_build_message_session_turn", table_name="build_message")
+    op.drop_table("build_message")
+
+    # ==========================================================================
+    # SNAPSHOT TABLE
+    # ==========================================================================
+
+    op.drop_index("ix_snapshot_session_created", table_name="snapshot")
+    op.drop_table("snapshot")
+
+    # ==========================================================================
+    # ARTIFACT TABLE
+    # ==========================================================================
+
+    op.drop_index("ix_artifact_type", table_name="artifact")
+    op.drop_index("ix_artifact_session_created", table_name="artifact")
+    op.drop_table("artifact")
+    sa.Enum(name="artifacttype").drop(op.get_bind(), checkfirst=True)
+
+    # ==========================================================================
+    # SANDBOX TABLE
+    # ==========================================================================
+
+    op.drop_index("ix_sandbox_container_id", table_name="sandbox")
+    op.drop_index("ix_sandbox_status", table_name="sandbox")
+    op.drop_table("sandbox")
+    sa.Enum(name="sandboxstatus").drop(op.get_bind(), checkfirst=True)
+
+    # ==========================================================================
+    # BUILD_SESSION TABLE
+    # ==========================================================================
+
+    op.drop_index("ix_build_session_status", table_name="build_session")
+    op.drop_index("ix_build_session_user_created", table_name="build_session")
+    op.drop_table("build_session")
+    sa.Enum(name="buildsessionstatus").drop(op.get_bind(), checkfirst=True)

backend/alembic/versions/72aa7de2e5cf_make_processing_mode_default_all_caps.py

@@ -0,0 +1,45 @@
+"""make processing mode default all caps
+
+Revision ID: 72aa7de2e5cf
+Revises: 2020d417ec84
+Create Date: 2026-01-26 18:58:47.705253
+
+This migration fixes the ProcessingMode enum value mismatch:
+- SQLAlchemy's Enum with native_enum=False uses enum member NAMES as valid values
+- The original migration stored lowercase VALUES ('regular', 'file_system')
+- This converts existing data to uppercase NAMES ('REGULAR', 'FILE_SYSTEM')
+- Also drops any spurious native PostgreSQL enum type that may have been auto-created
+"""
+
+from alembic import op
+
+
+# revision identifiers, used by Alembic.
+revision = "72aa7de2e5cf"
+down_revision = "2020d417ec84"
+branch_labels = None
+depends_on = None
+
+
+def upgrade() -> None:
+    # Convert existing lowercase values to uppercase to match enum member names
+    op.execute(
+        "UPDATE connector_credential_pair SET processing_mode = 'REGULAR' "
+        "WHERE processing_mode = 'regular'"
+    )
+    op.execute(
+        "UPDATE connector_credential_pair SET processing_mode = 'FILE_SYSTEM' "
+        "WHERE processing_mode = 'file_system'"
+    )
+
+    # Update the server default to use uppercase
+    op.alter_column(
+        "connector_credential_pair",
+        "processing_mode",
+        server_default="REGULAR",
+    )
+
+
+def downgrade() -> None:
+    # State prior to this was broken, so we don't want to revert back to it
+    pass

backend/alembic/versions/78ebc66946a0_remove_reranking_from_search_settings.py

@@ -0,0 +1,58 @@
+"""remove reranking from search_settings
+
+Revision ID: 78ebc66946a0
+Revises: 849b21c732f8
+Create Date: 2026-01-28
+
+"""
+
+from alembic import op
+import sqlalchemy as sa
+
+# revision identifiers, used by Alembic.
+revision = "78ebc66946a0"
+down_revision = "849b21c732f8"
+branch_labels: None = None
+depends_on: None = None
+
+
+def upgrade() -> None:
+    op.drop_column("search_settings", "disable_rerank_for_streaming")
+    op.drop_column("search_settings", "rerank_model_name")
+    op.drop_column("search_settings", "rerank_provider_type")
+    op.drop_column("search_settings", "rerank_api_key")
+    op.drop_column("search_settings", "rerank_api_url")
+    op.drop_column("search_settings", "num_rerank")
+
+
+def downgrade() -> None:
+    op.add_column(
+        "search_settings",
+        sa.Column(
+            "disable_rerank_for_streaming",
+            sa.Boolean(),
+            nullable=False,
+            server_default="false",
+        ),
+    )
+    op.add_column(
+        "search_settings", sa.Column("rerank_model_name", sa.String(), nullable=True)
+    )
+    op.add_column(
+        "search_settings", sa.Column("rerank_provider_type", sa.String(), nullable=True)
+    )
+    op.add_column(
+        "search_settings", sa.Column("rerank_api_key", sa.String(), nullable=True)
+    )
+    op.add_column(
+        "search_settings", sa.Column("rerank_api_url", sa.String(), nullable=True)
+    )
+    op.add_column(
+        "search_settings",
+        sa.Column(
+            "num_rerank",
+            sa.Integer(),
+            nullable=False,
+            server_default=str(20),
+        ),
+    )

backend/alembic/versions/81c22b1e2e78_hierarchy_nodes_v1.py

@@ -0,0 +1,349 @@
+"""hierarchy_nodes_v1
+
+Revision ID: 81c22b1e2e78
+Revises: 72aa7de2e5cf
+Create Date: 2026-01-13 18:10:01.021451
+
+"""
+
+from alembic import op
+import sqlalchemy as sa
+from sqlalchemy.dialects import postgresql
+
+from onyx.configs.constants import DocumentSource
+
+
+# revision identifiers, used by Alembic.
+revision = "81c22b1e2e78"
+down_revision = "72aa7de2e5cf"
+branch_labels = None
+depends_on = None
+
+
+# Human-readable display names for each source
+SOURCE_DISPLAY_NAMES: dict[str, str] = {
+    "ingestion_api": "Ingestion API",
+    "slack": "Slack",
+    "web": "Web",
+    "google_drive": "Google Drive",
+    "gmail": "Gmail",
+    "requesttracker": "Request Tracker",
+    "github": "GitHub",
+    "gitbook": "GitBook",
+    "gitlab": "GitLab",
+    "guru": "Guru",
+    "bookstack": "BookStack",
+    "outline": "Outline",
+    "confluence": "Confluence",
+    "jira": "Jira",
+    "slab": "Slab",
+    "productboard": "Productboard",
+    "file": "File",
+    "coda": "Coda",
+    "notion": "Notion",
+    "zulip": "Zulip",
+    "linear": "Linear",
+    "hubspot": "HubSpot",
+    "document360": "Document360",
+    "gong": "Gong",
+    "google_sites": "Google Sites",
+    "zendesk": "Zendesk",
+    "loopio": "Loopio",
+    "dropbox": "Dropbox",
+    "sharepoint": "SharePoint",
+    "teams": "Teams",
+    "salesforce": "Salesforce",
+    "discourse": "Discourse",
+    "axero": "Axero",
+    "clickup": "ClickUp",
+    "mediawiki": "MediaWiki",
+    "wikipedia": "Wikipedia",
+    "asana": "Asana",
+    "s3": "S3",
+    "r2": "R2",
+    "google_cloud_storage": "Google Cloud Storage",
+    "oci_storage": "OCI Storage",
+    "xenforo": "XenForo",
+    "not_applicable": "Not Applicable",
+    "discord": "Discord",
+    "freshdesk": "Freshdesk",
+    "fireflies": "Fireflies",
+    "egnyte": "Egnyte",
+    "airtable": "Airtable",
+    "highspot": "Highspot",
+    "drupal_wiki": "Drupal Wiki",
+    "imap": "IMAP",
+    "bitbucket": "Bitbucket",
+    "testrail": "TestRail",
+    "mock_connector": "Mock Connector",
+    "user_file": "User File",
+}
+
+
+def upgrade() -> None:
+    # 1. Create hierarchy_node table
+    op.create_table(
+        "hierarchy_node",
+        sa.Column("id", sa.Integer(), nullable=False),
+        sa.Column("raw_node_id", sa.String(), nullable=False),
+        sa.Column("display_name", sa.String(), nullable=False),
+        sa.Column("link", sa.String(), nullable=True),
+        sa.Column("source", sa.String(), nullable=False),
+        sa.Column("node_type", sa.String(), nullable=False),
+        sa.Column("document_id", sa.String(), nullable=True),
+        sa.Column("parent_id", sa.Integer(), nullable=True),
+        # Permission fields - same pattern as Document table
+        sa.Column(
+            "external_user_emails",
+            postgresql.ARRAY(sa.String()),
+            nullable=True,
+        ),
+        sa.Column(
+            "external_user_group_ids",
+            postgresql.ARRAY(sa.String()),
+            nullable=True,
+        ),
+        sa.Column("is_public", sa.Boolean(), nullable=False, server_default="false"),
+        sa.PrimaryKeyConstraint("id"),
+        # When document is deleted, just unlink (node can exist without document)
+        sa.ForeignKeyConstraint(["document_id"], ["document.id"], ondelete="SET NULL"),
+        # When parent node is deleted, orphan children (cleanup via pruning)
+        sa.ForeignKeyConstraint(
+            ["parent_id"], ["hierarchy_node.id"], ondelete="SET NULL"
+        ),
+        sa.UniqueConstraint(
+            "raw_node_id", "source", name="uq_hierarchy_node_raw_id_source"
+        ),
+    )
+    op.create_index("ix_hierarchy_node_parent_id", "hierarchy_node", ["parent_id"])
+    op.create_index(
+        "ix_hierarchy_node_source_type", "hierarchy_node", ["source", "node_type"]
+    )
+
+    # Add partial unique index to ensure only one SOURCE-type node per source
+    # This prevents duplicate source root nodes from being created
+    # NOTE: node_type stores enum NAME ('SOURCE'), not value ('source')
+    op.execute(
+        sa.text(
+            """
+            CREATE UNIQUE INDEX uq_hierarchy_node_one_source_per_type
+            ON hierarchy_node (source)
+            WHERE node_type = 'SOURCE'
+            """
+        )
+    )
+
+    # 2. Create hierarchy_fetch_attempt table
+    op.create_table(
+        "hierarchy_fetch_attempt",
+        sa.Column("id", postgresql.UUID(as_uuid=True), nullable=False),
+        sa.Column("connector_credential_pair_id", sa.Integer(), nullable=False),
+        sa.Column("status", sa.String(), nullable=False),
+        sa.Column("nodes_fetched", sa.Integer(), nullable=True, server_default="0"),
+        sa.Column("nodes_updated", sa.Integer(), nullable=True, server_default="0"),
+        sa.Column("error_msg", sa.Text(), nullable=True),
+        sa.Column("full_exception_trace", sa.Text(), nullable=True),
+        sa.Column(
+            "time_created",
+            sa.DateTime(timezone=True),
+            server_default=sa.func.now(),
+            nullable=False,
+        ),
+        sa.Column("time_started", sa.DateTime(timezone=True), nullable=True),
+        sa.Column(
+            "time_updated",
+            sa.DateTime(timezone=True),
+            server_default=sa.func.now(),
+            nullable=False,
+        ),
+        sa.PrimaryKeyConstraint("id"),
+        sa.ForeignKeyConstraint(
+            ["connector_credential_pair_id"],
+            ["connector_credential_pair.id"],
+            ondelete="CASCADE",
+        ),
+    )
+    op.create_index(
+        "ix_hierarchy_fetch_attempt_status", "hierarchy_fetch_attempt", ["status"]
+    )
+    op.create_index(
+        "ix_hierarchy_fetch_attempt_time_created",
+        "hierarchy_fetch_attempt",
+        ["time_created"],
+    )
+    op.create_index(
+        "ix_hierarchy_fetch_attempt_cc_pair",
+        "hierarchy_fetch_attempt",
+        ["connector_credential_pair_id"],
+    )
+
+    # 3. Insert SOURCE-type hierarchy nodes for each DocumentSource
+    # We insert these so every existing document can have a parent hierarchy node
+    # NOTE: SQLAlchemy's Enum with native_enum=False stores the enum NAME (e.g., 'GOOGLE_DRIVE'),
+    # not the VALUE (e.g., 'google_drive'). We must use .name for source and node_type columns.
+    # SOURCE nodes are always public since they're just categorical roots.
+    for source in DocumentSource:
+        source_name = (
+            source.name
+        )  # e.g., 'GOOGLE_DRIVE' - what SQLAlchemy stores/expects
+        source_value = source.value  # e.g., 'google_drive' - the raw_node_id
+        display_name = SOURCE_DISPLAY_NAMES.get(
+            source_value, source_value.replace("_", " ").title()
+        )
+        op.execute(
+            sa.text(
+                """
+                INSERT INTO hierarchy_node (raw_node_id, display_name, source, node_type, parent_id, is_public)
+                VALUES (:raw_node_id, :display_name, :source, 'SOURCE', NULL, true)
+                ON CONFLICT (raw_node_id, source) DO NOTHING
+                """
+            ).bindparams(
+                raw_node_id=source_value,  # Use .value for raw_node_id (human-readable identifier)
+                display_name=display_name,
+                source=source_name,  # Use .name for source column (SQLAlchemy enum storage)
+            )
+        )
+
+    # 4. Add parent_hierarchy_node_id column to document table
+    op.add_column(
+        "document",
+        sa.Column("parent_hierarchy_node_id", sa.Integer(), nullable=True),
+    )
+    # When hierarchy node is deleted, just unlink the document (SET NULL)
+    op.create_foreign_key(
+        "fk_document_parent_hierarchy_node",
+        "document",
+        "hierarchy_node",
+        ["parent_hierarchy_node_id"],
+        ["id"],
+        ondelete="SET NULL",
+    )
+    op.create_index(
+        "ix_document_parent_hierarchy_node_id",
+        "document",
+        ["parent_hierarchy_node_id"],
+    )
+
+    # 5. Set all existing documents' parent_hierarchy_node_id to their source's SOURCE node
+    # For documents with multiple connectors, we pick one source deterministically (MIN connector_id)
+    # NOTE: Both connector.source and hierarchy_node.source store enum NAMEs (e.g., 'GOOGLE_DRIVE')
+    # because SQLAlchemy Enum(native_enum=False) uses the enum name for storage.
+    op.execute(
+        sa.text(
+            """
+            UPDATE document d
+            SET parent_hierarchy_node_id = hn.id
+            FROM (
+                -- Get the source for each document (pick MIN connector_id for determinism)
+                SELECT DISTINCT ON (dbcc.id)
+                    dbcc.id as doc_id,
+                    c.source as source
+                FROM document_by_connector_credential_pair dbcc
+                JOIN connector c ON dbcc.connector_id = c.id
+                ORDER BY dbcc.id, dbcc.connector_id
+            ) doc_source
+            JOIN hierarchy_node hn ON hn.source = doc_source.source AND hn.node_type = 'SOURCE'
+            WHERE d.id = doc_source.doc_id
+            """
+        )
+    )
+
+    # Create the persona__hierarchy_node association table
+    op.create_table(
+        "persona__hierarchy_node",
+        sa.Column("persona_id", sa.Integer(), nullable=False),
+        sa.Column("hierarchy_node_id", sa.Integer(), nullable=False),
+        sa.ForeignKeyConstraint(
+            ["persona_id"],
+            ["persona.id"],
+            ondelete="CASCADE",
+        ),
+        sa.ForeignKeyConstraint(
+            ["hierarchy_node_id"],
+            ["hierarchy_node.id"],
+            ondelete="CASCADE",
+        ),
+        sa.PrimaryKeyConstraint("persona_id", "hierarchy_node_id"),
+    )
+
+    # Add index for efficient lookups
+    op.create_index(
+        "ix_persona__hierarchy_node_hierarchy_node_id",
+        "persona__hierarchy_node",
+        ["hierarchy_node_id"],
+    )
+
+    # Create the persona__document association table for attaching individual
+    # documents directly to assistants
+    op.create_table(
+        "persona__document",
+        sa.Column("persona_id", sa.Integer(), nullable=False),
+        sa.Column("document_id", sa.String(), nullable=False),
+        sa.ForeignKeyConstraint(
+            ["persona_id"],
+            ["persona.id"],
+            ondelete="CASCADE",
+        ),
+        sa.ForeignKeyConstraint(
+            ["document_id"],
+            ["document.id"],
+            ondelete="CASCADE",
+        ),
+        sa.PrimaryKeyConstraint("persona_id", "document_id"),
+    )
+
+    # Add index for efficient lookups by document_id
+    op.create_index(
+        "ix_persona__document_document_id",
+        "persona__document",
+        ["document_id"],
+    )
+
+    # 6. Add last_time_hierarchy_fetch column to connector_credential_pair table
+    op.add_column(
+        "connector_credential_pair",
+        sa.Column(
+            "last_time_hierarchy_fetch", sa.DateTime(timezone=True), nullable=True
+        ),
+    )
+
+
+def downgrade() -> None:
+    # Remove last_time_hierarchy_fetch from connector_credential_pair
+    op.drop_column("connector_credential_pair", "last_time_hierarchy_fetch")
+
+    # Drop persona__document table
+    op.drop_index("ix_persona__document_document_id", table_name="persona__document")
+    op.drop_table("persona__document")
+
+    # Drop persona__hierarchy_node table
+    op.drop_index(
+        "ix_persona__hierarchy_node_hierarchy_node_id",
+        table_name="persona__hierarchy_node",
+    )
+    op.drop_table("persona__hierarchy_node")
+
+    # Remove parent_hierarchy_node_id from document
+    op.drop_index("ix_document_parent_hierarchy_node_id", table_name="document")
+    op.drop_constraint(
+        "fk_document_parent_hierarchy_node", "document", type_="foreignkey"
+    )
+    op.drop_column("document", "parent_hierarchy_node_id")
+
+    # Drop hierarchy_fetch_attempt table
+    op.drop_index(
+        "ix_hierarchy_fetch_attempt_cc_pair", table_name="hierarchy_fetch_attempt"
+    )
+    op.drop_index(
+        "ix_hierarchy_fetch_attempt_time_created", table_name="hierarchy_fetch_attempt"
+    )
+    op.drop_index(
+        "ix_hierarchy_fetch_attempt_status", table_name="hierarchy_fetch_attempt"
+    )
+    op.drop_table("hierarchy_fetch_attempt")
+
+    # Drop hierarchy_node table
+    op.drop_index("uq_hierarchy_node_one_source_per_type", table_name="hierarchy_node")
+    op.drop_index("ix_hierarchy_node_source_type", table_name="hierarchy_node")
+    op.drop_index("ix_hierarchy_node_parent_id", table_name="hierarchy_node")
+    op.drop_table("hierarchy_node")

backend/alembic/versions/849b21c732f8_add_demo_data_enabled_to_build_session.py

@@ -0,0 +1,32 @@
+"""add demo_data_enabled to build_session
+
+Revision ID: 849b21c732f8
+Revises: 81c22b1e2e78
+Create Date: 2026-01-28 10:00:00.000000
+
+"""
+
+from alembic import op
+import sqlalchemy as sa
+
+# revision identifiers, used by Alembic.
+revision = "849b21c732f8"
+down_revision = "81c22b1e2e78"
+branch_labels = None
+depends_on = None
+
+
+def upgrade() -> None:
+    op.add_column(
+        "build_session",
+        sa.Column(
+            "demo_data_enabled",
+            sa.Boolean(),
+            nullable=False,
+            server_default=sa.text("true"),
+        ),
+    )
+
+
+def downgrade() -> None:
+    op.drop_column("build_session", "demo_data_enabled")

backend/alembic/versions/90b409d06e50_add_chat_compression_fields.py

@@ -0,0 +1,36 @@
+"""add_chat_compression_fields
+
+Revision ID: 90b409d06e50
+Revises: f220515df7b4
+Create Date: 2026-01-26 09:13:09.635427
+
+"""
+
+from alembic import op
+import sqlalchemy as sa
+
+
+# revision identifiers, used by Alembic.
+revision = "90b409d06e50"
+down_revision = "f220515df7b4"
+branch_labels = None
+depends_on = None
+
+
+def upgrade() -> None:
+    # Add last_summarized_message_id to chat_message
+    # This field marks a message as a summary and indicates the last message it covers.
+    # Summaries are branch-aware via their parent_message_id pointing to the branch.
+    op.add_column(
+        "chat_message",
+        sa.Column(
+            "last_summarized_message_id",
+            sa.Integer(),
+            sa.ForeignKey("chat_message.id", ondelete="SET NULL"),
+            nullable=True,
+        ),
+    )
+
+
+def downgrade() -> None:
+    op.drop_column("chat_message", "last_summarized_message_id")

backend/alembic/versions/9d1543a37106_add_processing_duration_seconds_to_chat_.py

@@ -0,0 +1,27 @@
+"""add processing_duration_seconds to chat_message
+
+Revision ID: 9d1543a37106
+Revises: cbc03e08d0f3
+Create Date: 2026-01-21 11:42:18.546188
+
+"""
+
+from alembic import op
+import sqlalchemy as sa
+
+# revision identifiers, used by Alembic.
+revision = "9d1543a37106"
+down_revision = "cbc03e08d0f3"
+branch_labels = None
+depends_on = None
+
+
+def upgrade() -> None:
+    op.add_column(
+        "chat_message",
+        sa.Column("processing_duration_seconds", sa.Float(), nullable=True),
+    )
+
+
+def downgrade() -> None:
+    op.drop_column("chat_message", "processing_duration_seconds")

backend/alembic/versions/be87a654d5af_persona_new_default_model_configuration_.py

@@ -0,0 +1,40 @@
+"""Persona new default model configuration id column
+
+Revision ID: be87a654d5af
+Revises: e7f8a9b0c1d2
+Create Date: 2026-01-30 11:14:17.306275
+
+"""
+
+from alembic import op
+import sqlalchemy as sa
+
+
+# revision identifiers, used by Alembic.
+revision = "be87a654d5af"
+down_revision = "e7f8a9b0c1d2"
+branch_labels = None
+depends_on = None
+
+
+def upgrade() -> None:
+    op.add_column(
+        "persona",
+        sa.Column("default_model_configuration_id", sa.Integer(), nullable=True),
+    )
+    op.create_foreign_key(
+        "fk_persona_default_model_configuration_id",
+        "persona",
+        "model_configuration",
+        ["default_model_configuration_id"],
+        ["id"],
+        ondelete="SET NULL",
+    )
+
+
+def downgrade() -> None:
+    op.drop_constraint(
+        "fk_persona_default_model_configuration_id", "persona", type_="foreignkey"
+    )
+
+    op.drop_column("persona", "default_model_configuration_id")

backend/alembic/versions/cbc03e08d0f3_add_opensearch_migration_tables.py

@@ -0,0 +1,128 @@
+"""add_opensearch_migration_tables
+
+Revision ID: cbc03e08d0f3
+Revises: be87a654d5af
+Create Date: 2026-01-31 17:00:45.176604
+
+"""
+
+from alembic import op
+import sqlalchemy as sa
+
+
+# revision identifiers, used by Alembic.
+revision = "cbc03e08d0f3"
+down_revision = "be87a654d5af"
+branch_labels = None
+depends_on = None
+
+
+def upgrade() -> None:
+    # 1. Create opensearch_document_migration_record table.
+    op.create_table(
+        "opensearch_document_migration_record",
+        sa.Column("document_id", sa.String(), nullable=False),
+        sa.Column("status", sa.String(), nullable=False, server_default="pending"),
+        sa.Column("error_message", sa.Text(), nullable=True),
+        sa.Column("attempts_count", sa.Integer(), nullable=False, server_default="0"),
+        sa.Column("last_attempt_at", sa.DateTime(timezone=True), nullable=True),
+        sa.Column(
+            "created_at",
+            sa.DateTime(timezone=True),
+            server_default=sa.func.now(),
+            nullable=False,
+        ),
+        sa.PrimaryKeyConstraint("document_id"),
+        sa.ForeignKeyConstraint(
+            ["document_id"],
+            ["document.id"],
+            ondelete="CASCADE",
+        ),
+    )
+    # 2. Create indices.
+    op.create_index(
+        "ix_opensearch_document_migration_record_status",
+        "opensearch_document_migration_record",
+        ["status"],
+    )
+    op.create_index(
+        "ix_opensearch_document_migration_record_attempts_count",
+        "opensearch_document_migration_record",
+        ["attempts_count"],
+    )
+    op.create_index(
+        "ix_opensearch_document_migration_record_created_at",
+        "opensearch_document_migration_record",
+        ["created_at"],
+    )
+
+    # 3. Create opensearch_tenant_migration_record table (singleton).
+    op.create_table(
+        "opensearch_tenant_migration_record",
+        sa.Column("id", sa.Integer(), nullable=False),
+        sa.Column(
+            "document_migration_record_table_population_status",
+            sa.String(),
+            nullable=False,
+            server_default="pending",
+        ),
+        sa.Column(
+            "num_times_observed_no_additional_docs_to_populate_migration_table",
+            sa.Integer(),
+            nullable=False,
+            server_default="0",
+        ),
+        sa.Column(
+            "overall_document_migration_status",
+            sa.String(),
+            nullable=False,
+            server_default="pending",
+        ),
+        sa.Column(
+            "num_times_observed_no_additional_docs_to_migrate",
+            sa.Integer(),
+            nullable=False,
+            server_default="0",
+        ),
+        sa.Column(
+            "last_updated_at",
+            sa.DateTime(timezone=True),
+            server_default=sa.func.now(),
+            nullable=False,
+        ),
+        sa.PrimaryKeyConstraint("id"),
+    )
+
+    # 4. Create unique index on constant to enforce singleton pattern.
+    op.execute(
+        sa.text(
+            """
+            CREATE UNIQUE INDEX idx_opensearch_tenant_migration_singleton
+            ON opensearch_tenant_migration_record ((true))
+            """
+        )
+    )
+
+
+def downgrade() -> None:
+    # Drop opensearch_tenant_migration_record.
+    op.drop_index(
+        "idx_opensearch_tenant_migration_singleton",
+        table_name="opensearch_tenant_migration_record",
+    )
+    op.drop_table("opensearch_tenant_migration_record")
+
+    # Drop opensearch_document_migration_record.
+    op.drop_index(
+        "ix_opensearch_document_migration_record_created_at",
+        table_name="opensearch_document_migration_record",
+    )
+    op.drop_index(
+        "ix_opensearch_document_migration_record_attempts_count",
+        table_name="opensearch_document_migration_record",
+    )
+    op.drop_index(
+        "ix_opensearch_document_migration_record_status",
+        table_name="opensearch_document_migration_record",
+    )
+    op.drop_table("opensearch_document_migration_record")

backend/alembic/versions/d5c86e2c6dc6_add_cascade_delete_to_search_query_user_.py

@@ -0,0 +1,35 @@
+"""add_cascade_delete_to_search_query_user_id
+
+Revision ID: d5c86e2c6dc6
+Revises: 90b409d06e50
+Create Date: 2026-02-04 16:05:04.749804
+
+"""
+
+from alembic import op
+
+
+# revision identifiers, used by Alembic.
+revision = "d5c86e2c6dc6"
+down_revision = "90b409d06e50"
+branch_labels = None
+depends_on = None
+
+
+def upgrade() -> None:
+    op.drop_constraint("search_query_user_id_fkey", "search_query", type_="foreignkey")
+    op.create_foreign_key(
+        "search_query_user_id_fkey",
+        "search_query",
+        "user",
+        ["user_id"],
+        ["id"],
+        ondelete="CASCADE",
+    )
+
+
+def downgrade() -> None:
+    op.drop_constraint("search_query_user_id_fkey", "search_query", type_="foreignkey")
+    op.create_foreign_key(
+        "search_query_user_id_fkey", "search_query", "user", ["user_id"], ["id"]
+    )

backend/alembic/versions/e7f8a9b0c1d2_create_anonymous_user.py

@@ -0,0 +1,125 @@
+"""create_anonymous_user
+
+This migration creates a permanent anonymous user in the database.
+When anonymous access is enabled, unauthenticated requests will use this user
+instead of returning user_id=NULL.
+
+Revision ID: e7f8a9b0c1d2
+Revises: f7ca3e2f45d9
+Create Date: 2026-01-15 14:00:00.000000
+
+"""
+
+from alembic import op
+import sqlalchemy as sa
+
+
+# revision identifiers, used by Alembic.
+revision = "e7f8a9b0c1d2"
+down_revision = "f7ca3e2f45d9"
+branch_labels = None
+depends_on = None
+
+# Must match constants in onyx/configs/constants.py file
+ANONYMOUS_USER_UUID = "00000000-0000-0000-0000-000000000002"
+ANONYMOUS_USER_EMAIL = "anonymous@onyx.app"
+
+# Tables with user_id foreign key that may need migration
+TABLES_WITH_USER_ID = [
+    "chat_session",
+    "credential",
+    "document_set",
+    "persona",
+    "tool",
+    "notification",
+    "inputprompt",
+]
+
+
+def upgrade() -> None:
+    """
+    Create the anonymous user for anonymous access feature.
+    Also migrates any remaining user_id=NULL records to the anonymous user.
+    """
+    connection = op.get_bind()
+
+    # Create the anonymous user (using ON CONFLICT to be idempotent)
+    connection.execute(
+        sa.text(
+            """
+            INSERT INTO "user" (id, email, hashed_password, is_active, is_superuser, is_verified, role)
+            VALUES (:id, :email, :hashed_password, :is_active, :is_superuser, :is_verified, :role)
+            ON CONFLICT (id) DO NOTHING
+            """
+        ),
+        {
+            "id": ANONYMOUS_USER_UUID,
+            "email": ANONYMOUS_USER_EMAIL,
+            "hashed_password": "",  # Empty password - user cannot log in directly
+            "is_active": True,  # Active so it can be used for anonymous access
+            "is_superuser": False,
+            "is_verified": True,  # Verified since no email verification needed
+            "role": "LIMITED",  # Anonymous users have limited role to restrict access
+        },
+    )
+
+    # Migrate any remaining user_id=NULL records to anonymous user
+    for table in TABLES_WITH_USER_ID:
+        try:
+            # Exclude public credential (id=0) which must remain user_id=NULL
+            # Exclude builtin tools (in_code_tool_id IS NOT NULL) which must remain user_id=NULL
+            # Exclude builtin personas (builtin_persona=True) which must remain user_id=NULL
+            # Exclude system input prompts (is_public=True with user_id=NULL) which must remain user_id=NULL
+            if table == "credential":
+                condition = "user_id IS NULL AND id != 0"
+            elif table == "tool":
+                condition = "user_id IS NULL AND in_code_tool_id IS NULL"
+            elif table == "persona":
+                condition = "user_id IS NULL AND builtin_persona = false"
+            elif table == "inputprompt":
+                condition = "user_id IS NULL AND is_public = false"
+            else:
+                condition = "user_id IS NULL"
+            result = connection.execute(
+                sa.text(
+                    f"""
+                    UPDATE "{table}"
+                    SET user_id = :user_id
+                    WHERE {condition}
+                    """
+                ),
+                {"user_id": ANONYMOUS_USER_UUID},
+            )
+            if result.rowcount > 0:
+                print(f"Updated {result.rowcount} rows in {table} to anonymous user")
+        except Exception as e:
+            print(f"Skipping {table}: {e}")
+
+
+def downgrade() -> None:
+    """
+    Set anonymous user's records back to NULL and delete the anonymous user.
+    """
+    connection = op.get_bind()
+
+    # Set records back to NULL
+    for table in TABLES_WITH_USER_ID:
+        try:
+            connection.execute(
+                sa.text(
+                    f"""
+                    UPDATE "{table}"
+                    SET user_id = NULL
+                    WHERE user_id = :user_id
+                    """
+                ),
+                {"user_id": ANONYMOUS_USER_UUID},
+            )
+        except Exception:
+            pass
+
+    # Delete the anonymous user
+    connection.execute(
+        sa.text('DELETE FROM "user" WHERE id = :user_id'),
+        {"user_id": ANONYMOUS_USER_UUID},
+    )

backend/alembic/versions/f220515df7b4_add_flow_mapping_table.py

@@ -0,0 +1,57 @@
+"""Add flow mapping table
+
+Revision ID: f220515df7b4
+Revises: cbc03e08d0f3
+Create Date: 2026-01-30 12:21:24.955922
+
+"""
+
+from onyx.db.enums import LLMModelFlowType
+from alembic import op
+import sqlalchemy as sa
+
+
+# revision identifiers, used by Alembic.
+revision = "f220515df7b4"
+down_revision = "9d1543a37106"
+branch_labels = None
+depends_on = None
+
+
+def upgrade() -> None:
+    op.create_table(
+        "llm_model_flow",
+        sa.Column("id", sa.Integer(), nullable=False),
+        sa.Column(
+            "llm_model_flow_type",
+            sa.Enum(LLMModelFlowType, name="llmmodelflowtype", native_enum=False),
+            nullable=False,
+        ),
+        sa.Column(
+            "is_default", sa.Boolean(), nullable=False, server_default=sa.text("false")
+        ),
+        sa.Column("model_configuration_id", sa.Integer(), nullable=False),
+        sa.PrimaryKeyConstraint("id"),
+        sa.ForeignKeyConstraint(
+            ["model_configuration_id"], ["model_configuration.id"], ondelete="CASCADE"
+        ),
+        sa.UniqueConstraint(
+            "llm_model_flow_type",
+            "model_configuration_id",
+            name="uq_model_config_per_llm_model_flow_type",
+        ),
+    )
+
+    # Partial unique index so that there is at most one default for each flow type
+    op.create_index(
+        "ix_one_default_per_llm_model_flow",
+        "llm_model_flow",
+        ["llm_model_flow_type"],
+        unique=True,
+        postgresql_where=sa.text("is_default IS TRUE"),
+    )
+
+
+def downgrade() -> None:
+    # Drop the llm_model_flow table (index is dropped automatically with table)
+    op.drop_table("llm_model_flow")

backend/alembic/versions/f7ca3e2f45d9_migrate_no_auth_data_to_placeholder.py

@@ -0,0 +1,281 @@
+"""migrate_no_auth_data_to_placeholder
+
+This migration handles the transition from AUTH_TYPE=disabled to requiring
+authentication. It creates a placeholder user and assigns all data that was
+created without a user (user_id=NULL) to this placeholder.
+
+A database trigger is installed that automatically transfers all data from
+the placeholder user to the first real user who registers, then drops itself.
+
+Revision ID: f7ca3e2f45d9
+Revises: 78ebc66946a0
+Create Date: 2026-01-15 12:49:53.802741
+
+"""
+
+import os
+
+from alembic import op
+import sqlalchemy as sa
+
+from shared_configs.configs import MULTI_TENANT
+
+
+# revision identifiers, used by Alembic.
+revision = "f7ca3e2f45d9"
+down_revision = "78ebc66946a0"
+branch_labels = None
+depends_on = None
+
+# Must match constants in onyx/configs/constants.py file
+NO_AUTH_PLACEHOLDER_USER_UUID = "00000000-0000-0000-0000-000000000001"
+NO_AUTH_PLACEHOLDER_USER_EMAIL = "no-auth-placeholder@onyx.app"
+
+# Trigger and function names
+TRIGGER_NAME = "trg_migrate_no_auth_data"
+FUNCTION_NAME = "migrate_no_auth_data_to_user"
+
+# Trigger function that migrates data from placeholder to first real user
+MIGRATE_NO_AUTH_TRIGGER_FUNCTION = f"""
+CREATE OR REPLACE FUNCTION {FUNCTION_NAME}()
+RETURNS TRIGGER AS $$
+DECLARE
+    placeholder_uuid UUID := '00000000-0000-0000-0000-000000000001'::uuid;
+    anonymous_uuid UUID := '00000000-0000-0000-0000-000000000002'::uuid;
+    placeholder_row RECORD;
+    schema_name TEXT;
+BEGIN
+    -- Skip if this is the placeholder user being inserted
+    IF NEW.id = placeholder_uuid THEN
+        RETURN NULL;
+    END IF;
+
+    -- Skip if this is the anonymous user being inserted (not a real user)
+    IF NEW.id = anonymous_uuid THEN
+        RETURN NULL;
+    END IF;
+
+    -- Skip if the new user is not active
+    IF NEW.is_active = FALSE THEN
+        RETURN NULL;
+    END IF;
+
+    -- Get current schema for self-cleanup
+    schema_name := current_schema();
+
+    -- Try to lock the placeholder user row with FOR UPDATE SKIP LOCKED
+    -- This ensures only one concurrent transaction can proceed with migration
+    -- SKIP LOCKED means if another transaction has the lock, we skip (don't wait)
+    SELECT id INTO placeholder_row
+    FROM "user"
+    WHERE id = placeholder_uuid
+    FOR UPDATE SKIP LOCKED;
+
+    IF NOT FOUND THEN
+        -- Either placeholder doesn't exist or another transaction has it locked
+        -- Either way, drop the trigger and return without making admin
+        EXECUTE format('DROP TRIGGER IF EXISTS {TRIGGER_NAME} ON %I."user"', schema_name);
+        EXECUTE format('DROP FUNCTION IF EXISTS %I.{FUNCTION_NAME}()', schema_name);
+        RETURN NULL;
+    END IF;
+
+    -- We have exclusive lock on placeholder - proceed with migration
+    -- The INSERT has already completed (AFTER INSERT), so NEW.id exists in the table
+
+    -- Migrate chat_session
+    UPDATE "chat_session" SET user_id = NEW.id WHERE user_id = placeholder_uuid;
+
+    -- Migrate credential (exclude public credential id=0)
+    UPDATE "credential" SET user_id = NEW.id WHERE user_id = placeholder_uuid AND id != 0;
+
+    -- Migrate document_set
+    UPDATE "document_set" SET user_id = NEW.id WHERE user_id = placeholder_uuid;
+
+    -- Migrate persona (exclude builtin personas)
+    UPDATE "persona" SET user_id = NEW.id WHERE user_id = placeholder_uuid AND builtin_persona = FALSE;
+
+    -- Migrate tool (exclude builtin tools)
+    UPDATE "tool" SET user_id = NEW.id WHERE user_id = placeholder_uuid AND in_code_tool_id IS NULL;
+
+    -- Migrate notification
+    UPDATE "notification" SET user_id = NEW.id WHERE user_id = placeholder_uuid;
+
+    -- Migrate inputprompt (exclude system/public prompts)
+    UPDATE "inputprompt" SET user_id = NEW.id WHERE user_id = placeholder_uuid AND is_public = FALSE;
+
+    -- Make the new user an admin (they had admin access in no-auth mode)
+    -- In AFTER INSERT trigger, we must UPDATE the row since it already exists
+    UPDATE "user" SET role = 'ADMIN' WHERE id = NEW.id;
+
+    -- Delete the placeholder user (we hold the lock so this is safe)
+    DELETE FROM "user" WHERE id = placeholder_uuid;
+
+    -- Drop the trigger and function (self-cleanup)
+    EXECUTE format('DROP TRIGGER IF EXISTS {TRIGGER_NAME} ON %I."user"', schema_name);
+    EXECUTE format('DROP FUNCTION IF EXISTS %I.{FUNCTION_NAME}()', schema_name);
+
+    RETURN NULL;
+END;
+$$ LANGUAGE plpgsql;
+"""
+
+MIGRATE_NO_AUTH_TRIGGER = f"""
+CREATE TRIGGER {TRIGGER_NAME}
+AFTER INSERT ON "user"
+FOR EACH ROW
+EXECUTE FUNCTION {FUNCTION_NAME}();
+"""
+
+
+def upgrade() -> None:
+    """
+    Create a placeholder user and assign all NULL user_id records to it.
+    Install a trigger that migrates data to the first real user and self-destructs.
+    Only runs if AUTH_TYPE is currently disabled/none.
+
+    Skipped in multi-tenant mode - each tenant starts fresh with no legacy data.
+    """
+    # Skip in multi-tenant mode - this migration handles single-tenant
+    # AUTH_TYPE=disabled -> auth transitions only
+    if MULTI_TENANT:
+        return
+
+    # Only run if AUTH_TYPE is currently disabled/none
+    # If they've already switched to auth-enabled, NULL data is stale anyway
+    auth_type = (os.environ.get("AUTH_TYPE") or "").lower()
+    if auth_type not in ("disabled", "none", ""):
+        print(f"AUTH_TYPE is '{auth_type}', not disabled. Skipping migration.")
+        return
+
+    connection = op.get_bind()
+
+    # Check if there are any NULL user_id records that need migration
+    tables_to_check = [
+        "chat_session",
+        "credential",
+        "document_set",
+        "persona",
+        "tool",
+        "notification",
+        "inputprompt",
+    ]
+
+    has_null_records = False
+    for table in tables_to_check:
+        try:
+            result = connection.execute(
+                sa.text(f'SELECT 1 FROM "{table}" WHERE user_id IS NULL LIMIT 1')
+            )
+            if result.fetchone():
+                has_null_records = True
+                break
+        except Exception:
+            # Table might not exist
+            pass
+
+    if not has_null_records:
+        return
+
+    # Create the placeholder user
+    connection.execute(
+        sa.text(
+            """
+            INSERT INTO "user" (id, email, hashed_password, is_active, is_superuser, is_verified, role)
+            VALUES (:id, :email, :hashed_password, :is_active, :is_superuser, :is_verified, :role)
+            """
+        ),
+        {
+            "id": NO_AUTH_PLACEHOLDER_USER_UUID,
+            "email": NO_AUTH_PLACEHOLDER_USER_EMAIL,
+            "hashed_password": "",  # Empty password - user cannot log in
+            "is_active": False,  # Inactive - user cannot log in
+            "is_superuser": False,
+            "is_verified": False,
+            "role": "BASIC",
+        },
+    )
+
+    # Assign NULL user_id records to the placeholder user
+    for table in tables_to_check:
+        try:
+            # Base condition for all tables
+            condition = "user_id IS NULL"
+            # Exclude public credential (id=0) which must remain user_id=NULL
+            if table == "credential":
+                condition += " AND id != 0"
+            # Exclude builtin tools (in_code_tool_id IS NOT NULL) which must remain user_id=NULL
+            elif table == "tool":
+                condition += " AND in_code_tool_id IS NULL"
+            # Exclude builtin personas which must remain user_id=NULL
+            elif table == "persona":
+                condition += " AND builtin_persona = FALSE"
+            # Exclude system/public input prompts which must remain user_id=NULL
+            elif table == "inputprompt":
+                condition += " AND is_public = FALSE"
+            result = connection.execute(
+                sa.text(
+                    f"""
+                    UPDATE "{table}"
+                    SET user_id = :user_id
+                    WHERE {condition}
+                    """
+                ),
+                {"user_id": NO_AUTH_PLACEHOLDER_USER_UUID},
+            )
+            if result.rowcount > 0:
+                print(f"Updated {result.rowcount} rows in {table}")
+        except Exception as e:
+            print(f"Skipping {table}: {e}")
+
+    # Install the trigger function and trigger for automatic migration on first user registration
+    connection.execute(sa.text(MIGRATE_NO_AUTH_TRIGGER_FUNCTION))
+    connection.execute(sa.text(MIGRATE_NO_AUTH_TRIGGER))
+    print("Installed trigger for automatic data migration on first user registration")
+
+
+def downgrade() -> None:
+    """
+    Drop trigger and function, set placeholder user's records back to NULL,
+    and delete the placeholder user.
+    """
+    # Skip in multi-tenant mode for consistency with upgrade
+    if MULTI_TENANT:
+        return
+
+    connection = op.get_bind()
+
+    # Drop trigger and function if they exist (they may have already self-destructed)
+    connection.execute(sa.text(f'DROP TRIGGER IF EXISTS {TRIGGER_NAME} ON "user"'))
+    connection.execute(sa.text(f"DROP FUNCTION IF EXISTS {FUNCTION_NAME}()"))
+
+    tables_to_update = [
+        "chat_session",
+        "credential",
+        "document_set",
+        "persona",
+        "tool",
+        "notification",
+        "inputprompt",
+    ]
+
+    # Set records back to NULL
+    for table in tables_to_update:
+        try:
+            connection.execute(
+                sa.text(
+                    f"""
+                    UPDATE "{table}"
+                    SET user_id = NULL
+                    WHERE user_id = :user_id
+                    """
+                ),
+                {"user_id": NO_AUTH_PLACEHOLDER_USER_UUID},
+            )
+        except Exception:
+            pass
+
+    # Delete the placeholder user
+    connection.execute(
+        sa.text('DELETE FROM "user" WHERE id = :user_id'),
+        {"user_id": NO_AUTH_PLACEHOLDER_USER_UUID},
+    )

backend/alembic_tenants/env.py

@@ -39,7 +39,7 @@ EXCLUDE_TABLES = {"kombu_queue", "kombu_message"}
 
 
 def include_object(
-    object: SchemaItem,
+    object: SchemaItem,  # noqa: ARG001
     name: str | None,
     type_: Literal[
         "schema",
@@ -49,8 +49,8 @@ def include_object(
         "unique_constraint",
         "foreign_key_constraint",
     ],
-    reflected: bool,
-    compare_to: SchemaItem | None,
+    reflected: bool,  # noqa: ARG001
+    compare_to: SchemaItem | None,  # noqa: ARG001
 ) -> bool:
     if type_ == "table" and name in EXCLUDE_TABLES:
         return False

backend/ee/onyx/access/access.py

@@ -116,7 +116,7 @@ def _get_access_for_documents(
     return access_map
 
 
-def _get_acl_for_user(user: User | None, db_session: Session) -> set[str]:
+def _get_acl_for_user(user: User, db_session: Session) -> set[str]:
     """Returns a list of ACL entries that the user has access to. This is meant to be
     used downstream to filter out documents that the user does not have access to. The
     user should have access to a document if at least one entry in the document's ACL
@@ -124,13 +124,16 @@ def _get_acl_for_user(user: User | None, db_session: Session) -> set[str]:
 
     NOTE: is imported in onyx.access.access by `fetch_versioned_implementation`
     DO NOT REMOVE."""
-    db_user_groups = fetch_user_groups_for_user(db_session, user.id) if user else []
+    is_anonymous = user.is_anonymous
+    db_user_groups = (
+        [] if is_anonymous else fetch_user_groups_for_user(db_session, user.id)
+    )
     prefixed_user_groups = [
         prefix_user_group(db_user_group.name) for db_user_group in db_user_groups
     ]
 
     db_external_groups = (
-        fetch_external_groups_for_user(db_session, user.id) if user else []
+        [] if is_anonymous else fetch_external_groups_for_user(db_session, user.id)
     )
     prefixed_external_groups = [
         prefix_external_group(db_external_group.external_user_group_id)

backend/ee/onyx/access/hierarchy_access.py

@@ -0,0 +1,11 @@
+from sqlalchemy.orm import Session
+
+from ee.onyx.db.external_perm import fetch_external_groups_for_user
+from onyx.db.models import User
+
+
+def _get_user_external_group_ids(db_session: Session, user: User) -> list[str]:
+    if not user:
+        return []
+    external_groups = fetch_external_groups_for_user(db_session, user.id)
+    return [external_group.external_user_group_id for external_group in external_groups]

backend/ee/onyx/auth/users.py

@@ -33,8 +33,8 @@ def get_default_admin_user_emails_() -> list[str]:
 
 async def current_cloud_superuser(
     request: Request,
-    user: User | None = Depends(current_admin_user),
-) -> User | None:
+    user: User = Depends(current_admin_user),
+) -> User:
     api_key = request.headers.get("Authorization", "").replace("Bearer ", "")
     if api_key != SUPER_CLOUD_API_KEY:
         raise HTTPException(status_code=401, detail="Invalid API key")

backend/ee/onyx/background/celery/tasks/doc_permission_syncing/tasks.py

@@ -25,6 +25,7 @@ from ee.onyx.db.connector_credential_pair import get_all_auto_sync_cc_pairs
 from ee.onyx.db.document import upsert_document_external_perms
 from ee.onyx.external_permissions.sync_params import get_source_perm_sync_config
 from onyx.access.models import DocExternalAccess
+from onyx.access.models import ElementExternalAccess
 from onyx.background.celery.apps.app_base import task_logger
 from onyx.background.celery.celery_redis import celery_find_task
 from onyx.background.celery.celery_redis import celery_get_queue_length
@@ -55,6 +56,9 @@ from onyx.db.enums import AccessType
 from onyx.db.enums import ConnectorCredentialPairStatus
 from onyx.db.enums import SyncStatus
 from onyx.db.enums import SyncType
+from onyx.db.hierarchy import (
+    update_hierarchy_node_permissions as db_update_hierarchy_node_permissions,
+)
 from onyx.db.models import ConnectorCredentialPair
 from onyx.db.permission_sync_attempt import complete_doc_permission_sync_attempt
 from onyx.db.permission_sync_attempt import create_doc_permission_sync_attempt
@@ -637,18 +641,25 @@ def connector_permission_sync_generator_task(
     ),
     stop=stop_after_delay(DOCUMENT_PERMISSIONS_UPDATE_STOP_AFTER),
 )
-def document_update_permissions(
+def element_update_permissions(
     tenant_id: str,
-    permissions: DocExternalAccess,
+    permissions: ElementExternalAccess,
     source_type_str: str,
     connector_id: int,
     credential_id: int,
 ) -> bool:
+    """Update permissions for a document or hierarchy node."""
     start = time.monotonic()
-
-    doc_id = permissions.doc_id
     external_access = permissions.external_access
 
+    # Determine element type and identifier for logging
+    if isinstance(permissions, DocExternalAccess):
+        element_id = permissions.doc_id
+        element_type = "doc"
+    else:
+        element_id = permissions.raw_node_id
+        element_type = "node"
+
     try:
         with get_session_with_tenant(tenant_id=tenant_id) as db_session:
             # Add the users to the DB if they don't exist
@@ -657,39 +668,57 @@ def document_update_permissions(
                 emails=list(external_access.external_user_emails),
                 continue_on_error=True,
             )
-            # Then upsert the document's external permissions
-            created_new_doc = upsert_document_external_perms(
-                db_session=db_session,
-                doc_id=doc_id,
-                external_access=external_access,
-                source_type=DocumentSource(source_type_str),
-            )
 
-            if created_new_doc:
-                # If a new document was created, we associate it with the cc_pair
-                upsert_document_by_connector_credential_pair(
+            if isinstance(permissions, DocExternalAccess):
+                # Document permission update
+                created_new_doc = upsert_document_external_perms(
                     db_session=db_session,
-                    connector_id=connector_id,
-                    credential_id=credential_id,
-                    document_ids=[doc_id],
+                    doc_id=permissions.doc_id,
+                    external_access=external_access,
+                    source_type=DocumentSource(source_type_str),
+                )
+
+                if created_new_doc:
+                    # If a new document was created, we associate it with the cc_pair
+                    upsert_document_by_connector_credential_pair(
+                        db_session=db_session,
+                        connector_id=connector_id,
+                        credential_id=credential_id,
+                        document_ids=[permissions.doc_id],
+                    )
+            else:
+                # Hierarchy node permission update
+                db_update_hierarchy_node_permissions(
+                    db_session=db_session,
+                    raw_node_id=permissions.raw_node_id,
+                    source=DocumentSource(permissions.source),
+                    is_public=external_access.is_public,
+                    external_user_emails=(
+                        list(external_access.external_user_emails)
+                        if external_access.external_user_emails
+                        else None
+                    ),
+                    external_user_group_ids=(
+                        list(external_access.external_user_group_ids)
+                        if external_access.external_user_group_ids
+                        else None
+                    ),
                 )
 
             elapsed = time.monotonic() - start
             task_logger.info(
-                f"connector_id={connector_id} "
-                f"doc={doc_id} "
+                f"{element_type}={element_id} "
                 f"action=update_permissions "
                 f"elapsed={elapsed:.2f}"
             )
     except Exception as e:
         task_logger.exception(
-            f"document_update_permissions exceptioned: "
-            f"connector_id={connector_id} doc_id={doc_id}"
+            f"element_update_permissions exceptioned: {element_type}={element_id}, {connector_id=} {credential_id=}"
         )
         raise e
     finally:
         task_logger.info(
-            f"document_update_permissions completed: connector_id={connector_id} doc={doc_id}"
+            f"element_update_permissions completed: {element_type}={element_id}, {connector_id=} {credential_id=}"
         )
 
     return True
@@ -922,7 +951,7 @@ class PermissionSyncCallback(IndexingHeartbeatInterface):
 
         return False
 
-    def progress(self, tag: str, amount: int) -> None:
+    def progress(self, tag: str, amount: int) -> None:  # noqa: ARG002
         try:
             self.redis_connector.permissions.set_active()
 
@@ -953,7 +982,7 @@ class PermissionSyncCallback(IndexingHeartbeatInterface):
 
 
 def monitor_ccpair_permissions_taskset(
-    tenant_id: str, key_bytes: bytes, r: Redis, db_session: Session
+    tenant_id: str, key_bytes: bytes, r: Redis, db_session: Session  # noqa: ARG001
 ) -> None:
     fence_key = key_bytes.decode("utf-8")
     cc_pair_id_str = RedisConnector.get_id_from_fence_key(fence_key)

backend/ee/onyx/background/celery/tasks/external_group_syncing/tasks.py

@@ -259,7 +259,7 @@ def check_for_external_group_sync(self: Task, *, tenant_id: str) -> bool | None:
 def try_creating_external_group_sync_task(
     app: Celery,
     cc_pair_id: int,
-    r: Redis,
+    r: Redis,  # noqa: ARG001
     tenant_id: str,
 ) -> str | None:
     """Returns an int if syncing is needed. The int represents the number of sync tasks generated.
@@ -344,7 +344,7 @@ def try_creating_external_group_sync_task(
     bind=True,
 )
 def connector_external_group_sync_generator_task(
-    self: Task,
+    self: Task,  # noqa: ARG001
     cc_pair_id: int,
     tenant_id: str,
 ) -> None:
@@ -590,8 +590,8 @@ def _perform_external_group_sync(
 
 def validate_external_group_sync_fences(
     tenant_id: str,
-    celery_app: Celery,
-    r: Redis,
+    celery_app: Celery,  # noqa: ARG001
+    r: Redis,  # noqa: ARG001
     r_replica: Redis,
     r_celery: Redis,
     lock_beat: RedisLock,

backend/ee/onyx/background/celery/tasks/query_history/tasks.py

@@ -40,7 +40,7 @@ def export_query_history_task(
     end: datetime,
     start_time: datetime,
     # Need to include the tenant_id since the TenantAwareTask needs this
-    tenant_id: str,
+    tenant_id: str,  # noqa: ARG001
 ) -> None:
     if not self.request.id:
         raise RuntimeError("No task id defined for this task; cannot identify it")

backend/ee/onyx/background/celery/tasks/tenant_provisioning/tasks.py

@@ -43,7 +43,7 @@ _TENANT_PROVISIONING_TIME_LIMIT = 60 * 10  # 10 minutes
     trail=False,
     bind=True,
 )
-def check_available_tenants(self: Task) -> None:
+def check_available_tenants(self: Task) -> None:  # noqa: ARG001
     """
     Check if we have enough pre-provisioned tenants available.
     If not, trigger the pre-provisioning of new tenants.

backend/ee/onyx/background/celery/tasks/usage_reporting/tasks.py

@@ -21,9 +21,9 @@ logger = setup_logger()
     trail=False,
 )
 def generate_usage_report_task(
-    self: Task,
+    self: Task,  # noqa: ARG001
     *,
-    tenant_id: str,
+    tenant_id: str,  # noqa: ARG001
     user_id: str | None = None,
     period_from: str | None = None,
     period_to: str | None = None,

backend/ee/onyx/background/task_name_builders.py

@@ -7,7 +7,7 @@ QUERY_HISTORY_TASK_NAME_PREFIX = OnyxCeleryTask.EXPORT_QUERY_HISTORY_TASK
 
 
 def name_chat_ttl_task(
-    retention_limit_days: float, tenant_id: str | None = None
+    retention_limit_days: float, tenant_id: str | None = None  # noqa: ARG001
 ) -> str:
     return f"chat_ttl_{retention_limit_days}_days"
 

backend/ee/onyx/configs/app_configs.py

@@ -122,6 +122,9 @@ SUPER_CLOUD_API_KEY = os.environ.get("SUPER_CLOUD_API_KEY", "api_key")
 # when the capture is called. These defaults prevent Posthog issues from breaking the Onyx app
 POSTHOG_API_KEY = os.environ.get("POSTHOG_API_KEY") or "FooBar"
 POSTHOG_HOST = os.environ.get("POSTHOG_HOST") or "https://us.i.posthog.com"
+POSTHOG_DEBUG_LOGS_ENABLED = (
+    os.environ.get("POSTHOG_DEBUG_LOGS_ENABLED", "").lower() == "true"
+)
 
 MARKETING_POSTHOG_API_KEY = os.environ.get("MARKETING_POSTHOG_API_KEY")
 
@@ -133,3 +136,9 @@ GATED_TENANTS_KEY = "gated_tenants"
 LICENSE_ENFORCEMENT_ENABLED = (
     os.environ.get("LICENSE_ENFORCEMENT_ENABLED", "").lower() == "true"
 )
+
+# Cloud data plane URL - self-hosted instances call this to reach cloud proxy endpoints
+# Used when MULTI_TENANT=false (self-hosted mode)
+CLOUD_DATA_PLANE_URL = os.environ.get(
+    "CLOUD_DATA_PLANE_URL", "https://cloud.onyx.app/api"
+)

backend/ee/onyx/configs/license_enforcement_config.py

@@ -0,0 +1,73 @@
+"""Constants for license enforcement.
+
+This file is the single source of truth for:
+1. Paths that bypass license enforcement (always accessible)
+2. Paths that require an EE license (EE-only features)
+
+Import these constants in both production code and tests to ensure consistency.
+"""
+
+# Paths that are ALWAYS accessible, even when license is expired/gated.
+# These enable users to:
+#   /auth - Log in/out (users can't fix billing if locked out of auth)
+#   /license - Fetch, upload, or check license status
+#   /health - Health checks for load balancers/orchestrators
+#   /me - Basic user info needed for UI rendering
+#   /settings, /enterprise-settings - View app status and branding
+#   /billing - Unified billing API
+#   /proxy - Self-hosted proxy endpoints (have own license-based auth)
+#   /tenants/billing-* - Legacy billing endpoints (backwards compatibility)
+#   /manage/users, /users - User management (needed for seat limit resolution)
+#   /notifications - Needed for UI to load properly
+LICENSE_ENFORCEMENT_ALLOWED_PREFIXES: frozenset[str] = frozenset(
+    {
+        "/auth",
+        "/license",
+        "/health",
+        "/me",
+        "/settings",
+        "/enterprise-settings",
+        # Billing endpoints (unified API for both MT and self-hosted)
+        "/billing",
+        "/admin/billing",
+        # Proxy endpoints for self-hosted billing (no tenant context)
+        "/proxy",
+        # Legacy tenant billing endpoints (kept for backwards compatibility)
+        "/tenants/billing-information",
+        "/tenants/create-customer-portal-session",
+        "/tenants/create-subscription-session",
+        # User management - needed to remove users when seat limit exceeded
+        "/manage/users",
+        "/manage/admin/users",
+        "/manage/admin/valid-domains",
+        "/manage/admin/deactivate-user",
+        "/manage/admin/delete-user",
+        "/users",
+        # Notifications - needed for UI to load properly
+        "/notifications",
+    }
+)
+
+# EE-only paths that require a valid license.
+# Users without a license (community edition) cannot access these.
+# These are blocked even when user has never subscribed (no license).
+EE_ONLY_PATH_PREFIXES: frozenset[str] = frozenset(
+    {
+        # User groups and access control
+        "/manage/admin/user-group",
+        # Analytics and reporting
+        "/analytics",
+        # Query history (admin chat session endpoints)
+        "/admin/chat-sessions",
+        "/admin/chat-session-history",
+        "/admin/query-history",
+        # Usage reporting/export
+        "/admin/usage-report",
+        # Standard answers (canned responses)
+        "/manage/admin/standard-answer",
+        # Token rate limits
+        "/admin/token-rate-limits",
+        # Evals
+        "/evals",
+    }
+)

backend/ee/onyx/db/analytics.py

@@ -334,11 +334,9 @@ def fetch_assistant_unique_users_total(
 # Users can view assistant stats if they created the persona,
 # or if they are an admin
 def user_can_view_assistant_stats(
-    db_session: Session, user: User | None, assistant_id: int
+    db_session: Session, user: User, assistant_id: int
 ) -> bool:
-    # If user is None and auth is disabled, assume the user is an admin
-
-    if user is None or user.role == UserRole.ADMIN:
+    if user.role == UserRole.ADMIN:
         return True
 
     # Check if the user created the persona

backend/ee/onyx/db/document_set.py

@@ -54,7 +54,7 @@ def delete_document_set_privacy__no_commit(
 def fetch_document_sets(
     user_id: UUID | None,
     db_session: Session,
-    include_outdated: bool = True,  # Parameter only for versioned implementation, unused
+    include_outdated: bool = True,  # Parameter only for versioned implementation, unused  # noqa: ARG001
 ) -> list[tuple[DocumentSet, list[ConnectorCredentialPair]]]:
     assert user_id is not None
 

backend/ee/onyx/db/hierarchy.py

@@ -0,0 +1,67 @@
+"""EE version of hierarchy node access control.
+
+This module provides permission-aware hierarchy node access for Enterprise Edition.
+It filters hierarchy nodes based on user email and external group membership.
+"""
+
+from sqlalchemy import any_
+from sqlalchemy import or_
+from sqlalchemy import select
+from sqlalchemy.dialects import postgresql
+from sqlalchemy.orm import Session
+from sqlalchemy.sql.elements import ColumnElement
+
+from onyx.configs.constants import DocumentSource
+from onyx.db.models import HierarchyNode
+
+
+def _build_hierarchy_access_filter(
+    user_email: str | None,
+    external_group_ids: list[str],
+) -> ColumnElement[bool]:
+    """Build SQLAlchemy filter for hierarchy node access.
+
+    A user can access a hierarchy node if any of the following are true:
+    - The node is marked as public (is_public=True)
+    - The user's email is in the node's external_user_emails list
+    - Any of the user's external group IDs overlap with the node's external_user_group_ids
+    """
+    access_filters: list[ColumnElement[bool]] = [HierarchyNode.is_public.is_(True)]
+    if user_email:
+        access_filters.append(any_(HierarchyNode.external_user_emails) == user_email)
+    if external_group_ids:
+        access_filters.append(
+            HierarchyNode.external_user_group_ids.overlap(
+                postgresql.array(external_group_ids)
+            )
+        )
+    return or_(*access_filters)
+
+
+def _get_accessible_hierarchy_nodes_for_source(
+    db_session: Session,
+    source: DocumentSource,
+    user_email: str | None,
+    external_group_ids: list[str],
+) -> list[HierarchyNode]:
+    """
+    EE version: Returns hierarchy nodes filtered by user permissions.
+
+    A user can access a hierarchy node if any of the following are true:
+    - The node is marked as public (is_public=True)
+    - The user's email is in the node's external_user_emails list
+    - Any of the user's external group IDs overlap with the node's external_user_group_ids
+
+    Args:
+        db_session: SQLAlchemy session
+        source: Document source type
+        user_email: User's email for permission checking
+        external_group_ids: User's external group IDs for permission checking
+
+    Returns:
+        List of HierarchyNode objects the user has access to
+    """
+    stmt = select(HierarchyNode).where(HierarchyNode.source == source)
+    stmt = stmt.where(_build_hierarchy_access_filter(user_email, external_group_ids))
+    stmt = stmt.order_by(HierarchyNode.display_name)
+    return list(db_session.execute(stmt).scalars().all())

backend/ee/onyx/db/license.py

@@ -1,6 +1,7 @@
 """Database and cache operations for the license table."""
 
 from datetime import datetime
+from typing import NamedTuple
 
 from sqlalchemy import func
 from sqlalchemy import select
@@ -9,6 +10,8 @@ from sqlalchemy.orm import Session
 from ee.onyx.server.license.models import LicenseMetadata
 from ee.onyx.server.license.models import LicensePayload
 from ee.onyx.server.license.models import LicenseSource
+from onyx.auth.schemas import UserRole
+from onyx.configs.constants import ANONYMOUS_USER_EMAIL
 from onyx.db.models import License
 from onyx.db.models import User
 from onyx.redis.redis_pool import get_redis_client
@@ -23,6 +26,13 @@ LICENSE_METADATA_KEY = "license:metadata"
 LICENSE_CACHE_TTL_SECONDS = 86400  # 24 hours
 
 
+class SeatAvailabilityResult(NamedTuple):
+    """Result of a seat availability check."""
+
+    available: bool
+    error_message: str | None = None
+
+
 # -----------------------------------------------------------------------------
 # Database CRUD Operations
 # -----------------------------------------------------------------------------
@@ -95,23 +105,32 @@ def delete_license(db_session: Session) -> bool:
 
 def get_used_seats(tenant_id: str | None = None) -> int:
     """
-    Get current seat usage.
+    Get current seat usage directly from database.
 
     For multi-tenant: counts users in UserTenantMapping for this tenant.
-    For self-hosted: counts all active users (includes both Onyx UI users
-    and Slack users who have been converted to Onyx users).
+    For self-hosted: counts all active users (excludes EXT_PERM_USER role
+    and the anonymous system user).
+
+    TODO: Exclude API key dummy users from seat counting. API keys create
+    users with emails like `__DANSWER_API_KEY_*` that should not count toward
+    seat limits. See: https://linear.app/onyx-app/issue/ENG-3518
     """
     if MULTI_TENANT:
         from ee.onyx.server.tenants.user_mapping import get_tenant_count
 
         return get_tenant_count(tenant_id or get_current_tenant_id())
     else:
-        # Self-hosted: count all active users (Onyx + converted Slack users)
         from onyx.db.engine.sql_engine import get_session_with_current_tenant
 
         with get_session_with_current_tenant() as db_session:
             result = db_session.execute(
-                select(func.count()).select_from(User).where(User.is_active)  # type: ignore
+                select(func.count())
+                .select_from(User)
+                .where(
+                    User.is_active == True,  # type: ignore  # noqa: E712
+                    User.role != UserRole.EXT_PERM_USER,
+                    User.email != ANONYMOUS_USER_EMAIL,  # type: ignore
+                )
             )
             return result.scalar() or 0
 
@@ -211,10 +230,10 @@ def update_license_cache(
         stripe_subscription_id=payload.stripe_subscription_id,
     )
 
-    redis_client.setex(
+    redis_client.set(
         LICENSE_METADATA_KEY,
-        LICENSE_CACHE_TTL_SECONDS,
         metadata.model_dump_json(),
+        ex=LICENSE_CACHE_TTL_SECONDS,
     )
 
     logger.info(f"License cache updated: {metadata.seats} seats, status={status.value}")
@@ -276,3 +295,43 @@ def get_license_metadata(
 
     # Refresh from database
     return refresh_license_cache(db_session, tenant_id)
+
+
+def check_seat_availability(
+    db_session: Session,
+    seats_needed: int = 1,
+    tenant_id: str | None = None,
+) -> SeatAvailabilityResult:
+    """
+    Check if there are enough seats available to add users.
+
+    Args:
+        db_session: Database session
+        seats_needed: Number of seats needed (default 1)
+        tenant_id: Tenant ID (for multi-tenant deployments)
+
+    Returns:
+        SeatAvailabilityResult with available=True if seats are available,
+        or available=False with error_message if limit would be exceeded.
+        Returns available=True if no license exists (self-hosted = unlimited).
+    """
+    metadata = get_license_metadata(db_session, tenant_id)
+
+    # No license = no enforcement (self-hosted without license)
+    if metadata is None:
+        return SeatAvailabilityResult(available=True)
+
+    # Calculate current usage directly from DB (not cache) for accuracy
+    current_used = get_used_seats(tenant_id)
+    total_seats = metadata.seats
+
+    # Use > (not >=) to allow filling to exactly 100% capacity
+    would_exceed_limit = current_used + seats_needed > total_seats
+    if would_exceed_limit:
+        return SeatAvailabilityResult(
+            available=False,
+            error_message=f"Seat limit would be exceeded: {current_used} of {total_seats} seats used, "
+            f"cannot add {seats_needed} more user(s).",
+        )
+
+    return SeatAvailabilityResult(available=True)

backend/ee/onyx/db/token_limit.py

@@ -7,7 +7,6 @@ from sqlalchemy import select
 from sqlalchemy.orm import aliased
 from sqlalchemy.orm import Session
 
-from onyx.configs.app_configs import DISABLE_AUTH
 from onyx.configs.constants import TokenRateLimitScope
 from onyx.db.models import TokenRateLimit
 from onyx.db.models import TokenRateLimit__UserGroup
@@ -18,13 +17,15 @@ from onyx.db.models import UserRole
 from onyx.server.token_rate_limits.models import TokenRateLimitArgs
 
 
-def _add_user_filters(
-    stmt: Select, user: User | None, get_editable: bool = True
-) -> Select:
-    # If user is None and auth is disabled, assume the user is an admin
-    if (user is None and DISABLE_AUTH) or (user and user.role == UserRole.ADMIN):
+def _add_user_filters(stmt: Select, user: User, get_editable: bool = True) -> Select:
+    if user.role == UserRole.ADMIN:
         return stmt
 
+    # If anonymous user, only show global/public token_rate_limits
+    if user.is_anonymous:
+        where_clause = TokenRateLimit.scope == TokenRateLimitScope.GLOBAL
+        return stmt.where(where_clause)
+
     stmt = stmt.distinct()
     TRLimit_UG = aliased(TokenRateLimit__UserGroup)
     User__UG = aliased(User__UserGroup)
@@ -49,11 +50,6 @@ def _add_user_filters(
     - if we are not editing, we show all token_rate_limits in the groups the user curates
     """
 
-    # If user is None, this is an anonymous user and we should only show public token_rate_limits
-    if user is None:
-        where_clause = TokenRateLimit.scope == TokenRateLimitScope.GLOBAL
-        return stmt.where(where_clause)
-
     where_clause = User__UG.user_id == user.id
     if user.role == UserRole.CURATOR and get_editable:
         where_clause &= User__UG.is_curator == True  # noqa: E712
@@ -114,7 +110,7 @@ def insert_user_group_token_rate_limit(
 def fetch_user_group_token_rate_limits_for_user(
     db_session: Session,
     group_id: int,
-    user: User | None,
+    user: User,
     enabled_only: bool = False,
     ordered: bool = True,
     get_editable: bool = True,

backend/ee/onyx/db/user_group.py

@@ -125,7 +125,7 @@ def _cleanup_document_set__user_group_relationships__no_commit(
 
 def validate_object_creation_for_user(
     db_session: Session,
-    user: User | None,
+    user: User,
     target_group_ids: list[int] | None = None,
     object_is_public: bool | None = None,
     object_is_perm_sync: bool | None = None,
@@ -144,7 +144,8 @@ def validate_object_creation_for_user(
     if object_is_perm_sync and not target_group_ids:
         return
 
-    if not user or user.role == UserRole.ADMIN:
+    # Admins are allowed
+    if user.role == UserRole.ADMIN:
         return
 
     # Allow curators and global curators to create public objects
@@ -474,14 +475,15 @@ def remove_curator_status__no_commit(db_session: Session, user: User) -> None:
 def _validate_curator_relationship_update_requester(
     db_session: Session,
     user_group_id: int,
-    user_making_change: User | None = None,
+    user_making_change: User,
 ) -> None:
     """
     This function validates that the user making the change has the necessary permissions
     to update the curator relationship for the target user in the given user group.
     """
 
-    if user_making_change is None or user_making_change.role == UserRole.ADMIN:
+    # Admins can update curator relationships for any group
+    if user_making_change.role == UserRole.ADMIN:
         return
 
     # check if the user making the change is a curator in the group they are changing the curator relationship for
@@ -550,7 +552,7 @@ def update_user_curator_relationship(
     db_session: Session,
     user_group_id: int,
     set_curator_request: SetCuratorRequest,
-    user_making_change: User | None = None,
+    user_making_change: User,
 ) -> None:
     target_user = fetch_user_by_id(db_session, set_curator_request.user_id)
     if not target_user:
@@ -599,7 +601,7 @@ def update_user_curator_relationship(
 
 def add_users_to_user_group(
     db_session: Session,
-    user: User | None,
+    user: User,
     user_group_id: int,
     user_ids: list[UUID],
 ) -> UserGroup:
@@ -641,7 +643,7 @@ def add_users_to_user_group(
 
 def update_user_group(
     db_session: Session,
-    user: User | None,
+    user: User,  # noqa: ARG001
     user_group_id: int,
     user_group_update: UserGroupUpdate,
 ) -> UserGroup:

backend/ee/onyx/external_permissions/confluence/doc_sync.py

@@ -8,7 +8,7 @@ from collections.abc import Generator
 from ee.onyx.external_permissions.perm_sync_types import FetchAllDocumentsFunction
 from ee.onyx.external_permissions.perm_sync_types import FetchAllDocumentsIdsFunction
 from ee.onyx.external_permissions.utils import generic_doc_sync
-from onyx.access.models import DocExternalAccess
+from onyx.access.models import ElementExternalAccess
 from onyx.configs.constants import DocumentSource
 from onyx.connectors.confluence.connector import ConfluenceConnector
 from onyx.connectors.credentials_provider import OnyxDBCredentialsProvider
@@ -25,10 +25,10 @@ CONFLUENCE_DOC_SYNC_LABEL = "confluence_doc_sync"
 
 def confluence_doc_sync(
     cc_pair: ConnectorCredentialPair,
-    fetch_all_existing_docs_fn: FetchAllDocumentsFunction,
+    fetch_all_existing_docs_fn: FetchAllDocumentsFunction,  # noqa: ARG001
     fetch_all_existing_docs_ids_fn: FetchAllDocumentsIdsFunction,
     callback: IndexingHeartbeatInterface | None,
-) -> Generator[DocExternalAccess, None, None]:
+) -> Generator[ElementExternalAccess, None, None]:
     """
     Fetches document permissions from Confluence and yields DocExternalAccess objects.
     Compares fetched documents against existing documents in the DB for the connector.

backend/ee/onyx/external_permissions/confluence/page_access.py

@@ -1,6 +1,8 @@
 from typing import Any
 
 from onyx.access.models import ExternalAccess
+from onyx.access.utils import build_ext_group_name_for_onyx
+from onyx.configs.constants import DocumentSource
 from onyx.connectors.confluence.onyx_confluence import (
     get_user_email_from_username__server,
 )
@@ -72,6 +74,7 @@ def get_page_restrictions(
     page_id: str,
     page_restrictions: dict[str, Any],
     ancestors: list[dict[str, Any]],
+    add_prefix: bool = False,
 ) -> ExternalAccess | None:
     """
     This function gets the restrictions for a page. In Confluence, a child can have
@@ -79,6 +82,9 @@ def get_page_restrictions(
 
     If no restrictions are found anywhere, then return None, indicating that the page
     should inherit the space's restrictions.
+
+    add_prefix: When True, prefix group IDs with source type (for indexing path).
+               When False (default), leave unprefixed (for permission sync path).
     """
     found_user_emails: set[str] = set()
     found_group_names: set[str] = set()
@@ -92,13 +98,22 @@ def get_page_restrictions(
             restrictions=page_restrictions,
         )
     )
+
+    def _maybe_prefix_groups(group_names: set[str]) -> set[str]:
+        if add_prefix:
+            return {
+                build_ext_group_name_for_onyx(g, DocumentSource.CONFLUENCE)
+                for g in group_names
+            }
+        return group_names
+
     # if there are individual page-level restrictions, then this is the accurate
     # restriction for the page. You cannot both have page-level restrictions AND
     # inherit restrictions from the parent.
     if found_any_page_level_restriction:
         return ExternalAccess(
             external_user_emails=found_user_emails,
-            external_user_group_ids=found_group_names,
+            external_user_group_ids=_maybe_prefix_groups(found_group_names),
             is_public=False,
         )
 
@@ -125,7 +140,7 @@ def get_page_restrictions(
             )
             return ExternalAccess(
                 external_user_emails=ancestor_user_emails,
-                external_user_group_ids=ancestor_group_names,
+                external_user_group_ids=_maybe_prefix_groups(ancestor_group_names),
                 is_public=False,
             )
 

backend/ee/onyx/external_permissions/confluence/space_access.py

@@ -3,6 +3,8 @@ from ee.onyx.external_permissions.confluence.constants import ALL_CONF_EMAILS_GR
 from ee.onyx.external_permissions.confluence.constants import REQUEST_PAGINATION_LIMIT
 from ee.onyx.external_permissions.confluence.constants import VIEWSPACE_PERMISSION_TYPE
 from onyx.access.models import ExternalAccess
+from onyx.access.utils import build_ext_group_name_for_onyx
+from onyx.configs.constants import DocumentSource
 from onyx.connectors.confluence.onyx_confluence import (
     get_user_email_from_username__server,
 )
@@ -112,6 +114,7 @@ def get_space_permission(
     confluence_client: OnyxConfluence,
     space_key: str,
     is_cloud: bool,
+    add_prefix: bool = False,
 ) -> ExternalAccess:
     if is_cloud:
         space_permissions = _get_cloud_space_permissions(confluence_client, space_key)
@@ -130,13 +133,32 @@ def get_space_permission(
             f"permissions for space '{space_key}'"
         )
 
+    # Prefix group IDs with source type if requested (for indexing path)
+    if add_prefix and space_permissions.external_user_group_ids:
+        prefixed_groups = {
+            build_ext_group_name_for_onyx(g, DocumentSource.CONFLUENCE)
+            for g in space_permissions.external_user_group_ids
+        }
+        return ExternalAccess(
+            external_user_emails=space_permissions.external_user_emails,
+            external_user_group_ids=prefixed_groups,
+            is_public=space_permissions.is_public,
+        )
+
     return space_permissions
 
 
 def get_all_space_permissions(
     confluence_client: OnyxConfluence,
     is_cloud: bool,
+    add_prefix: bool = False,
 ) -> dict[str, ExternalAccess]:
+    """
+    Get access permissions for all spaces in Confluence.
+
+    add_prefix: When True, prefix group IDs with source type (for indexing path).
+               When False (default), leave unprefixed (for permission sync path).
+    """
     logger.debug("Getting space permissions")
     # Gets all the spaces in the Confluence instance
     all_space_keys = [
@@ -151,7 +173,9 @@ def get_all_space_permissions(
     logger.debug(f"Got {len(all_space_keys)} spaces from confluence")
     space_permissions_by_space_key: dict[str, ExternalAccess] = {}
     for space_key in all_space_keys:
-        space_permissions = get_space_permission(confluence_client, space_key, is_cloud)
+        space_permissions = get_space_permission(
+            confluence_client, space_key, is_cloud, add_prefix
+        )
 
         # Stores the permissions for each space
         space_permissions_by_space_key[space_key] = space_permissions

backend/ee/onyx/external_permissions/github/doc_sync.py

@@ -34,7 +34,7 @@ GITHUB_DOC_SYNC_LABEL = "github_doc_sync"
 def github_doc_sync(
     cc_pair: ConnectorCredentialPair,
     fetch_all_existing_docs_fn: FetchAllDocumentsFunction,
-    fetch_all_existing_docs_ids_fn: FetchAllDocumentsIdsFunction,
+    fetch_all_existing_docs_ids_fn: FetchAllDocumentsIdsFunction,  # noqa: ARG001
     callback: IndexingHeartbeatInterface | None = None,
 ) -> Generator[DocExternalAccess, None, None]:
     """

backend/ee/onyx/external_permissions/github/group_sync.py

@@ -12,7 +12,7 @@ logger = setup_logger()
 
 
 def github_group_sync(
-    tenant_id: str,
+    tenant_id: str,  # noqa: ARG001
     cc_pair: ConnectorCredentialPair,
 ) -> Generator[ExternalUserGroup, None, None]:
     github_connector: GithubConnector = GithubConnector(

backend/ee/onyx/external_permissions/github/utils.py

@@ -91,7 +91,7 @@ class TeamInfo(BaseModel):
 
 
 def _fetch_organization_members(
-    github_client: Github, org_name: str, retry_count: int = 0
+    github_client: Github, org_name: str, retry_count: int = 0  # noqa: ARG001
 ) -> List[UserInfo]:
     """Fetch all organization members including owners and regular members."""
     org_members: List[UserInfo] = []
@@ -124,7 +124,7 @@ def _fetch_organization_members(
 
 
 def _fetch_repository_teams_detailed(
-    repo: Repository, github_client: Github, retry_count: int = 0
+    repo: Repository, github_client: Github, retry_count: int = 0  # noqa: ARG001
 ) -> List[TeamInfo]:
     """Fetch teams with access to the repository and their members."""
     teams_data: List[TeamInfo] = []
@@ -167,7 +167,7 @@ def _fetch_repository_teams_detailed(
 
 
 def fetch_repository_team_slugs(
-    repo: Repository, github_client: Github, retry_count: int = 0
+    repo: Repository, github_client: Github, retry_count: int = 0  # noqa: ARG001
 ) -> List[str]:
     """Fetch team slugs with access to the repository."""
     logger.info(f"Fetching team slugs for repository {repo.full_name}")

backend/ee/onyx/external_permissions/gmail/doc_sync.py

@@ -5,8 +5,12 @@ from datetime import timezone
 from ee.onyx.external_permissions.perm_sync_types import FetchAllDocumentsFunction
 from ee.onyx.external_permissions.perm_sync_types import FetchAllDocumentsIdsFunction
 from onyx.access.models import DocExternalAccess
+from onyx.access.models import ElementExternalAccess
+from onyx.access.models import NodeExternalAccess
+from onyx.configs.constants import DocumentSource
 from onyx.connectors.gmail.connector import GmailConnector
 from onyx.connectors.interfaces import GenerateSlimDocumentOutput
+from onyx.connectors.models import HierarchyNode
 from onyx.db.models import ConnectorCredentialPair
 from onyx.indexing.indexing_heartbeat import IndexingHeartbeatInterface
 from onyx.utils.logger import setup_logger
@@ -35,15 +39,15 @@ def _get_slim_doc_generator(
 
 def gmail_doc_sync(
     cc_pair: ConnectorCredentialPair,
-    fetch_all_existing_docs_fn: FetchAllDocumentsFunction,
-    fetch_all_existing_docs_ids_fn: FetchAllDocumentsIdsFunction,
+    fetch_all_existing_docs_fn: FetchAllDocumentsFunction,  # noqa: ARG001
+    fetch_all_existing_docs_ids_fn: FetchAllDocumentsIdsFunction,  # noqa: ARG001
     callback: IndexingHeartbeatInterface | None,
-) -> Generator[DocExternalAccess, None, None]:
+) -> Generator[ElementExternalAccess, None, None]:
     """
-    Adds the external permissions to the documents in postgres
-    if the document doesn't already exists in postgres, we create
+    Adds the external permissions to the documents and hierarchy nodes in postgres.
+    If the document doesn't already exist in postgres, we create
     it in postgres so that when it gets created later, the permissions are
-    already populated
+    already populated.
     """
     gmail_connector = GmailConnector(**cc_pair.connector.connector_specific_config)
     gmail_connector.load_credentials(cc_pair.credential.credential_json)
@@ -60,6 +64,15 @@ def gmail_doc_sync(
 
                 callback.progress("gmail_doc_sync", 1)
 
+            if isinstance(slim_doc, HierarchyNode):
+                # Yield hierarchy node permissions to be processed in outer layer
+                if slim_doc.external_access:
+                    yield NodeExternalAccess(
+                        external_access=slim_doc.external_access,
+                        raw_node_id=slim_doc.raw_node_id,
+                        source=DocumentSource.GMAIL.value,
+                    )
+                continue
             if slim_doc.external_access is None:
                 logger.warning(f"No permissions found for document {slim_doc.id}")
                 continue

backend/ee/onyx/external_permissions/google_drive/doc_sync.py

@@ -10,11 +10,16 @@ from ee.onyx.external_permissions.google_drive.permission_retrieval import (
 from ee.onyx.external_permissions.perm_sync_types import FetchAllDocumentsFunction
 from ee.onyx.external_permissions.perm_sync_types import FetchAllDocumentsIdsFunction
 from onyx.access.models import DocExternalAccess
+from onyx.access.models import ElementExternalAccess
 from onyx.access.models import ExternalAccess
+from onyx.access.models import NodeExternalAccess
+from onyx.access.utils import build_ext_group_name_for_onyx
+from onyx.configs.constants import DocumentSource
 from onyx.connectors.google_drive.connector import GoogleDriveConnector
 from onyx.connectors.google_drive.models import GoogleDriveFileType
 from onyx.connectors.google_utils.resources import GoogleDriveService
 from onyx.connectors.interfaces import GenerateSlimDocumentOutput
+from onyx.connectors.models import HierarchyNode
 from onyx.db.models import ConnectorCredentialPair
 from onyx.indexing.indexing_heartbeat import IndexingHeartbeatInterface
 from onyx.utils.logger import setup_logger
@@ -63,11 +68,17 @@ def get_external_access_for_raw_gdrive_file(
     company_domain: str,
     retriever_drive_service: GoogleDriveService | None,
     admin_drive_service: GoogleDriveService,
+    add_prefix: bool = False,
 ) -> ExternalAccess:
     """
     Get the external access for a raw Google Drive file.
 
     Assumes the file we retrieved has EITHER `permissions` or `permission_ids`
+
+    add_prefix: When this method is called during the initial indexing via the connector,
+                set add_prefix to True so group IDs are prefixed with the source type.
+                When invoked from doc_sync (permission sync), use the default (False)
+                since upsert_document_external_perms handles prefixing.
     """
     doc_id = file.get("id")
     if not doc_id:
@@ -160,6 +171,13 @@ def get_external_access_for_raw_gdrive_file(
         | ({drive_id} if drive_id is not None else set())
     )
 
+    # Prefix group IDs with source type if requested (for indexing path)
+    if add_prefix:
+        group_ids = {
+            build_ext_group_name_for_onyx(group_id, DocumentSource.GOOGLE_DRIVE)
+            for group_id in group_ids
+        }
+
     return ExternalAccess(
         external_user_emails=user_emails,
         external_user_group_ids=group_ids,
@@ -167,17 +185,112 @@ def get_external_access_for_raw_gdrive_file(
     )
 
 
+def get_external_access_for_folder(
+    folder: GoogleDriveFileType,
+    google_domain: str,
+    drive_service: GoogleDriveService,
+    add_prefix: bool = False,
+) -> ExternalAccess:
+    """
+    Extract ExternalAccess from a folder's permissions.
+
+    This fetches permissions using the Drive API (via permissionIds) and extracts
+    user emails, group emails, and public access status.
+
+    Args:
+        folder: The folder metadata from Google Drive API (must include permissionIds field)
+        google_domain: The company's Google Workspace domain (e.g., "company.com")
+        drive_service: Google Drive service for fetching permission details
+        add_prefix: When True, prefix group IDs with source type (for indexing path).
+                   When False (default), leave unprefixed (for permission sync path).
+
+    Returns:
+        ExternalAccess with extracted permission info
+    """
+    folder_id = folder.get("id")
+    if not folder_id:
+        logger.warning("Folder missing ID, returning empty permissions")
+        return ExternalAccess(
+            external_user_emails=set(),
+            external_user_group_ids=set(),
+            is_public=False,
+        )
+
+    # Get permission IDs from folder metadata
+    permission_ids = folder.get("permissionIds") or []
+    if not permission_ids:
+        logger.debug(f"No permissionIds found for folder {folder_id}")
+        return ExternalAccess(
+            external_user_emails=set(),
+            external_user_group_ids=set(),
+            is_public=False,
+        )
+
+    # Fetch full permission objects using the permission IDs
+    permissions_list = get_permissions_by_ids(
+        drive_service=drive_service,
+        doc_id=folder_id,
+        permission_ids=permission_ids,
+    )
+
+    user_emails: set[str] = set()
+    group_emails: set[str] = set()
+    is_public = False
+
+    for permission in permissions_list:
+        if permission.type == PermissionType.USER:
+            if permission.email_address:
+                user_emails.add(permission.email_address)
+            else:
+                logger.warning(f"User permission without email for folder {folder_id}")
+        elif permission.type == PermissionType.GROUP:
+            # Groups are represented as email addresses in Google Drive
+            if permission.email_address:
+                group_emails.add(permission.email_address)
+            else:
+                logger.warning(f"Group permission without email for folder {folder_id}")
+        elif permission.type == PermissionType.DOMAIN:
+            # Domain permission - check if it matches company domain
+            if permission.domain == google_domain:
+                # Only public if discoverable (allowFileDiscovery is not False)
+                # If allowFileDiscovery is False, it's "link only" access
+                is_public = permission.allow_file_discovery is not False
+            else:
+                logger.debug(
+                    f"Domain permission for {permission.domain} does not match "
+                    f"company domain {google_domain} for folder {folder_id}"
+                )
+        elif permission.type == PermissionType.ANYONE:
+            # Only public if discoverable (allowFileDiscovery is not False)
+            # If allowFileDiscovery is False, it's "link only" access
+            is_public = permission.allow_file_discovery is not False
+
+    # Prefix group IDs with source type if requested (for indexing path)
+    group_ids: set[str] = group_emails
+    if add_prefix:
+        group_ids = {
+            build_ext_group_name_for_onyx(group_id, DocumentSource.GOOGLE_DRIVE)
+            for group_id in group_emails
+        }
+
+    return ExternalAccess(
+        external_user_emails=user_emails,
+        external_user_group_ids=group_ids,
+        is_public=is_public,
+    )
+
+
 def gdrive_doc_sync(
     cc_pair: ConnectorCredentialPair,
-    fetch_all_existing_docs_fn: FetchAllDocumentsFunction,
-    fetch_all_existing_docs_ids_fn: FetchAllDocumentsIdsFunction,
+    fetch_all_existing_docs_fn: FetchAllDocumentsFunction,  # noqa: ARG001
+    fetch_all_existing_docs_ids_fn: FetchAllDocumentsIdsFunction,  # noqa: ARG001
     callback: IndexingHeartbeatInterface | None,
-) -> Generator[DocExternalAccess, None, None]:
+) -> Generator[ElementExternalAccess, None, None]:
     """
-    Adds the external permissions to the documents in postgres
-    if the document doesn't already exists in postgres, we create
+    Adds the external permissions to the documents and hierarchy nodes in postgres.
+    If the document doesn't already exist in postgres, we create
     it in postgres so that when it gets created later, the permissions are
-    already populated
+    already populated.
     """
     google_drive_connector = GoogleDriveConnector(
         **cc_pair.connector.connector_specific_config
@@ -195,7 +308,15 @@ def gdrive_doc_sync(
                     raise RuntimeError("gdrive_doc_sync: Stop signal detected")
 
                 callback.progress("gdrive_doc_sync", 1)
-
+            if isinstance(slim_doc, HierarchyNode):
+                # Yield hierarchy node permissions to be processed in outer layer
+                if slim_doc.external_access:
+                    yield NodeExternalAccess(
+                        external_access=slim_doc.external_access,
+                        raw_node_id=slim_doc.raw_node_id,
+                        source=DocumentSource.GOOGLE_DRIVE.value,
+                    )
+                continue
             if slim_doc.external_access is None:
                 raise ValueError(
                     f"Drive perm sync: No external access for document {slim_doc.id}"

backend/ee/onyx/external_permissions/google_drive/group_sync.py

@@ -384,7 +384,7 @@ def _build_onyx_groups(
 
 
 def gdrive_group_sync(
-    tenant_id: str,
+    tenant_id: str,  # noqa: ARG001
     cc_pair: ConnectorCredentialPair,
 ) -> Generator[ExternalUserGroup, None, None]:
     # Initialize connector and build credential/service objects

backend/ee/onyx/external_permissions/google_drive/models.py

@@ -30,6 +30,10 @@ class GoogleDrivePermission(BaseModel):
     type: PermissionType
     domain: str | None  # only applies to domain permissions
     permission_details: GoogleDrivePermissionDetails | None
+    # Whether this permission makes the file discoverable in search
+    # False means "anyone with the link" (not searchable/discoverable)
+    # Only applicable for domain/anyone permission types
+    allow_file_discovery: bool | None
 
     @classmethod
     def from_drive_permission(
@@ -46,6 +50,7 @@ class GoogleDrivePermission(BaseModel):
             email_address=drive_permission.get("emailAddress"),
             type=PermissionType(drive_permission["type"]),
             domain=drive_permission.get("domain"),
+            allow_file_discovery=drive_permission.get("allowFileDiscovery"),
             permission_details=(
                 GoogleDrivePermissionDetails(
                     permission_type=permission_details.get("type"),

backend/ee/onyx/external_permissions/google_drive/permission_retrieval.py

@@ -36,7 +36,7 @@ def get_permissions_by_ids(
         retrieval_function=drive_service.permissions().list,
         list_key="permissions",
         fileId=doc_id,
-        fields="permissions(id, emailAddress, type, domain, permissionDetails),nextPageToken",
+        fields="permissions(id, emailAddress, type, domain, allowFileDiscovery, permissionDetails),nextPageToken",
         supportsAllDrives=True,
         continue_on_404_or_403=True,
     )

backend/ee/onyx/external_permissions/jira/doc_sync.py

@@ -3,7 +3,7 @@ from collections.abc import Generator
 from ee.onyx.external_permissions.perm_sync_types import FetchAllDocumentsFunction
 from ee.onyx.external_permissions.perm_sync_types import FetchAllDocumentsIdsFunction
 from ee.onyx.external_permissions.utils import generic_doc_sync
-from onyx.access.models import DocExternalAccess
+from onyx.access.models import ElementExternalAccess
 from onyx.configs.constants import DocumentSource
 from onyx.connectors.jira.connector import JiraConnector
 from onyx.db.models import ConnectorCredentialPair
@@ -17,10 +17,10 @@ JIRA_DOC_SYNC_TAG = "jira_doc_sync"
 
 def jira_doc_sync(
     cc_pair: ConnectorCredentialPair,
-    fetch_all_existing_docs_fn: FetchAllDocumentsFunction,
+    fetch_all_existing_docs_fn: FetchAllDocumentsFunction,  # noqa: ARG001
     fetch_all_existing_docs_ids_fn: FetchAllDocumentsIdsFunction,
     callback: IndexingHeartbeatInterface | None = None,
-) -> Generator[DocExternalAccess, None, None]:
+) -> Generator[ElementExternalAccess, None, None]:
     jira_connector = JiraConnector(
         **cc_pair.connector.connector_specific_config,
     )

backend/ee/onyx/external_permissions/jira/group_sync.py

@@ -102,7 +102,7 @@ def _build_group_member_email_map(
 
 
 def jira_group_sync(
-    tenant_id: str,
+    tenant_id: str,  # noqa: ARG001
     cc_pair: ConnectorCredentialPair,
 ) -> Generator[ExternalUserGroup, None, None]:
     """

backend/ee/onyx/external_permissions/jira/page_access.py

@@ -8,6 +8,8 @@ from ee.onyx.external_permissions.jira.models import Holder
 from ee.onyx.external_permissions.jira.models import Permission
 from ee.onyx.external_permissions.jira.models import User
 from onyx.access.models import ExternalAccess
+from onyx.access.utils import build_ext_group_name_for_onyx
+from onyx.configs.constants import DocumentSource
 from onyx.utils.logger import setup_logger
 
 HolderMap = dict[str, list[Holder]]
@@ -252,7 +254,14 @@ def _build_external_access_from_holder_map(
 def get_project_permissions(
     jira_client: JIRA,
     jira_project: str,
+    add_prefix: bool = False,
 ) -> ExternalAccess | None:
+    """
+    Get project permissions from Jira.
+
+    add_prefix: When True, prefix group IDs with source type (for indexing path).
+               When False (default), leave unprefixed (for permission sync path).
+    """
     project_permissions: PermissionScheme = jira_client.project_permissionscheme(
         project=jira_project
     )
@@ -267,6 +276,20 @@ def get_project_permissions(
 
     holder_map = _build_holder_map(permissions=project_permissions.permissions)
 
-    return _build_external_access_from_holder_map(
+    external_access = _build_external_access_from_holder_map(
         jira_client=jira_client, jira_project=jira_project, holder_map=holder_map
     )
+
+    # Prefix group IDs with source type if requested (for indexing path)
+    if add_prefix and external_access and external_access.external_user_group_ids:
+        prefixed_groups = {
+            build_ext_group_name_for_onyx(g, DocumentSource.JIRA)
+            for g in external_access.external_user_group_ids
+        }
+        return ExternalAccess(
+            external_user_emails=external_access.external_user_emails,
+            external_user_group_ids=prefixed_groups,
+            is_public=external_access.is_public,
+        )
+
+    return external_access

backend/ee/onyx/external_permissions/perm_sync_types.py

@@ -5,6 +5,8 @@ from typing import Protocol
 
 from ee.onyx.db.external_perm import ExternalUserGroup  # noqa
 from onyx.access.models import DocExternalAccess  # noqa
+from onyx.access.models import ElementExternalAccess  # noqa
+from onyx.access.models import NodeExternalAccess  # noqa
 from onyx.context.search.models import InferenceChunk
 from onyx.db.models import ConnectorCredentialPair  # noqa
 from onyx.db.utils import DocumentRow
@@ -53,7 +55,7 @@ DocSyncFuncType = Callable[
         FetchAllDocumentsIdsFunction,
         Optional[IndexingHeartbeatInterface],
     ],
-    Generator[DocExternalAccess, None, None],
+    Generator[ElementExternalAccess, None, None],
 ]
 
 GroupSyncFuncType = Callable[

backend/ee/onyx/external_permissions/post_query_censoring.py

@@ -34,21 +34,21 @@ def _get_all_censoring_enabled_sources() -> set[DocumentSource]:
 # NOTE: This is only called if ee is enabled.
 def _post_query_chunk_censoring(
     chunks: list[InferenceChunk],
-    user: User | None,
+    user: User,
 ) -> list[InferenceChunk]:
     """
     This function checks all chunks to see if they need to be sent to a censoring
     function. If they do, it sends them to the censoring function and returns the
     censored chunks. If they don't, it returns the original chunks.
     """
-    if user is None:
-        # if user is None, permissions are not enforced
-        return chunks
+    sources_to_censor = _get_all_censoring_enabled_sources()
+
+    # Anonymous users can only access public (non-permission-synced) content
+    if user.is_anonymous:
+        return [chunk for chunk in chunks if chunk.source_type not in sources_to_censor]
 
     final_chunk_dict: dict[str, InferenceChunk] = {}
     chunks_to_process: dict[DocumentSource, list[InferenceChunk]] = {}
-
-    sources_to_censor = _get_all_censoring_enabled_sources()
     for chunk in chunks:
         # Separate out chunks that require permission post-processing by source
         if chunk.source_type in sources_to_censor:

backend/ee/onyx/external_permissions/salesforce/postprocessing.py

@@ -23,7 +23,7 @@ ContentRange = tuple[int, int | None]  # (start_index, end_index) None means to
 
 # NOTE: Used for testing timing
 def _get_dummy_object_access_map(
-    object_ids: set[str], user_email: str, chunks: list[InferenceChunk]
+    object_ids: set[str], user_email: str, chunks: list[InferenceChunk]  # noqa: ARG001
 ) -> dict[str, bool]:
     time.sleep(0.15)
     # return {object_id: True for object_id in object_ids}

backend/ee/onyx/external_permissions/sharepoint/doc_sync.py

@@ -3,7 +3,7 @@ from collections.abc import Generator
 from ee.onyx.external_permissions.perm_sync_types import FetchAllDocumentsFunction
 from ee.onyx.external_permissions.perm_sync_types import FetchAllDocumentsIdsFunction
 from ee.onyx.external_permissions.utils import generic_doc_sync
-from onyx.access.models import DocExternalAccess
+from onyx.access.models import ElementExternalAccess
 from onyx.configs.constants import DocumentSource
 from onyx.connectors.sharepoint.connector import SharepointConnector
 from onyx.db.models import ConnectorCredentialPair
@@ -17,10 +17,10 @@ SHAREPOINT_DOC_SYNC_TAG = "sharepoint_doc_sync"
 
 def sharepoint_doc_sync(
     cc_pair: ConnectorCredentialPair,
-    fetch_all_existing_docs_fn: FetchAllDocumentsFunction,
+    fetch_all_existing_docs_fn: FetchAllDocumentsFunction,  # noqa: ARG001
     fetch_all_existing_docs_ids_fn: FetchAllDocumentsIdsFunction,
     callback: IndexingHeartbeatInterface | None = None,
-) -> Generator[DocExternalAccess, None, None]:
+) -> Generator[ElementExternalAccess, None, None]:
     sharepoint_connector = SharepointConnector(
         **cc_pair.connector.connector_specific_config,
     )

backend/ee/onyx/external_permissions/sharepoint/group_sync.py

@@ -15,7 +15,7 @@ logger = setup_logger()
 
 
 def sharepoint_group_sync(
-    tenant_id: str,
+    tenant_id: str,  # noqa: ARG001
     cc_pair: ConnectorCredentialPair,
 ) -> Generator[ExternalUserGroup, None, None]:
     """Sync SharePoint groups and their members"""

backend/ee/onyx/external_permissions/slack/doc_sync.py

@@ -8,6 +8,7 @@ from ee.onyx.external_permissions.slack.utils import fetch_user_id_to_email_map
 from onyx.access.models import DocExternalAccess
 from onyx.access.models import ExternalAccess
 from onyx.connectors.credentials_provider import OnyxDBCredentialsProvider
+from onyx.connectors.models import HierarchyNode
 from onyx.connectors.slack.connector import get_channels
 from onyx.connectors.slack.connector import make_paginated_slack_api_call
 from onyx.connectors.slack.connector import SlackConnector
@@ -102,7 +103,7 @@ def _fetch_channel_permissions(
 
 def _get_slack_document_access(
     slack_connector: SlackConnector,
-    channel_permissions: dict[str, ExternalAccess],
+    channel_permissions: dict[str, ExternalAccess],  # noqa: ARG001
     callback: IndexingHeartbeatInterface | None,
 ) -> Generator[DocExternalAccess, None, None]:
     slim_doc_generator = slack_connector.retrieve_all_slim_docs_perm_sync(
@@ -111,6 +112,9 @@ def _get_slack_document_access(
 
     for doc_metadata_batch in slim_doc_generator:
         for doc_metadata in doc_metadata_batch:
+            if isinstance(doc_metadata, HierarchyNode):
+                # TODO: handle hierarchynodes during sync
+                continue
             if doc_metadata.external_access is None:
                 raise ValueError(
                     f"No external access for document {doc_metadata.id}. "
@@ -132,8 +136,8 @@ def _get_slack_document_access(
 
 def slack_doc_sync(
     cc_pair: ConnectorCredentialPair,
-    fetch_all_existing_docs_fn: FetchAllDocumentsFunction,
-    fetch_all_existing_docs_ids_fn: FetchAllDocumentsIdsFunction,
+    fetch_all_existing_docs_fn: FetchAllDocumentsFunction,  # noqa: ARG001
+    fetch_all_existing_docs_ids_fn: FetchAllDocumentsIdsFunction,  # noqa: ARG001
     callback: IndexingHeartbeatInterface | None,
 ) -> Generator[DocExternalAccess, None, None]:
     """

backend/ee/onyx/external_permissions/sync_params.py

@@ -72,10 +72,10 @@ class SyncConfig(BaseModel):
 
 # Mock doc sync function for testing (no-op)
 def mock_doc_sync(
-    cc_pair: "ConnectorCredentialPair",
-    fetch_all_docs_fn: FetchAllDocumentsFunction,
-    fetch_all_docs_ids_fn: FetchAllDocumentsIdsFunction,
-    callback: Optional["IndexingHeartbeatInterface"],
+    cc_pair: "ConnectorCredentialPair",  # noqa: ARG001
+    fetch_all_docs_fn: FetchAllDocumentsFunction,  # noqa: ARG001
+    fetch_all_docs_ids_fn: FetchAllDocumentsIdsFunction,  # noqa: ARG001
+    callback: Optional["IndexingHeartbeatInterface"],  # noqa: ARG001
 ) -> Generator["DocExternalAccess", None, None]:
     """Mock doc sync function for testing - returns empty list since permissions are fetched during indexing"""
     yield from []

backend/ee/onyx/external_permissions/teams/doc_sync.py

@@ -3,7 +3,7 @@ from collections.abc import Generator
 from ee.onyx.external_permissions.perm_sync_types import FetchAllDocumentsFunction
 from ee.onyx.external_permissions.perm_sync_types import FetchAllDocumentsIdsFunction
 from ee.onyx.external_permissions.utils import generic_doc_sync
-from onyx.access.models import DocExternalAccess
+from onyx.access.models import ElementExternalAccess
 from onyx.configs.constants import DocumentSource
 from onyx.connectors.teams.connector import TeamsConnector
 from onyx.db.models import ConnectorCredentialPair
@@ -18,10 +18,10 @@ TEAMS_DOC_SYNC_LABEL = "teams_doc_sync"
 
 def teams_doc_sync(
     cc_pair: ConnectorCredentialPair,
-    fetch_all_existing_docs_fn: FetchAllDocumentsFunction,
+    fetch_all_existing_docs_fn: FetchAllDocumentsFunction,  # noqa: ARG001
     fetch_all_existing_docs_ids_fn: FetchAllDocumentsIdsFunction,
     callback: IndexingHeartbeatInterface | None,
-) -> Generator[DocExternalAccess, None, None]:
+) -> Generator[ElementExternalAccess, None, None]:
     teams_connector = TeamsConnector(
         **cc_pair.connector.connector_specific_config,
     )

backend/ee/onyx/external_permissions/utils.py

@@ -2,9 +2,12 @@ from collections.abc import Generator
 
 from ee.onyx.external_permissions.perm_sync_types import FetchAllDocumentsIdsFunction
 from onyx.access.models import DocExternalAccess
+from onyx.access.models import ElementExternalAccess
 from onyx.access.models import ExternalAccess
+from onyx.access.models import NodeExternalAccess
 from onyx.configs.constants import DocumentSource
 from onyx.connectors.interfaces import SlimConnectorWithPermSync
+from onyx.connectors.models import HierarchyNode
 from onyx.db.models import ConnectorCredentialPair
 from onyx.indexing.indexing_heartbeat import IndexingHeartbeatInterface
 from onyx.utils.logger import setup_logger
@@ -19,7 +22,7 @@ def generic_doc_sync(
     doc_source: DocumentSource,
     slim_connector: SlimConnectorWithPermSync,
     label: str,
-) -> Generator[DocExternalAccess, None, None]:
+) -> Generator[ElementExternalAccess, None, None]:
     """
     A convenience function for performing a generic document synchronization.
 
@@ -29,7 +32,7 @@ def generic_doc_sync(
         - fetching *all* new (slim) docs
         - yielding external-access permissions for existing docs which do not exist in the newly fetched slim-docs set (with their
         `external_access` set to "private")
-        - yielding external-access permissions for newly fetched docs
+        - yielding external-access permissions for newly fetched docs and hierarchy nodes
 
     Returns:
         A `Generator` which yields existing and newly fetched external-access permissions.
@@ -49,6 +52,15 @@ def generic_doc_sync(
             callback.progress(label, 1)
 
         for doc in doc_batch:
+            if isinstance(doc, HierarchyNode):
+                # Yield hierarchy node permissions to be processed in outer layer
+                if doc.external_access:
+                    yield NodeExternalAccess(
+                        external_access=doc.external_access,
+                        raw_node_id=doc.raw_node_id,
+                        source=doc_source.value,
+                    )
+                continue
             if not doc.external_access:
                 raise RuntimeError(
                     f"No external access found for document ID; {cc_pair.id=} {doc_source=} {doc.id=}"

backend/ee/onyx/main.py

@@ -4,8 +4,10 @@ from contextlib import asynccontextmanager
 from fastapi import FastAPI
 from httpx_oauth.clients.google import GoogleOAuth2
 
+from ee.onyx.configs.app_configs import LICENSE_ENFORCEMENT_ENABLED
 from ee.onyx.server.analytics.api import router as analytics_router
 from ee.onyx.server.auth_check import check_ee_router_auth
+from ee.onyx.server.billing.api import router as billing_router
 from ee.onyx.server.documents.cc_pair import router as ee_document_cc_pair_router
 from ee.onyx.server.enterprise_settings.api import (
     admin_router as enterprise_settings_admin_router,
@@ -85,10 +87,11 @@ def get_application() -> FastAPI:
 
     if MULTI_TENANT:
         add_api_server_tenant_id_middleware(application, logger)
-
-    # Add license enforcement middleware (runs after tenant tracking)
-    # This blocks access when license is expired/gated
-    add_license_enforcement_middleware(application, logger)
+    else:
+        # License enforcement middleware for self-hosted deployments only
+        # Checks LICENSE_ENFORCEMENT_ENABLED at runtime (can be toggled without restart)
+        # MT deployments use control plane gating via is_tenant_gated() instead
+        add_license_enforcement_middleware(application, logger)
 
     if AUTH_TYPE == AuthType.CLOUD:
         # For Google OAuth, refresh tokens are requested by:
@@ -148,6 +151,13 @@ def get_application() -> FastAPI:
     # License management
     include_router_with_global_prefix_prepended(application, license_router)
 
+    # Unified billing API - available when license system is enabled
+    # Works for both self-hosted and cloud deployments
+    # TODO(ENG-3533): Once frontend migrates to /admin/billing/*, this becomes the
+    # primary billing API and /tenants/* billing endpoints can be removed
+    if LICENSE_ENFORCEMENT_ENABLED:
+        include_router_with_global_prefix_prepended(application, billing_router)
+
     if MULTI_TENANT:
         # Tenant management
         include_router_with_global_prefix_prepended(application, tenants_router)

backend/ee/onyx/search/process_search_query.py

@@ -41,7 +41,7 @@ def _run_single_search(
     query: str,
     filters: BaseFilters | None,
     document_index: DocumentIndex,
-    user: User | None,
+    user: User,
     db_session: Session,
     num_hits: int | None = None,
 ) -> list[InferenceChunk]:
@@ -63,7 +63,7 @@ def _run_single_search(
 
 def stream_search_query(
     request: SendSearchQueryRequest,
-    user: User | None,
+    user: User,
     db_session: Session,
 ) -> Generator[
     SearchQueriesPacket | SearchDocsPacket | LLMSelectedDocsPacket | SearchErrorPacket,
@@ -101,8 +101,7 @@ def stream_search_query(
     # Build list of all executed queries for tracking
     all_executed_queries = [original_query] + keyword_expansions
 
-    # TODO remove this check, user should not be None
-    if user is not None:
+    if not user.is_anonymous:
         create_search_query(
             db_session=db_session,
             user_id=user.id,

backend/ee/onyx/server/analytics/api.py

@@ -40,7 +40,7 @@ class QueryAnalyticsResponse(BaseModel):
 def get_query_analytics(
     start: datetime.datetime | None = None,
     end: datetime.datetime | None = None,
-    _: User | None = Depends(current_admin_user),
+    _: User = Depends(current_admin_user),
     db_session: Session = Depends(get_session),
 ) -> list[QueryAnalyticsResponse]:
     daily_query_usage_info = fetch_query_analytics(
@@ -71,7 +71,7 @@ class UserAnalyticsResponse(BaseModel):
 def get_user_analytics(
     start: datetime.datetime | None = None,
     end: datetime.datetime | None = None,
-    _: User | None = Depends(current_admin_user),
+    _: User = Depends(current_admin_user),
     db_session: Session = Depends(get_session),
 ) -> list[UserAnalyticsResponse]:
     daily_query_usage_info_per_user = fetch_per_user_query_analytics(
@@ -105,7 +105,7 @@ class OnyxbotAnalyticsResponse(BaseModel):
 def get_onyxbot_analytics(
     start: datetime.datetime | None = None,
     end: datetime.datetime | None = None,
-    _: User | None = Depends(current_admin_user),
+    _: User = Depends(current_admin_user),
     db_session: Session = Depends(get_session),
 ) -> list[OnyxbotAnalyticsResponse]:
     daily_onyxbot_info = fetch_onyxbot_analytics(
@@ -141,7 +141,7 @@ def get_persona_messages(
     persona_id: int,
     start: datetime.datetime | None = None,
     end: datetime.datetime | None = None,
-    _: User | None = Depends(current_admin_user),
+    _: User = Depends(current_admin_user),
     db_session: Session = Depends(get_session),
 ) -> list[PersonaMessageAnalyticsResponse]:
     """Fetch daily message counts for a single persona within the given time range."""
@@ -179,7 +179,7 @@ def get_persona_unique_users(
     persona_id: int,
     start: datetime.datetime,
     end: datetime.datetime,
-    _: User | None = Depends(current_admin_user),
+    _: User = Depends(current_admin_user),
     db_session: Session = Depends(get_session),
 ) -> list[PersonaUniqueUsersResponse]:
     """Get unique users per day for a single persona."""
@@ -218,7 +218,7 @@ def get_assistant_stats(
     assistant_id: int,
     start: datetime.datetime | None = None,
     end: datetime.datetime | None = None,
-    user: User | None = Depends(current_user),
+    user: User = Depends(current_user),
     db_session: Session = Depends(get_session),
 ) -> AssistantStatsResponse:
     """

backend/ee/onyx/server/auth_check.py

@@ -12,6 +12,14 @@ EE_PUBLIC_ENDPOINT_SPECS = PUBLIC_ENDPOINT_SPECS + [
     ("/enterprise-settings/custom-analytics-script", {"GET"}),
     # Stripe publishable key is safe to expose publicly
     ("/tenants/stripe-publishable-key", {"GET"}),
+    ("/admin/billing/stripe-publishable-key", {"GET"}),
+    # Proxy endpoints use license-based auth, not user auth
+    ("/proxy/create-checkout-session", {"POST"}),
+    ("/proxy/claim-license", {"POST"}),
+    ("/proxy/create-customer-portal-session", {"POST"}),
+    ("/proxy/billing-information", {"GET"}),
+    ("/proxy/license/{tenant_id}", {"GET"}),
+    ("/proxy/seats/update", {"POST"}),
 ]
 
 

backend/ee/onyx/server/billing/api.py

@@ -0,0 +1,393 @@
+"""Unified Billing API endpoints.
+
+These endpoints provide Stripe billing functionality for both cloud and
+self-hosted deployments. The service layer routes requests appropriately:
+
+- Self-hosted: Routes through cloud data plane proxy
+  Flow: Backend /admin/billing/* → Cloud DP /proxy/* → Control plane
+
+- Cloud (MULTI_TENANT): Routes directly to control plane
+  Flow: Backend /admin/billing/* → Control plane
+
+License claiming is handled separately by /license/claim endpoint (self-hosted only).
+
+Migration Note (ENG-3533):
+This /admin/billing/* API replaces the older /tenants/* billing endpoints:
+- /tenants/billing-information            -> /admin/billing/billing-information
+- /tenants/create-customer-portal-session -> /admin/billing/create-customer-portal-session
+- /tenants/create-subscription-session    -> /admin/billing/create-checkout-session
+- /tenants/stripe-publishable-key         -> /admin/billing/stripe-publishable-key
+
+See: https://linear.app/onyx-app/issue/ENG-3533/migrate-tenantsbilling-adminbilling
+"""
+
+import asyncio
+
+import httpx
+from fastapi import APIRouter
+from fastapi import Depends
+from fastapi import HTTPException
+from pydantic import BaseModel
+from sqlalchemy.orm import Session
+
+from ee.onyx.auth.users import current_admin_user
+from ee.onyx.db.license import get_license
+from ee.onyx.db.license import get_used_seats
+from ee.onyx.server.billing.models import BillingInformationResponse
+from ee.onyx.server.billing.models import CreateCheckoutSessionRequest
+from ee.onyx.server.billing.models import CreateCheckoutSessionResponse
+from ee.onyx.server.billing.models import CreateCustomerPortalSessionRequest
+from ee.onyx.server.billing.models import CreateCustomerPortalSessionResponse
+from ee.onyx.server.billing.models import SeatUpdateRequest
+from ee.onyx.server.billing.models import SeatUpdateResponse
+from ee.onyx.server.billing.models import StripePublishableKeyResponse
+from ee.onyx.server.billing.models import SubscriptionStatusResponse
+from ee.onyx.server.billing.service import BillingServiceError
+from ee.onyx.server.billing.service import (
+    create_checkout_session as create_checkout_service,
+)
+from ee.onyx.server.billing.service import (
+    create_customer_portal_session as create_portal_service,
+)
+from ee.onyx.server.billing.service import (
+    get_billing_information as get_billing_service,
+)
+from ee.onyx.server.billing.service import update_seat_count as update_seat_service
+from onyx.auth.users import User
+from onyx.configs.app_configs import STRIPE_PUBLISHABLE_KEY_OVERRIDE
+from onyx.configs.app_configs import STRIPE_PUBLISHABLE_KEY_URL
+from onyx.configs.app_configs import WEB_DOMAIN
+from onyx.db.engine.sql_engine import get_session
+from onyx.redis.redis_pool import get_shared_redis_client
+from onyx.utils.logger import setup_logger
+from shared_configs.configs import MULTI_TENANT
+from shared_configs.contextvars import get_current_tenant_id
+
+logger = setup_logger()
+
+router = APIRouter(prefix="/admin/billing")
+
+# Cache for Stripe publishable key to avoid hitting S3 on every request
+_stripe_publishable_key_cache: str | None = None
+_stripe_key_lock = asyncio.Lock()
+
+# Redis key for billing circuit breaker (self-hosted only)
+# When set, billing requests to Stripe are disabled until user manually retries
+BILLING_CIRCUIT_BREAKER_KEY = "billing_circuit_open"
+# Circuit breaker auto-expires after 1 hour (user can manually retry sooner)
+BILLING_CIRCUIT_BREAKER_TTL_SECONDS = 3600
+
+
+def _is_billing_circuit_open() -> bool:
+    """Check if the billing circuit breaker is open (self-hosted only)."""
+    if MULTI_TENANT:
+        return False
+    try:
+        redis_client = get_shared_redis_client()
+        is_open = bool(redis_client.exists(BILLING_CIRCUIT_BREAKER_KEY))
+        logger.debug(
+            f"Circuit breaker check: key={BILLING_CIRCUIT_BREAKER_KEY}, is_open={is_open}"
+        )
+        return is_open
+    except Exception as e:
+        logger.error(f"Failed to check circuit breaker: {e}")
+        return False
+
+
+def _open_billing_circuit() -> None:
+    """Open the billing circuit breaker after a failure (self-hosted only)."""
+    if MULTI_TENANT:
+        return
+    try:
+        redis_client = get_shared_redis_client()
+        redis_client.set(
+            BILLING_CIRCUIT_BREAKER_KEY,
+            "1",
+            ex=BILLING_CIRCUIT_BREAKER_TTL_SECONDS,
+        )
+        # Verify it was set
+        exists = redis_client.exists(BILLING_CIRCUIT_BREAKER_KEY)
+        logger.warning(
+            f"Billing circuit breaker opened (TTL={BILLING_CIRCUIT_BREAKER_TTL_SECONDS}s, "
+            f"verified={exists}). Stripe billing requests are disabled until manually reset."
+        )
+    except Exception as e:
+        logger.error(f"Failed to open circuit breaker: {e}")
+
+
+def _close_billing_circuit() -> None:
+    """Close the billing circuit breaker (re-enable Stripe requests)."""
+    if MULTI_TENANT:
+        return
+    try:
+        redis_client = get_shared_redis_client()
+        redis_client.delete(BILLING_CIRCUIT_BREAKER_KEY)
+        logger.info(
+            "Billing circuit breaker closed. Stripe billing requests re-enabled."
+        )
+    except Exception as e:
+        logger.error(f"Failed to close circuit breaker: {e}")
+
+
+def _get_license_data(db_session: Session) -> str | None:
+    """Get license data from database if exists (self-hosted only)."""
+    if MULTI_TENANT:
+        return None
+    license_record = get_license(db_session)
+    return license_record.license_data if license_record else None
+
+
+def _get_tenant_id() -> str | None:
+    """Get tenant ID for cloud deployments."""
+    if MULTI_TENANT:
+        return get_current_tenant_id()
+    return None
+
+
+@router.post("/create-checkout-session")
+async def create_checkout_session(
+    request: CreateCheckoutSessionRequest | None = None,
+    _: User = Depends(current_admin_user),
+    db_session: Session = Depends(get_session),
+) -> CreateCheckoutSessionResponse:
+    """Create a Stripe checkout session for new subscription or renewal.
+
+    For new customers, no license/tenant is required.
+    For renewals, existing license (self-hosted) or tenant_id (cloud) is used.
+
+    After checkout completion:
+    - Self-hosted: Use /license/claim to retrieve the license
+    - Cloud: Subscription is automatically activated
+    """
+    license_data = _get_license_data(db_session)
+    tenant_id = _get_tenant_id()
+    billing_period = request.billing_period if request else "monthly"
+    seats = request.seats if request else None
+    email = request.email if request else None
+
+    # Validate that requested seats is not less than current used seats
+    if seats is not None:
+        used_seats = get_used_seats(tenant_id)
+        if seats < used_seats:
+            raise HTTPException(
+                status_code=400,
+                detail=f"Cannot subscribe with fewer seats than current usage. "
+                f"You have {used_seats} active users/integrations but requested {seats} seats.",
+            )
+
+    # Build redirect URL for after checkout completion
+    redirect_url = f"{WEB_DOMAIN}/admin/billing?checkout=success"
+
+    try:
+        return await create_checkout_service(
+            billing_period=billing_period,
+            seats=seats,
+            email=email,
+            license_data=license_data,
+            redirect_url=redirect_url,
+            tenant_id=tenant_id,
+        )
+    except BillingServiceError as e:
+        raise HTTPException(status_code=e.status_code, detail=e.message)
+
+
+@router.post("/create-customer-portal-session")
+async def create_customer_portal_session(
+    request: CreateCustomerPortalSessionRequest | None = None,
+    _: User = Depends(current_admin_user),
+    db_session: Session = Depends(get_session),
+) -> CreateCustomerPortalSessionResponse:
+    """Create a Stripe customer portal session for managing subscription.
+
+    Requires existing license (self-hosted) or active tenant (cloud).
+    """
+    license_data = _get_license_data(db_session)
+    tenant_id = _get_tenant_id()
+
+    # Self-hosted requires license
+    if not MULTI_TENANT and not license_data:
+        raise HTTPException(status_code=400, detail="No license found")
+
+    return_url = request.return_url if request else f"{WEB_DOMAIN}/admin/billing"
+
+    try:
+        return await create_portal_service(
+            license_data=license_data,
+            return_url=return_url,
+            tenant_id=tenant_id,
+        )
+    except BillingServiceError as e:
+        raise HTTPException(status_code=e.status_code, detail=e.message)
+
+
+@router.get("/billing-information")
+async def get_billing_information(
+    _: User = Depends(current_admin_user),
+    db_session: Session = Depends(get_session),
+) -> BillingInformationResponse | SubscriptionStatusResponse:
+    """Get billing information for the current subscription.
+
+    Returns subscription status and details from Stripe.
+    For self-hosted: If the circuit breaker is open (previous failure),
+    returns a 503 error without making the request.
+    """
+    license_data = _get_license_data(db_session)
+    tenant_id = _get_tenant_id()
+
+    # Self-hosted without license = no subscription
+    if not MULTI_TENANT and not license_data:
+        return SubscriptionStatusResponse(subscribed=False)
+
+    # Check circuit breaker (self-hosted only)
+    if _is_billing_circuit_open():
+        raise HTTPException(
+            status_code=503,
+            detail="Stripe connection temporarily disabled. Click 'Connect to Stripe' to retry.",
+        )
+
+    try:
+        return await get_billing_service(
+            license_data=license_data,
+            tenant_id=tenant_id,
+        )
+    except BillingServiceError as e:
+        # Open circuit breaker on connection failures (self-hosted only)
+        if e.status_code in (502, 503, 504):
+            _open_billing_circuit()
+        raise HTTPException(status_code=e.status_code, detail=e.message)
+
+
+@router.post("/seats/update")
+async def update_seats(
+    request: SeatUpdateRequest,
+    _: User = Depends(current_admin_user),
+    db_session: Session = Depends(get_session),
+) -> SeatUpdateResponse:
+    """Update the seat count for the current subscription.
+
+    Handles Stripe proration and license regeneration via control plane.
+    For self-hosted, the frontend should call /license/claim after a short delay
+    to fetch the regenerated license.
+    """
+    license_data = _get_license_data(db_session)
+    tenant_id = _get_tenant_id()
+
+    # Self-hosted requires license
+    if not MULTI_TENANT and not license_data:
+        raise HTTPException(status_code=400, detail="No license found")
+
+    # Validate that new seat count is not less than current used seats
+    used_seats = get_used_seats(tenant_id)
+    if request.new_seat_count < used_seats:
+        raise HTTPException(
+            status_code=400,
+            detail=f"Cannot reduce seats below current usage. "
+            f"You have {used_seats} active users/integrations but requested {request.new_seat_count} seats.",
+        )
+
+    try:
+        result = await update_seat_service(
+            new_seat_count=request.new_seat_count,
+            license_data=license_data,
+            tenant_id=tenant_id,
+        )
+
+        # Note: Don't store license here - the control plane may still be processing
+        # the subscription update. The frontend should call /license/claim after a
+        # short delay to get the freshly generated license.
+
+        return result
+    except BillingServiceError as e:
+        raise HTTPException(status_code=e.status_code, detail=e.message)
+
+
+@router.get("/stripe-publishable-key")
+async def get_stripe_publishable_key() -> StripePublishableKeyResponse:
+    """Fetch the Stripe publishable key.
+
+    Priority: env var override (for testing) > S3 bucket (production).
+    This endpoint is public (no auth required) since publishable keys are safe to expose.
+    The key is cached in memory to avoid hitting S3 on every request.
+    """
+    global _stripe_publishable_key_cache
+
+    # Fast path: return cached value without lock
+    if _stripe_publishable_key_cache:
+        return StripePublishableKeyResponse(
+            publishable_key=_stripe_publishable_key_cache
+        )
+
+    # Use lock to prevent concurrent S3 requests
+    async with _stripe_key_lock:
+        # Double-check after acquiring lock (another request may have populated cache)
+        if _stripe_publishable_key_cache:
+            return StripePublishableKeyResponse(
+                publishable_key=_stripe_publishable_key_cache
+            )
+
+        # Check for env var override first (for local testing with pk_test_* keys)
+        if STRIPE_PUBLISHABLE_KEY_OVERRIDE:
+            key = STRIPE_PUBLISHABLE_KEY_OVERRIDE.strip()
+            if not key.startswith("pk_"):
+                raise HTTPException(
+                    status_code=500,
+                    detail="Invalid Stripe publishable key format",
+                )
+            _stripe_publishable_key_cache = key
+            return StripePublishableKeyResponse(publishable_key=key)
+
+        # Fall back to S3 bucket
+        if not STRIPE_PUBLISHABLE_KEY_URL:
+            raise HTTPException(
+                status_code=500,
+                detail="Stripe publishable key is not configured",
+            )
+
+        try:
+            async with httpx.AsyncClient() as client:
+                response = await client.get(STRIPE_PUBLISHABLE_KEY_URL)
+                response.raise_for_status()
+                key = response.text.strip()
+
+                # Validate key format
+                if not key.startswith("pk_"):
+                    raise HTTPException(
+                        status_code=500,
+                        detail="Invalid Stripe publishable key format",
+                    )
+
+                _stripe_publishable_key_cache = key
+                return StripePublishableKeyResponse(publishable_key=key)
+        except httpx.HTTPError:
+            raise HTTPException(
+                status_code=500,
+                detail="Failed to fetch Stripe publishable key",
+            )
+
+
+class ResetConnectionResponse(BaseModel):
+    success: bool
+    message: str
+
+
+@router.post("/reset-connection")
+async def reset_stripe_connection(
+    _: User = Depends(current_admin_user),
+) -> ResetConnectionResponse:
+    """Reset the Stripe connection circuit breaker.
+
+    Called when user clicks "Connect to Stripe" to retry after a previous failure.
+    This clears the circuit breaker flag, allowing billing requests to proceed again.
+    Self-hosted only - cloud deployments don't use the circuit breaker.
+    """
+    if MULTI_TENANT:
+        return ResetConnectionResponse(
+            success=True,
+            message="Circuit breaker not applicable for cloud deployments",
+        )
+
+    _close_billing_circuit()
+    return ResetConnectionResponse(
+        success=True,
+        message="Stripe connection reset. Billing requests re-enabled.",
+    )

backend/ee/onyx/server/billing/models.py

@@ -0,0 +1,77 @@
+"""Pydantic models for the billing API."""
+
+from datetime import datetime
+from typing import Literal
+
+from pydantic import BaseModel
+
+
+class CreateCheckoutSessionRequest(BaseModel):
+    """Request to create a Stripe checkout session."""
+
+    billing_period: Literal["monthly", "annual"] = "monthly"
+    seats: int | None = None
+    email: str | None = None
+
+
+class CreateCheckoutSessionResponse(BaseModel):
+    """Response containing the Stripe checkout session URL."""
+
+    stripe_checkout_url: str
+
+
+class CreateCustomerPortalSessionRequest(BaseModel):
+    """Request to create a Stripe customer portal session."""
+
+    return_url: str | None = None
+
+
+class CreateCustomerPortalSessionResponse(BaseModel):
+    """Response containing the Stripe customer portal URL."""
+
+    stripe_customer_portal_url: str
+
+
+class BillingInformationResponse(BaseModel):
+    """Billing information for the current subscription."""
+
+    tenant_id: str
+    status: str | None = None
+    plan_type: str | None = None
+    seats: int | None = None
+    billing_period: str | None = None
+    current_period_start: datetime | None = None
+    current_period_end: datetime | None = None
+    cancel_at_period_end: bool = False
+    canceled_at: datetime | None = None
+    trial_start: datetime | None = None
+    trial_end: datetime | None = None
+    payment_method_enabled: bool = False
+
+
+class SubscriptionStatusResponse(BaseModel):
+    """Response when no subscription exists."""
+
+    subscribed: bool = False
+
+
+class SeatUpdateRequest(BaseModel):
+    """Request to update seat count."""
+
+    new_seat_count: int
+
+
+class SeatUpdateResponse(BaseModel):
+    """Response from seat update operation."""
+
+    success: bool
+    current_seats: int
+    used_seats: int
+    message: str | None = None
+    license: str | None = None  # Regenerated license (self-hosted stores this)
+
+
+class StripePublishableKeyResponse(BaseModel):
+    """Response containing the Stripe publishable key."""
+
+    publishable_key: str

backend/ee/onyx/server/billing/service.py

@@ -0,0 +1,273 @@
+"""Service layer for billing operations.
+
+This module provides functions for billing operations that route differently
+based on deployment type:
+
+- Self-hosted (not MULTI_TENANT): Routes through cloud data plane proxy
+  Flow: Self-hosted backend → Cloud DP /proxy/* → Control plane
+
+- Cloud (MULTI_TENANT): Routes directly to control plane
+  Flow: Cloud backend → Control plane
+"""
+
+from typing import Literal
+
+import httpx
+
+from ee.onyx.configs.app_configs import CLOUD_DATA_PLANE_URL
+from ee.onyx.server.billing.models import BillingInformationResponse
+from ee.onyx.server.billing.models import CreateCheckoutSessionResponse
+from ee.onyx.server.billing.models import CreateCustomerPortalSessionResponse
+from ee.onyx.server.billing.models import SeatUpdateResponse
+from ee.onyx.server.billing.models import SubscriptionStatusResponse
+from ee.onyx.server.tenants.access import generate_data_plane_token
+from onyx.configs.app_configs import CONTROL_PLANE_API_BASE_URL
+from onyx.utils.logger import setup_logger
+from shared_configs.configs import MULTI_TENANT
+
+logger = setup_logger()
+
+# HTTP request timeout for billing service calls
+_REQUEST_TIMEOUT = 30.0
+
+
+class BillingServiceError(Exception):
+    """Exception raised for billing service errors."""
+
+    def __init__(self, message: str, status_code: int = 500):
+        self.message = message
+        self.status_code = status_code
+        super().__init__(self.message)
+
+
+def _get_proxy_headers(license_data: str | None) -> dict[str, str]:
+    """Build headers for proxy requests (self-hosted).
+
+    Self-hosted instances authenticate with their license.
+    """
+    headers = {"Content-Type": "application/json"}
+    if license_data:
+        headers["Authorization"] = f"Bearer {license_data}"
+    return headers
+
+
+def _get_direct_headers() -> dict[str, str]:
+    """Build headers for direct control plane requests (cloud).
+
+    Cloud instances authenticate with JWT.
+    """
+    token = generate_data_plane_token()
+    return {
+        "Content-Type": "application/json",
+        "Authorization": f"Bearer {token}",
+    }
+
+
+def _get_base_url() -> str:
+    """Get the base URL based on deployment type."""
+    if MULTI_TENANT:
+        return CONTROL_PLANE_API_BASE_URL
+    return f"{CLOUD_DATA_PLANE_URL}/proxy"
+
+
+def _get_headers(license_data: str | None) -> dict[str, str]:
+    """Get appropriate headers based on deployment type."""
+    if MULTI_TENANT:
+        return _get_direct_headers()
+    return _get_proxy_headers(license_data)
+
+
+async def _make_billing_request(
+    method: Literal["GET", "POST"],
+    path: str,
+    license_data: str | None = None,
+    body: dict | None = None,
+    params: dict | None = None,
+    error_message: str = "Billing service request failed",
+) -> dict:
+    """Make an HTTP request to the billing service.
+
+    Consolidates the common HTTP request pattern used by all billing operations.
+
+    Args:
+        method: HTTP method (GET or POST)
+        path: URL path (appended to base URL)
+        license_data: License for authentication (self-hosted)
+        body: Request body for POST requests
+        params: Query parameters for GET requests
+        error_message: Default error message if request fails
+
+    Returns:
+        Response JSON as dict
+
+    Raises:
+        BillingServiceError: If request fails
+    """
+
+    base_url = _get_base_url()
+    url = f"{base_url}{path}"
+    headers = _get_headers(license_data)
+
+    try:
+        async with httpx.AsyncClient(timeout=_REQUEST_TIMEOUT) as client:
+            if method == "GET":
+                response = await client.get(url, headers=headers, params=params)
+            else:
+                response = await client.post(url, headers=headers, json=body)
+
+            response.raise_for_status()
+            return response.json()
+
+    except httpx.HTTPStatusError as e:
+        detail = error_message
+        try:
+            error_data = e.response.json()
+            detail = error_data.get("detail", detail)
+        except Exception:
+            pass
+        logger.error(f"{error_message}: {e.response.status_code} - {detail}")
+        raise BillingServiceError(detail, e.response.status_code)
+
+    except httpx.RequestError:
+        logger.exception("Failed to connect to billing service")
+        raise BillingServiceError("Failed to connect to billing service", 502)
+
+
+async def create_checkout_session(
+    billing_period: str = "monthly",
+    seats: int | None = None,
+    email: str | None = None,
+    license_data: str | None = None,
+    redirect_url: str | None = None,
+    tenant_id: str | None = None,
+) -> CreateCheckoutSessionResponse:
+    """Create a Stripe checkout session.
+
+    Args:
+        billing_period: "monthly" or "annual"
+        seats: Number of seats to purchase (optional, uses default if not provided)
+        email: Customer email for new subscriptions
+        license_data: Existing license for renewals (self-hosted)
+        redirect_url: URL to redirect after successful checkout
+        tenant_id: Tenant ID (cloud only, for renewals)
+
+    Returns:
+        CreateCheckoutSessionResponse with checkout URL
+    """
+    body: dict = {"billing_period": billing_period}
+    if seats is not None:
+        body["seats"] = seats
+    if email:
+        body["email"] = email
+    if redirect_url:
+        body["redirect_url"] = redirect_url
+    if tenant_id and MULTI_TENANT:
+        body["tenant_id"] = tenant_id
+
+    data = await _make_billing_request(
+        method="POST",
+        path="/create-checkout-session",
+        license_data=license_data,
+        body=body,
+        error_message="Failed to create checkout session",
+    )
+    return CreateCheckoutSessionResponse(stripe_checkout_url=data["url"])
+
+
+async def create_customer_portal_session(
+    license_data: str | None = None,
+    return_url: str | None = None,
+    tenant_id: str | None = None,
+) -> CreateCustomerPortalSessionResponse:
+    """Create a Stripe customer portal session.
+
+    Args:
+        license_data: License blob for authentication (self-hosted)
+        return_url: URL to return to after portal session
+        tenant_id: Tenant ID (cloud only)
+
+    Returns:
+        CreateCustomerPortalSessionResponse with portal URL
+    """
+    body: dict = {}
+    if return_url:
+        body["return_url"] = return_url
+    if tenant_id and MULTI_TENANT:
+        body["tenant_id"] = tenant_id
+
+    data = await _make_billing_request(
+        method="POST",
+        path="/create-customer-portal-session",
+        license_data=license_data,
+        body=body,
+        error_message="Failed to create customer portal session",
+    )
+    return CreateCustomerPortalSessionResponse(stripe_customer_portal_url=data["url"])
+
+
+async def get_billing_information(
+    license_data: str | None = None,
+    tenant_id: str | None = None,
+) -> BillingInformationResponse | SubscriptionStatusResponse:
+    """Fetch billing information.
+
+    Args:
+        license_data: License blob for authentication (self-hosted)
+        tenant_id: Tenant ID (cloud only)
+
+    Returns:
+        BillingInformationResponse or SubscriptionStatusResponse if no subscription
+    """
+    params = {}
+    if tenant_id and MULTI_TENANT:
+        params["tenant_id"] = tenant_id
+
+    data = await _make_billing_request(
+        method="GET",
+        path="/billing-information",
+        license_data=license_data,
+        params=params or None,
+        error_message="Failed to fetch billing information",
+    )
+
+    # Check if no subscription
+    if isinstance(data, dict) and data.get("subscribed") is False:
+        return SubscriptionStatusResponse(subscribed=False)
+
+    return BillingInformationResponse(**data)
+
+
+async def update_seat_count(
+    new_seat_count: int,
+    license_data: str | None = None,
+    tenant_id: str | None = None,
+) -> SeatUpdateResponse:
+    """Update the seat count for the current subscription.
+
+    Args:
+        new_seat_count: New number of seats
+        license_data: License blob for authentication (self-hosted)
+        tenant_id: Tenant ID (cloud only)
+
+    Returns:
+        SeatUpdateResponse with updated seat information
+    """
+    body: dict = {"new_seat_count": new_seat_count}
+    if tenant_id and MULTI_TENANT:
+        body["tenant_id"] = tenant_id
+
+    data = await _make_billing_request(
+        method="POST",
+        path="/seats/update",
+        license_data=license_data,
+        body=body,
+        error_message="Failed to update seat count",
+    )
+
+    return SeatUpdateResponse(
+        success=data.get("success", False),
+        current_seats=data.get("current_seats", 0),
+        used_seats=data.get("used_seats", 0),
+        message=data.get("message"),
+        license=data.get("license"),
+    )

backend/ee/onyx/server/enterprise_settings/api.py

@@ -115,7 +115,7 @@ async def refresh_access_token(
 
 @admin_router.put("")
 def admin_ee_put_settings(
-    settings: EnterpriseSettings, _: User | None = Depends(current_admin_user)
+    settings: EnterpriseSettings, _: User = Depends(current_admin_user)
 ) -> None:
     store_settings(settings)
 
@@ -134,12 +134,12 @@ def ee_fetch_settings() -> EnterpriseSettings:
 def put_logo(
     file: UploadFile,
     is_logotype: bool = False,
-    _: User | None = Depends(current_admin_user),
+    _: User = Depends(current_admin_user),
 ) -> None:
     upload_logo(file=file, is_logotype=is_logotype)
 
 
-def fetch_logo_helper(db_session: Session) -> Response:
+def fetch_logo_helper(db_session: Session) -> Response:  # noqa: ARG001
     try:
         file_store = get_default_file_store()
         onyx_file = file_store.get_file_with_mime_type(get_logo_filename())
@@ -155,7 +155,7 @@ def fetch_logo_helper(db_session: Session) -> Response:
         return Response(content=onyx_file.data, media_type=onyx_file.mime_type)
 
 
-def fetch_logotype_helper(db_session: Session) -> Response:
+def fetch_logotype_helper(db_session: Session) -> Response:  # noqa: ARG001
     try:
         file_store = get_default_file_store()
         onyx_file = file_store.get_file_with_mime_type(get_logotype_filename())
@@ -187,7 +187,7 @@ def fetch_logo(
 
 @admin_router.put("/custom-analytics-script")
 def upload_custom_analytics_script(
-    script_upload: AnalyticsScriptUpload, _: User | None = Depends(current_admin_user)
+    script_upload: AnalyticsScriptUpload, _: User = Depends(current_admin_user)
 ) -> None:
     try:
         store_analytics_script(script_upload)

backend/ee/onyx/server/evals/api.py

@@ -17,7 +17,7 @@ router = APIRouter(prefix="/evals")
 @router.post("/eval_run", response_model=EvalRunAck)
 def eval_run(
     request: EvalConfigurationOptions,
-    user: User = Depends(current_cloud_superuser),
+    user: User = Depends(current_cloud_superuser),  # noqa: ARG001
 ) -> EvalRunAck:
     """
     Run an evaluation with the given message and optional dataset.