.

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

@@ -47,8 +47,7 @@ jobs:
           done
 
       - name: Publish Helm charts to gh-pages
-        # NOTE: HEAD of https://github.com/stefanprodan/helm-gh-pages/pull/43
-        uses: stefanprodan/helm-gh-pages@ad32ad3b8720abfeaac83532fd1e9bdfca5bbe27 # zizmor: ignore[impostor-commit]
+        uses: stefanprodan/helm-gh-pages@0ad2bb377311d61ac04ad9eb6f252fb68e207260 # ratchet:stefanprodan/helm-gh-pages@v1.7.0
         with:
           token: ${{ secrets.GITHUB_TOKEN }}
           charts_dir: deployment/helm/charts

.github/workflows/nightly-llm-provider-chat.yml

@@ -35,7 +35,6 @@ jobs:
     needs: [provider-chat-test]
     if: failure() && github.event_name == 'schedule'
     runs-on: ubuntu-slim
-    environment: ci-protected
     timeout-minutes: 5
     steps:
       - name: Checkout

.github/workflows/post-merge-beta-cherry-pick.yml

@@ -183,7 +183,6 @@ jobs:
       - cherry-pick-to-latest-release
     if: needs.resolve-cherry-pick-request.outputs.should_cherrypick == 'true' && needs.resolve-cherry-pick-request.result == 'success' && needs.cherry-pick-to-latest-release.result == 'success'
     runs-on: ubuntu-slim
-    environment: ci-protected
     timeout-minutes: 10
     steps:
       - name: Checkout
@@ -233,7 +232,6 @@ jobs:
       - cherry-pick-to-latest-release
     if: always() && needs.resolve-cherry-pick-request.outputs.should_cherrypick == 'true' && (needs.resolve-cherry-pick-request.result == 'failure' || needs.cherry-pick-to-latest-release.result == 'failure')
     runs-on: ubuntu-slim
-    environment: ci-protected
     timeout-minutes: 10
     steps:
       - name: Checkout

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

@@ -31,7 +31,6 @@ jobs:
       - runner=4cpu-linux-arm64
       - "run-id=${{ github.run_id }}-model-check"
       - "extras=ecr-cache"
-    environment: ci-protected
     timeout-minutes: 45
 
     env:

.github/workflows/preview.yml

@@ -15,7 +15,6 @@ permissions:
 jobs:
   Deploy-Preview:
     runs-on: ubuntu-latest
-    environment: ci-protected
     timeout-minutes: 30
     steps:
       - uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd

.github/workflows/release-cli.yml

@@ -13,6 +13,15 @@ jobs:
     permissions:
       id-token: write
     timeout-minutes: 10
+    strategy:
+      matrix:
+        os-arch:
+          - { goos: "linux", goarch: "amd64" }
+          - { goos: "linux", goarch: "arm64" }
+          - { goos: "windows", goarch: "amd64" }
+          - { goos: "windows", goarch: "arm64" }
+          - { goos: "darwin", goarch: "amd64" }
+          - { goos: "darwin", goarch: "arm64" }
     steps:
       - uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # ratchet:actions/checkout@v6
         with:
@@ -22,11 +31,9 @@ jobs:
           enable-cache: false
           version: "0.9.9"
       - run: |
-          for goos in linux windows darwin; do
-            for goarch in amd64 arm64; do
-              GOOS="$goos" GOARCH="$goarch" uv build --wheel
-            done
-          done
+          GOOS="${{ matrix.os-arch.goos }}" \
+          GOARCH="${{ matrix.os-arch.goarch }}" \
+          uv build --wheel
         working-directory: cli
       - run: uv publish
         working-directory: cli

.github/workflows/storybook-deploy.yml

@@ -25,7 +25,6 @@ permissions:
 jobs:
   Deploy-Storybook:
     runs-on: ubuntu-latest
-    environment: ci-protected
     timeout-minutes: 30
     steps:
       - uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # ratchet:actions/checkout@v4
@@ -55,7 +54,6 @@ jobs:
     needs: Deploy-Storybook
     if: always() && needs.Deploy-Storybook.result == 'failure'
     runs-on: ubuntu-latest
-    environment: ci-protected
     timeout-minutes: 10
     steps:
       - uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # ratchet:actions/checkout@v4

.github/workflows/sync_foss.yml

@@ -9,7 +9,6 @@ on:
 jobs:
   sync-foss:
     runs-on: ubuntu-latest
-    environment: ci-protected
     timeout-minutes: 45
     permissions:
       contents: read

.github/workflows/tag-nightly.yml

@@ -11,7 +11,6 @@ permissions:
 jobs:
   create-and-push-tag:
     runs-on: ubuntu-slim
-    environment: ci-protected
     timeout-minutes: 45
 
     steps:

.pre-commit-config.yaml

@@ -122,7 +122,7 @@ repos:
     rev: 5d1e709b7be35cb2025444e19de266b056b7b7ee # frozen: v2.10.1
     hooks:
       - id: golangci-lint
-        language_version: "1.26.1"
+        language_version: "1.26.0"
         entry: bash -c "find . -name go.mod -not -path './.venv/*' -print0 | xargs -0 -I{} bash -c 'cd \"$(dirname {})\" && golangci-lint run ./...'"
 
   - repo: https://github.com/astral-sh/ruff-pre-commit

README.md

@@ -35,7 +35,7 @@ Onyx comes loaded with advanced features like Agents, Web Search, RAG, MCP, Deep
 > [!TIP]
 > Run Onyx with one command (or see deployment section below):
 > ```
-> curl -fsSL https://onyx.app/install_onyx.sh | bash
+> curl -fsSL https://raw.githubusercontent.com/onyx-dot-app/onyx/main/deployment/docker_compose/install.sh > install.sh && chmod +x install.sh && ./install.sh
 > ```
 
 ****

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

@@ -28,7 +28,6 @@ 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_broker_client
 from onyx.background.celery.celery_redis import celery_get_queue_length
 from onyx.background.celery.celery_redis import celery_get_queued_task_ids
 from onyx.background.celery.celery_redis import celery_get_unacked_task_ids
@@ -188,6 +187,7 @@ def check_for_doc_permissions_sync(self: Task, *, tenant_id: str) -> bool | None
     # (which lives on a different db number)
     r = get_redis_client()
     r_replica = get_redis_replica_client()
+    r_celery: Redis = self.app.broker_connection().channel().client  # type: ignore
 
     lock_beat: RedisLock = r.lock(
         OnyxRedisLocks.CHECK_CONNECTOR_DOC_PERMISSIONS_SYNC_BEAT_LOCK,
@@ -227,7 +227,6 @@ def check_for_doc_permissions_sync(self: Task, *, tenant_id: str) -> bool | None
             # tasks can be in the queue in redis, in reserved tasks (prefetched by the worker),
             # or be currently executing
             try:
-                r_celery = celery_get_broker_client(self.app)
                 validate_permission_sync_fences(
                     tenant_id, r, r_replica, r_celery, lock_beat
                 )

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

@@ -29,7 +29,6 @@ from ee.onyx.external_permissions.sync_params import (
 from ee.onyx.external_permissions.sync_params import get_source_perm_sync_config
 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_broker_client
 from onyx.background.celery.celery_redis import celery_get_unacked_task_ids
 from onyx.background.celery.tasks.beat_schedule import CLOUD_BEAT_MULTIPLIER_DEFAULT
 from onyx.background.error_logging import emit_background_error
@@ -163,6 +162,7 @@ def check_for_external_group_sync(self: Task, *, tenant_id: str) -> bool | None:
     # (which lives on a different db number)
     r = get_redis_client()
     r_replica = get_redis_replica_client()
+    r_celery: Redis = self.app.broker_connection().channel().client  # type: ignore
 
     lock_beat: RedisLock = r.lock(
         OnyxRedisLocks.CHECK_CONNECTOR_EXTERNAL_GROUP_SYNC_BEAT_LOCK,
@@ -221,7 +221,6 @@ def check_for_external_group_sync(self: Task, *, tenant_id: str) -> bool | None:
             # tasks can be in the queue in redis, in reserved tasks (prefetched by the worker),
             # or be currently executing
             try:
-                r_celery = celery_get_broker_client(self.app)
                 validate_external_group_sync_fences(
                     tenant_id, self.app, r, r_replica, r_celery, lock_beat
                 )

backend/onyx/background/celery/celery_redis.py

@@ -1,6 +1,5 @@
 # These are helper objects for tracking the keys we need to write in redis
 import json
-import threading
 from typing import Any
 from typing import cast
 
@@ -8,59 +7,7 @@ from celery import Celery
 from redis import Redis
 
 from onyx.background.celery.configs.base import CELERY_SEPARATOR
-from onyx.configs.app_configs import REDIS_HEALTH_CHECK_INTERVAL
 from onyx.configs.constants import OnyxCeleryPriority
-from onyx.configs.constants import REDIS_SOCKET_KEEPALIVE_OPTIONS
-
-
-_broker_client: Redis | None = None
-_broker_url: str | None = None
-_broker_client_lock = threading.Lock()
-
-
-def celery_get_broker_client(app: Celery) -> Redis:
-    """Return a shared Redis client connected to the Celery broker DB.
-
-    Uses a module-level singleton so all tasks on a worker share one
-    connection instead of creating a new one per call. The client
-    connects directly to the broker Redis DB (parsed from the broker URL).
-
-    Thread-safe via lock — safe for use in Celery thread-pool workers.
-
-    Usage:
-        r_celery = celery_get_broker_client(self.app)
-        length = celery_get_queue_length(queue, r_celery)
-    """
-    global _broker_client, _broker_url
-    with _broker_client_lock:
-        url = app.conf.broker_url
-        if _broker_client is not None and _broker_url == url:
-            try:
-                _broker_client.ping()
-                return _broker_client
-            except Exception:
-                try:
-                    _broker_client.close()
-                except Exception:
-                    pass
-                _broker_client = None
-        elif _broker_client is not None:
-            try:
-                _broker_client.close()
-            except Exception:
-                pass
-            _broker_client = None
-
-        _broker_url = url
-        _broker_client = Redis.from_url(
-            url,
-            decode_responses=False,
-            health_check_interval=REDIS_HEALTH_CHECK_INTERVAL,
-            socket_keepalive=True,
-            socket_keepalive_options=REDIS_SOCKET_KEEPALIVE_OPTIONS,
-            retry_on_timeout=True,
-        )
-        return _broker_client
 
 
 def celery_get_unacked_length(r: Redis) -> int:

backend/onyx/background/celery/tasks/connector_deletion/tasks.py

@@ -14,7 +14,6 @@ from redis.lock import Lock as RedisLock
 from sqlalchemy.orm import Session
 
 from onyx.background.celery.apps.app_base import task_logger
-from onyx.background.celery.celery_redis import celery_get_broker_client
 from onyx.background.celery.celery_redis import celery_get_queue_length
 from onyx.background.celery.celery_redis import celery_get_queued_task_ids
 from onyx.configs.app_configs import JOB_TIMEOUT
@@ -133,6 +132,7 @@ def revoke_tasks_blocking_deletion(
 def check_for_connector_deletion_task(self: Task, *, tenant_id: str) -> bool | None:
     r = get_redis_client()
     r_replica = get_redis_replica_client()
+    r_celery: Redis = self.app.broker_connection().channel().client  # type: ignore
 
     lock_beat: RedisLock = r.lock(
         OnyxRedisLocks.CHECK_CONNECTOR_DELETION_BEAT_LOCK,
@@ -149,7 +149,6 @@ def check_for_connector_deletion_task(self: Task, *, tenant_id: str) -> bool | N
         if not r.exists(OnyxRedisSignals.BLOCK_VALIDATE_CONNECTOR_DELETION_FENCES):
             # clear fences that don't have associated celery tasks in progress
             try:
-                r_celery = celery_get_broker_client(self.app)
                 validate_connector_deletion_fences(
                     tenant_id, r, r_replica, r_celery, lock_beat
                 )

backend/onyx/background/celery/tasks/docprocessing/tasks.py

@@ -22,7 +22,6 @@ from sqlalchemy.orm import Session
 
 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_broker_client
 from onyx.background.celery.celery_redis import celery_get_unacked_task_ids
 from onyx.background.celery.celery_utils import httpx_init_vespa_pool
 from onyx.background.celery.memory_monitoring import emit_process_memory
@@ -450,7 +449,7 @@ def check_indexing_completion(
             ):
                 # Check if the task exists in the celery queue
                 # This handles the case where Redis dies after task creation but before task execution
-                redis_celery = celery_get_broker_client(task.app)
+                redis_celery = task.app.broker_connection().channel().client  # type: ignore
                 task_exists = celery_find_task(
                     attempt.celery_task_id,
                     OnyxCeleryQueues.CONNECTOR_DOC_FETCHING,

backend/onyx/background/celery/tasks/monitoring/tasks.py

@@ -1,5 +1,6 @@
 import json
 import time
+from collections.abc import Callable
 from datetime import timedelta
 from itertools import islice
 from typing import Any
@@ -18,7 +19,6 @@ from sqlalchemy import text
 from sqlalchemy.orm import Session
 
 from onyx.background.celery.apps.app_base import task_logger
-from onyx.background.celery.celery_redis import celery_get_broker_client
 from onyx.background.celery.celery_redis import celery_get_queue_length
 from onyx.background.celery.celery_redis import celery_get_unacked_task_ids
 from onyx.background.celery.memory_monitoring import emit_process_memory
@@ -698,27 +698,31 @@ def monitor_background_processes(self: Task, *, tenant_id: str) -> None:
         return None
 
     try:
+        # Get Redis client for Celery broker
+        redis_celery = self.app.broker_connection().channel().client  # type: ignore
         redis_std = get_redis_client()
 
-        # Collect queue metrics with broker connection
-        r_celery = celery_get_broker_client(self.app)
-        queue_metrics = _collect_queue_metrics(r_celery)
+        # Define metric collection functions and their dependencies
+        metric_functions: list[Callable[[], list[Metric]]] = [
+            lambda: _collect_queue_metrics(redis_celery),
+            lambda: _collect_connector_metrics(db_session, redis_std),
+            lambda: _collect_sync_metrics(db_session, redis_std),
+        ]
 
-        # Collect remaining metrics (no broker connection needed)
+        # Collect and log each metric
         with get_session_with_current_tenant() as db_session:
-            all_metrics: list[Metric] = queue_metrics
-            all_metrics.extend(_collect_connector_metrics(db_session, redis_std))
-            all_metrics.extend(_collect_sync_metrics(db_session, redis_std))
+            for metric_fn in metric_functions:
+                metrics = metric_fn()
+                for metric in metrics:
+                    # double check to make sure we aren't double-emitting metrics
+                    if metric.key is None or not _has_metric_been_emitted(
+                        redis_std, metric.key
+                    ):
+                        metric.log()
+                        metric.emit(tenant_id)
 
-            for metric in all_metrics:
-                if metric.key is None or not _has_metric_been_emitted(
-                    redis_std, metric.key
-                ):
-                    metric.log()
-                    metric.emit(tenant_id)
-
-                if metric.key is not None:
-                    _mark_metric_as_emitted(redis_std, metric.key)
+                    if metric.key is not None:
+                        _mark_metric_as_emitted(redis_std, metric.key)
 
         task_logger.info("Successfully collected background metrics")
     except SoftTimeLimitExceeded:
@@ -886,7 +890,7 @@ def monitor_celery_queues_helper(
 ) -> None:
     """A task to monitor all celery queue lengths."""
 
-    r_celery = celery_get_broker_client(task.app)
+    r_celery = task.app.broker_connection().channel().client  # type: ignore
     n_celery = celery_get_queue_length(OnyxCeleryQueues.PRIMARY, r_celery)
     n_docfetching = celery_get_queue_length(
         OnyxCeleryQueues.CONNECTOR_DOC_FETCHING, r_celery
@@ -1076,7 +1080,7 @@ def cloud_monitor_celery_pidbox(
     num_deleted = 0
 
     MAX_PIDBOX_IDLE = 24 * 3600  # 1 day in seconds
-    r_celery = celery_get_broker_client(self.app)
+    r_celery: Redis = self.app.broker_connection().channel().client  # type: ignore
     for key in r_celery.scan_iter("*.reply.celery.pidbox"):
         key_bytes = cast(bytes, key)
         key_str = key_bytes.decode("utf-8")

backend/onyx/background/celery/tasks/pruning/tasks.py

@@ -17,7 +17,6 @@ from sqlalchemy.orm import Session
 
 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_broker_client
 from onyx.background.celery.celery_redis import celery_get_queue_length
 from onyx.background.celery.celery_redis import celery_get_queued_task_ids
 from onyx.background.celery.celery_redis import celery_get_unacked_task_ids
@@ -204,6 +203,7 @@ def _is_pruning_due(cc_pair: ConnectorCredentialPair) -> bool:
 def check_for_pruning(self: Task, *, tenant_id: str) -> bool | None:
     r = get_redis_client()
     r_replica = get_redis_replica_client()
+    r_celery: Redis = self.app.broker_connection().channel().client  # type: ignore
 
     lock_beat: RedisLock = r.lock(
         OnyxRedisLocks.CHECK_PRUNE_BEAT_LOCK,
@@ -261,7 +261,6 @@ def check_for_pruning(self: Task, *, tenant_id: str) -> bool | None:
             # tasks can be in the queue in redis, in reserved tasks (prefetched by the worker),
             # or be currently executing
             try:
-                r_celery = celery_get_broker_client(self.app)
                 validate_pruning_fences(tenant_id, r, r_replica, r_celery, lock_beat)
             except Exception:
                 task_logger.exception("Exception while validating pruning fences")

backend/onyx/background/celery/tasks/user_file_processing/tasks.py

@@ -16,7 +16,6 @@ from sqlalchemy.orm import Session
 
 from onyx.access.access import build_access_for_user_files
 from onyx.background.celery.apps.app_base import task_logger
-from onyx.background.celery.celery_redis import celery_get_broker_client
 from onyx.background.celery.celery_redis import celery_get_queue_length
 from onyx.background.celery.celery_utils import httpx_init_vespa_pool
 from onyx.background.celery.tasks.shared.RetryDocumentIndex import RetryDocumentIndex
@@ -106,7 +105,7 @@ def _user_file_delete_queued_key(user_file_id: str | UUID) -> str:
 
 
 def get_user_file_project_sync_queue_depth(celery_app: Celery) -> int:
-    redis_celery = celery_get_broker_client(celery_app)
+    redis_celery: Redis = celery_app.broker_connection().channel().client  # type: ignore
     return celery_get_queue_length(
         OnyxCeleryQueues.USER_FILE_PROJECT_SYNC, redis_celery
     )
@@ -239,7 +238,7 @@ def check_user_file_processing(self: Task, *, tenant_id: str) -> None:
     skipped_guard = 0
     try:
         # --- Protection 1: queue depth backpressure ---
-        r_celery = celery_get_broker_client(self.app)
+        r_celery = self.app.broker_connection().channel().client  # type: ignore
         queue_len = celery_get_queue_length(
             OnyxCeleryQueues.USER_FILE_PROCESSING, r_celery
         )
@@ -592,7 +591,7 @@ def check_for_user_file_delete(self: Task, *, tenant_id: str) -> None:
         # --- Protection 1: queue depth backpressure ---
         # NOTE: must use the broker's Redis client (not redis_client) because
         # Celery queues live on a separate Redis DB with CELERY_SEPARATOR keys.
-        r_celery = celery_get_broker_client(self.app)
+        r_celery: Redis = self.app.broker_connection().channel().client  # type: ignore
         queue_len = celery_get_queue_length(OnyxCeleryQueues.USER_FILE_DELETE, r_celery)
         if queue_len > USER_FILE_DELETE_MAX_QUEUE_DEPTH:
             task_logger.warning(

backend/onyx/configs/app_configs.py

@@ -805,10 +805,6 @@ MINI_CHUNK_SIZE = 150
 # This is the number of regular chunks per large chunk
 LARGE_CHUNK_RATIO = 4
 
-# The maximum number of chunks that can be held for 1 document processing batch
-# The purpose of this is to set an upper bound on memory usage
-MAX_CHUNKS_PER_DOC_BATCH = int(os.environ.get("MAX_CHUNKS_PER_DOC_BATCH") or 1000)
-
 # Include the document level metadata in each chunk. If the metadata is too long, then it is thrown out
 # We don't want the metadata to overwhelm the actual contents of the chunk
 SKIP_METADATA_IN_CHUNK = os.environ.get("SKIP_METADATA_IN_CHUNK", "").lower() == "true"

backend/onyx/db/projects.py

@@ -7,6 +7,7 @@ from fastapi import HTTPException
 from fastapi import UploadFile
 from pydantic import BaseModel
 from pydantic import ConfigDict
+from pydantic import Field
 from sqlalchemy import func
 from sqlalchemy.orm import Session
 from starlette.background import BackgroundTasks
@@ -34,9 +35,19 @@ class CategorizedFilesResult(BaseModel):
     user_files: list[UserFile]
     rejected_files: list[RejectedFile]
     id_to_temp_id: dict[str, str]
+    # Filenames that should be stored but not indexed.
+    skip_indexing_filenames: set[str] = Field(default_factory=set)
     # Allow SQLAlchemy ORM models inside this result container
     model_config = ConfigDict(arbitrary_types_allowed=True)
 
+    @property
+    def indexable_files(self) -> list[UserFile]:
+        return [
+            uf
+            for uf in self.user_files
+            if (uf.name or "") not in self.skip_indexing_filenames
+        ]
+
 
 def build_hashed_file_key(file: UploadFile) -> str:
     name_prefix = (file.filename or "")[:50]
@@ -98,6 +109,7 @@ def create_user_files(
         user_files=user_files,
         rejected_files=rejected_files,
         id_to_temp_id=id_to_temp_id,
+        skip_indexing_filenames=categorized_files.skip_indexing,
     )
 
 
@@ -123,6 +135,7 @@ def upload_files_to_user_files_with_indexing(
     user_files = categorized_files_result.user_files
     rejected_files = categorized_files_result.rejected_files
     id_to_temp_id = categorized_files_result.id_to_temp_id
+    indexable_files = categorized_files_result.indexable_files
     # Trigger per-file processing immediately for the current tenant
     tenant_id = get_current_tenant_id()
     for rejected_file in rejected_files:
@@ -134,12 +147,12 @@ def upload_files_to_user_files_with_indexing(
         from onyx.background.task_utils import drain_processing_loop
 
         background_tasks.add_task(drain_processing_loop, tenant_id)
-        for user_file in user_files:
+        for user_file in indexable_files:
             logger.info(f"Queued in-process processing for user_file_id={user_file.id}")
     else:
         from onyx.background.celery.versioned_apps.client import app as client_app
 
-        for user_file in user_files:
+        for user_file in indexable_files:
             task = client_app.send_task(
                 OnyxCeleryTask.PROCESS_SINGLE_USER_FILE,
                 kwargs={"user_file_id": user_file.id, "tenant_id": tenant_id},
@@ -155,6 +168,7 @@ def upload_files_to_user_files_with_indexing(
         user_files=user_files,
         rejected_files=rejected_files,
         id_to_temp_id=id_to_temp_id,
+        skip_indexing_filenames=categorized_files_result.skip_indexing_filenames,
     )
 
 

backend/onyx/document_index/disabled.py

@@ -5,7 +5,6 @@ accidentally reaches the vector DB layer will fail loudly instead of timing
 out against a nonexistent Vespa/OpenSearch instance.
 """
 
-from collections.abc import Iterable
 from typing import Any
 
 from onyx.context.search.models import IndexFilters
@@ -67,7 +66,7 @@ class DisabledDocumentIndex(DocumentIndex):
     # ------------------------------------------------------------------
     def index(
         self,
-        chunks: Iterable[DocMetadataAwareIndexChunk],  # noqa: ARG002
+        chunks: list[DocMetadataAwareIndexChunk],  # noqa: ARG002
         index_batch_params: IndexBatchParams,  # noqa: ARG002
     ) -> set[DocumentInsertionRecord]:
         raise RuntimeError(VECTOR_DB_DISABLED_ERROR)

backend/onyx/document_index/interfaces.py

@@ -1,5 +1,4 @@
 import abc
-from collections.abc import Iterable
 from dataclasses import dataclass
 from datetime import datetime
 from typing import Any
@@ -207,7 +206,7 @@ class Indexable(abc.ABC):
     @abc.abstractmethod
     def index(
         self,
-        chunks: Iterable[DocMetadataAwareIndexChunk],
+        chunks: list[DocMetadataAwareIndexChunk],
         index_batch_params: IndexBatchParams,
     ) -> set[DocumentInsertionRecord]:
         """
@@ -227,8 +226,8 @@ class Indexable(abc.ABC):
         it is done automatically outside of this code.
 
         Parameters:
-        - chunks: Document chunks with all of the information needed for
-                indexing to the document index.
+        - chunks: Document chunks with all of the information needed for indexing to the document
+                index.
         - tenant_id: The tenant id of the user whose chunks are being indexed
         - large_chunks_enabled: Whether large chunks are enabled
 

backend/onyx/document_index/interfaces_new.py

@@ -1,5 +1,4 @@
 import abc
-from collections.abc import Iterable
 from typing import Self
 
 from pydantic import BaseModel
@@ -210,10 +209,10 @@ class Indexable(abc.ABC):
     @abc.abstractmethod
     def index(
         self,
-        chunks: Iterable[DocMetadataAwareIndexChunk],
+        chunks: list[DocMetadataAwareIndexChunk],
         indexing_metadata: IndexingMetadata,
     ) -> list[DocumentInsertionRecord]:
-        """Indexes an iterable of document chunks into the document index.
+        """Indexes a list of document chunks into the document index.
 
         This is often a batch operation including chunks from multiple
         documents.

backend/onyx/document_index/opensearch/opensearch_document_index.py

@@ -1,12 +1,11 @@
 import json
-from collections.abc import Iterable
+from collections import defaultdict
 from typing import Any
 
 import httpx
 from opensearchpy import NotFoundError
 
 from onyx.access.models import DocumentAccess
-from onyx.configs.app_configs import MAX_CHUNKS_PER_DOC_BATCH
 from onyx.configs.app_configs import VERIFY_CREATE_OPENSEARCH_INDEX_ON_INIT_MT
 from onyx.configs.chat_configs import NUM_RETURNED_HITS
 from onyx.configs.chat_configs import TITLE_CONTENT_RATIO
@@ -352,7 +351,7 @@ class OpenSearchOldDocumentIndex(OldDocumentIndex):
 
     def index(
         self,
-        chunks: Iterable[DocMetadataAwareIndexChunk],
+        chunks: list[DocMetadataAwareIndexChunk],
         index_batch_params: IndexBatchParams,
     ) -> set[OldDocumentInsertionRecord]:
         """
@@ -648,10 +647,10 @@ class OpenSearchDocumentIndex(DocumentIndex):
 
     def index(
         self,
-        chunks: Iterable[DocMetadataAwareIndexChunk],
-        indexing_metadata: IndexingMetadata,
+        chunks: list[DocMetadataAwareIndexChunk],
+        indexing_metadata: IndexingMetadata,  # noqa: ARG002
     ) -> list[DocumentInsertionRecord]:
-        """Indexes an iterable of document chunks into the document index.
+        """Indexes a list of document chunks into the document index.
 
         Groups chunks by document ID and for each document, deletes existing
         chunks and indexes the new chunks in bulk.
@@ -674,34 +673,29 @@ class OpenSearchDocumentIndex(DocumentIndex):
                 document is newly indexed or had already existed and was just
                 updated.
         """
-        total_chunks = sum(
-            cc.new_chunk_cnt
-            for cc in indexing_metadata.doc_id_to_chunk_cnt_diff.values()
+        # Group chunks by document ID.
+        doc_id_to_chunks: dict[str, list[DocMetadataAwareIndexChunk]] = defaultdict(
+            list
         )
+        for chunk in chunks:
+            doc_id_to_chunks[chunk.source_document.id].append(chunk)
         logger.debug(
-            f"[OpenSearchDocumentIndex] Indexing {total_chunks} chunks from {len(indexing_metadata.doc_id_to_chunk_cnt_diff)} "
+            f"[OpenSearchDocumentIndex] Indexing {len(chunks)} chunks from {len(doc_id_to_chunks)} "
             f"documents for index {self._index_name}."
         )
 
         document_indexing_results: list[DocumentInsertionRecord] = []
-        deleted_doc_ids: set[str] = set()
-        # Buffer chunks per document as they arrive from the iterable.
-        # When the document ID changes flush the buffered chunks.
-        current_doc_id: str | None = None
-        current_chunks: list[DocMetadataAwareIndexChunk] = []
-
-        def _flush_chunks(doc_chunks: list[DocMetadataAwareIndexChunk]) -> None:
-            assert len(doc_chunks) > 0, "doc_chunks is empty"
-
+        # Try to index per-document.
+        for _, chunks in doc_id_to_chunks.items():
             # Create a batch of OpenSearch-formatted chunks for bulk insertion.
-            # Since we are doing this in batches, an error occurring midway
-            # can result in a state where chunks are deleted and not all the
-            # new chunks have been indexed.
+            # Do this before deleting existing chunks to reduce the amount of
+            # time the document index has no content for a given document, and
+            # to reduce the chance of entering a state where we delete chunks,
+            # then some error happens, and never successfully index new chunks.
             chunk_batch: list[DocumentChunk] = [
-                _convert_onyx_chunk_to_opensearch_document(chunk)
-                for chunk in doc_chunks
+                _convert_onyx_chunk_to_opensearch_document(chunk) for chunk in chunks
             ]
-            onyx_document: Document = doc_chunks[0].source_document
+            onyx_document: Document = chunks[0].source_document
             # First delete the doc's chunks from the index. This is so that
             # there are no dangling chunks in the index, in the event that the
             # new document's content contains fewer chunks than the previous
@@ -710,43 +704,22 @@ class OpenSearchDocumentIndex(DocumentIndex):
             # if the chunk count has actually decreased. This assumes that
             # overlapping chunks are perfectly overwritten. If we can't
             # guarantee that then we need the code as-is.
-            if onyx_document.id not in deleted_doc_ids:
-                num_chunks_deleted = self.delete(
-                    onyx_document.id, onyx_document.chunk_count
-                )
-                deleted_doc_ids.add(onyx_document.id)
-                # If we see that chunks were deleted we assume the doc already
-                # existed. We record the result before bulk_index_documents
-                # runs. If indexing raises, this entire result list is discarded
-                # by the caller's retry logic, so early recording is safe.
-                document_indexing_results.append(
-                    DocumentInsertionRecord(
-                        document_id=onyx_document.id,
-                        already_existed=num_chunks_deleted > 0,
-                    )
-                )
+            num_chunks_deleted = self.delete(
+                onyx_document.id, onyx_document.chunk_count
+            )
+            # If we see that chunks were deleted we assume the doc already
+            # existed.
+            document_insertion_record = DocumentInsertionRecord(
+                document_id=onyx_document.id,
+                already_existed=num_chunks_deleted > 0,
+            )
             # Now index. This will raise if a chunk of the same ID exists, which
             # we do not expect because we should have deleted all chunks.
             self._client.bulk_index_documents(
                 documents=chunk_batch,
                 tenant_state=self._tenant_state,
             )
-
-        for chunk in chunks:
-            doc_id = chunk.source_document.id
-            if doc_id != current_doc_id:
-                if current_chunks:
-                    _flush_chunks(current_chunks)
-                current_doc_id = doc_id
-                current_chunks = [chunk]
-            elif len(current_chunks) >= MAX_CHUNKS_PER_DOC_BATCH:
-                _flush_chunks(current_chunks)
-                current_chunks = [chunk]
-            else:
-                current_chunks.append(chunk)
-
-        if current_chunks:
-            _flush_chunks(current_chunks)
+            document_indexing_results.append(document_insertion_record)
 
         return document_indexing_results
 

backend/onyx/document_index/vespa/index.py

@@ -6,7 +6,6 @@ import re
 import time
 import urllib
 import zipfile
-from collections.abc import Iterable
 from dataclasses import dataclass
 from datetime import datetime
 from datetime import timedelta
@@ -462,7 +461,7 @@ class VespaIndex(DocumentIndex):
 
     def index(
         self,
-        chunks: Iterable[DocMetadataAwareIndexChunk],
+        chunks: list[DocMetadataAwareIndexChunk],
         index_batch_params: IndexBatchParams,
     ) -> set[OldDocumentInsertionRecord]:
         """

backend/onyx/document_index/vespa/vespa_document_index.py

@@ -1,8 +1,6 @@
 import concurrent.futures
 import logging
 import random
-from collections.abc import Generator
-from collections.abc import Iterable
 from typing import Any
 from uuid import UUID
 
@@ -10,7 +8,6 @@ import httpx
 from pydantic import BaseModel
 from retry import retry
 
-from onyx.configs.app_configs import MAX_CHUNKS_PER_DOC_BATCH
 from onyx.configs.app_configs import RECENCY_BIAS_MULTIPLIER
 from onyx.configs.app_configs import RERANK_COUNT
 from onyx.configs.chat_configs import DOC_TIME_DECAY
@@ -321,7 +318,7 @@ class VespaDocumentIndex(DocumentIndex):
 
     def index(
         self,
-        chunks: Iterable[DocMetadataAwareIndexChunk],
+        chunks: list[DocMetadataAwareIndexChunk],
         indexing_metadata: IndexingMetadata,
     ) -> list[DocumentInsertionRecord]:
         doc_id_to_chunk_cnt_diff = indexing_metadata.doc_id_to_chunk_cnt_diff
@@ -341,31 +338,22 @@ class VespaDocumentIndex(DocumentIndex):
 
         # Vespa has restrictions on valid characters, yet document IDs come from
         # external w.r.t. this class. We need to sanitize them.
-        #
-        # Instead of materializing all cleaned chunks upfront, we stream them
-        # through a generator that cleans IDs and builds the original-ID mapping
-        # incrementally as chunks flow into Vespa.
-        def _clean_and_track(
-            chunks_iter: Iterable[DocMetadataAwareIndexChunk],
-            id_map: dict[str, str],
-            seen_ids: set[str],
-        ) -> Generator[DocMetadataAwareIndexChunk, None, None]:
-            """Cleans chunk IDs and builds the original-ID mapping
-            incrementally as chunks flow through, avoiding a separate
-            materialization pass."""
-            for chunk in chunks_iter:
-                original_id = chunk.source_document.id
-                cleaned = clean_chunk_id_copy(chunk)
-                cleaned_id = cleaned.source_document.id
-                # Needed so the final DocumentInsertionRecord returned can have
-                # the original document ID. cleaned_chunks might not contain IDs
-                # exactly as callers supplied them.
-                id_map[cleaned_id] = original_id
-                seen_ids.add(cleaned_id)
-                yield cleaned
+        cleaned_chunks: list[DocMetadataAwareIndexChunk] = [
+            clean_chunk_id_copy(chunk) for chunk in chunks
+        ]
+        assert len(cleaned_chunks) == len(
+            chunks
+        ), "Bug: Cleaned chunks and input chunks have different lengths."
 
-        new_document_id_to_original_document_id: dict[str, str] = {}
-        all_cleaned_doc_ids: set[str] = set()
+        # Needed so the final DocumentInsertionRecord returned can have the
+        # original document ID. cleaned_chunks might not contain IDs exactly as
+        # callers supplied them.
+        new_document_id_to_original_document_id: dict[str, str] = dict()
+        for i, cleaned_chunk in enumerate(cleaned_chunks):
+            old_chunk = chunks[i]
+            new_document_id_to_original_document_id[
+                cleaned_chunk.source_document.id
+            ] = old_chunk.source_document.id
 
         existing_docs: set[str] = set()
 
@@ -421,16 +409,8 @@ class VespaDocumentIndex(DocumentIndex):
                     executor=executor,
                 )
 
-            # Insert new Vespa documents, streaming through the cleaning
-            # pipeline so chunks are never fully materialized.
-            cleaned_chunks = _clean_and_track(
-                chunks,
-                new_document_id_to_original_document_id,
-                all_cleaned_doc_ids,
-            )
-            for chunk_batch in batch_generator(
-                cleaned_chunks, min(BATCH_SIZE, MAX_CHUNKS_PER_DOC_BATCH)
-            ):
+            # Insert new Vespa documents.
+            for chunk_batch in batch_generator(cleaned_chunks, BATCH_SIZE):
                 batch_index_vespa_chunks(
                     chunks=chunk_batch,
                     index_name=self._index_name,
@@ -439,6 +419,10 @@ class VespaDocumentIndex(DocumentIndex):
                     executor=executor,
                 )
 
+        all_cleaned_doc_ids: set[str] = {
+            chunk.source_document.id for chunk in cleaned_chunks
+        }
+
         return [
             DocumentInsertionRecord(
                 document_id=new_document_id_to_original_document_id[cleaned_doc_id],

backend/onyx/file_processing/extract_file_text.py

@@ -44,7 +44,6 @@ KNOWN_OPENPYXL_BUGS = [
     "Value must be either numerical or a string containing a wildcard",
     "File contains no valid workbook part",
     "Unable to read workbook: could not read stylesheet from None",
-    "Colors must be aRGB hex values",
 ]
 
 

backend/onyx/file_processing/file_types.py

@@ -15,6 +15,7 @@ PLAIN_TEXT_MIME_TYPE = "text/plain"
 class OnyxMimeTypes:
     IMAGE_MIME_TYPES = {"image/jpg", "image/jpeg", "image/png", "image/webp"}
     CSV_MIME_TYPES = {"text/csv"}
+    TABULAR_MIME_TYPES = CSV_MIME_TYPES | {SPREADSHEET_MIME_TYPE}
     TEXT_MIME_TYPES = {
         PLAIN_TEXT_MIME_TYPE,
         "text/markdown",
@@ -34,13 +35,12 @@ class OnyxMimeTypes:
         PDF_MIME_TYPE,
         WORD_PROCESSING_MIME_TYPE,
         PRESENTATION_MIME_TYPE,
-        SPREADSHEET_MIME_TYPE,
         "message/rfc822",
         "application/epub+zip",
     }
 
     ALLOWED_MIME_TYPES = IMAGE_MIME_TYPES.union(
-        TEXT_MIME_TYPES, DOCUMENT_MIME_TYPES, CSV_MIME_TYPES
+        TEXT_MIME_TYPES, DOCUMENT_MIME_TYPES, TABULAR_MIME_TYPES
     )
 
     EXCLUDED_IMAGE_TYPES = {

backend/onyx/file_store/models.py

@@ -13,13 +13,14 @@ class ChatFileType(str, Enum):
     DOC = "document"
     # Plain text only contain the text
     PLAIN_TEXT = "plain_text"
-    CSV = "csv"
+    # Tabular data files (CSV, TSV, XLSX) — metadata-only injection
+    TABULAR = "tabular"
 
     def is_text_file(self) -> bool:
         return self in (
             ChatFileType.PLAIN_TEXT,
             ChatFileType.DOC,
-            ChatFileType.CSV,
+            ChatFileType.TABULAR,
         )
 
 

backend/onyx/indexing/adapters/document_indexing_adapter.py

@@ -19,8 +19,7 @@ from onyx.db.document import update_docs_updated_at__no_commit
 from onyx.db.document_set import fetch_document_sets_for_documents
 from onyx.indexing.indexing_pipeline import DocumentBatchPrepareContext
 from onyx.indexing.indexing_pipeline import index_doc_batch_prepare
-from onyx.indexing.models import ChunkEnrichmentContext
-from onyx.indexing.models import DocAwareChunk
+from onyx.indexing.models import BuildMetadataAwareChunksResult
 from onyx.indexing.models import DocMetadataAwareIndexChunk
 from onyx.indexing.models import IndexChunk
 from onyx.indexing.models import UpdatableChunkData
@@ -86,21 +85,14 @@ class DocumentIndexingBatchAdapter:
         ) as transaction:
             yield transaction
 
-    def prepare_enrichment(
+    def build_metadata_aware_chunks(
         self,
-        context: DocumentBatchPrepareContext,
+        chunks_with_embeddings: list[IndexChunk],
+        chunk_content_scores: list[float],
         tenant_id: str,
-        chunks: list[DocAwareChunk],
-    ) -> "DocumentChunkEnricher":
-        """Do all DB lookups once and return a per-chunk enricher."""
-        updatable_ids = [doc.id for doc in context.updatable_docs]
-
-        doc_id_to_new_chunk_cnt: dict[str, int] = {
-            doc_id: 0 for doc_id in updatable_ids
-        }
-        for chunk in chunks:
-            if chunk.source_document.id in doc_id_to_new_chunk_cnt:
-                doc_id_to_new_chunk_cnt[chunk.source_document.id] += 1
+        context: DocumentBatchPrepareContext,
+    ) -> BuildMetadataAwareChunksResult:
+        """Enrich chunks with access, document sets, boosts, token counts, and hierarchy."""
 
         no_access = DocumentAccess.build(
             user_emails=[],
@@ -110,30 +102,67 @@ class DocumentIndexingBatchAdapter:
             is_public=False,
         )
 
-        return DocumentChunkEnricher(
-            doc_id_to_access_info=get_access_for_documents(
+        updatable_ids = [doc.id for doc in context.updatable_docs]
+
+        doc_id_to_access_info = get_access_for_documents(
+            document_ids=updatable_ids, db_session=self.db_session
+        )
+        doc_id_to_document_set = {
+            document_id: document_sets
+            for document_id, document_sets in fetch_document_sets_for_documents(
                 document_ids=updatable_ids, db_session=self.db_session
-            ),
-            doc_id_to_document_set={
-                document_id: document_sets
-                for document_id, document_sets in fetch_document_sets_for_documents(
-                    document_ids=updatable_ids, db_session=self.db_session
-                )
-            },
-            doc_id_to_ancestor_ids=self._get_ancestor_ids_for_documents(
-                context.updatable_docs, tenant_id
-            ),
-            id_to_boost_map=context.id_to_boost_map,
-            doc_id_to_previous_chunk_cnt={
-                document_id: chunk_count
-                for document_id, chunk_count in fetch_chunk_counts_for_documents(
-                    document_ids=updatable_ids,
-                    db_session=self.db_session,
-                )
-            },
-            doc_id_to_new_chunk_cnt=dict(doc_id_to_new_chunk_cnt),
-            no_access=no_access,
-            tenant_id=tenant_id,
+            )
+        }
+
+        doc_id_to_previous_chunk_cnt: dict[str, int] = {
+            document_id: chunk_count
+            for document_id, chunk_count in fetch_chunk_counts_for_documents(
+                document_ids=updatable_ids,
+                db_session=self.db_session,
+            )
+        }
+
+        doc_id_to_new_chunk_cnt: dict[str, int] = {
+            doc_id: 0 for doc_id in updatable_ids
+        }
+        for chunk in chunks_with_embeddings:
+            if chunk.source_document.id in doc_id_to_new_chunk_cnt:
+                doc_id_to_new_chunk_cnt[chunk.source_document.id] += 1
+
+        # Get ancestor hierarchy node IDs for each document
+        doc_id_to_ancestor_ids = self._get_ancestor_ids_for_documents(
+            context.updatable_docs, tenant_id
+        )
+
+        access_aware_chunks = [
+            DocMetadataAwareIndexChunk.from_index_chunk(
+                index_chunk=chunk,
+                access=doc_id_to_access_info.get(chunk.source_document.id, no_access),
+                document_sets=set(
+                    doc_id_to_document_set.get(chunk.source_document.id, [])
+                ),
+                user_project=[],
+                personas=[],
+                boost=(
+                    context.id_to_boost_map[chunk.source_document.id]
+                    if chunk.source_document.id in context.id_to_boost_map
+                    else DEFAULT_BOOST
+                ),
+                tenant_id=tenant_id,
+                aggregated_chunk_boost_factor=chunk_content_scores[chunk_num],
+                ancestor_hierarchy_node_ids=doc_id_to_ancestor_ids[
+                    chunk.source_document.id
+                ],
+            )
+            for chunk_num, chunk in enumerate(chunks_with_embeddings)
+        ]
+
+        return BuildMetadataAwareChunksResult(
+            chunks=access_aware_chunks,
+            doc_id_to_previous_chunk_cnt=doc_id_to_previous_chunk_cnt,
+            doc_id_to_new_chunk_cnt=doc_id_to_new_chunk_cnt,
+            user_file_id_to_raw_text={},
+            user_file_id_to_token_count={},
         )
 
     def _get_ancestor_ids_for_documents(
@@ -174,7 +203,7 @@ class DocumentIndexingBatchAdapter:
         context: DocumentBatchPrepareContext,
         updatable_chunk_data: list[UpdatableChunkData],
         filtered_documents: list[Document],
-        enrichment: ChunkEnrichmentContext,
+        result: BuildMetadataAwareChunksResult,
     ) -> None:
         """Finalize DB updates, store plaintext, and mark docs as indexed."""
         updatable_ids = [doc.id for doc in context.updatable_docs]
@@ -198,7 +227,7 @@ class DocumentIndexingBatchAdapter:
 
         update_docs_chunk_count__no_commit(
             document_ids=updatable_ids,
-            doc_id_to_chunk_count=enrichment.doc_id_to_new_chunk_cnt,
+            doc_id_to_chunk_count=result.doc_id_to_new_chunk_cnt,
             db_session=self.db_session,
         )
 
@@ -220,52 +249,3 @@ class DocumentIndexingBatchAdapter:
         )
 
         self.db_session.commit()
-
-
-class DocumentChunkEnricher:
-    """Pre-computed metadata for per-chunk enrichment of connector documents."""
-
-    def __init__(
-        self,
-        doc_id_to_access_info: dict[str, DocumentAccess],
-        doc_id_to_document_set: dict[str, list[str]],
-        doc_id_to_ancestor_ids: dict[str, list[int]],
-        id_to_boost_map: dict[str, int],
-        doc_id_to_previous_chunk_cnt: dict[str, int],
-        doc_id_to_new_chunk_cnt: dict[str, int],
-        no_access: DocumentAccess,
-        tenant_id: str,
-    ) -> None:
-        self._doc_id_to_access_info = doc_id_to_access_info
-        self._doc_id_to_document_set = doc_id_to_document_set
-        self._doc_id_to_ancestor_ids = doc_id_to_ancestor_ids
-        self._id_to_boost_map = id_to_boost_map
-        self._no_access = no_access
-        self._tenant_id = tenant_id
-        self.doc_id_to_previous_chunk_cnt = doc_id_to_previous_chunk_cnt
-        self.doc_id_to_new_chunk_cnt = doc_id_to_new_chunk_cnt
-
-    def enrich_chunk(
-        self, chunk: IndexChunk, score: float
-    ) -> DocMetadataAwareIndexChunk:
-        return DocMetadataAwareIndexChunk.from_index_chunk(
-            index_chunk=chunk,
-            access=self._doc_id_to_access_info.get(
-                chunk.source_document.id, self._no_access
-            ),
-            document_sets=set(
-                self._doc_id_to_document_set.get(chunk.source_document.id, [])
-            ),
-            user_project=[],
-            personas=[],
-            boost=(
-                self._id_to_boost_map[chunk.source_document.id]
-                if chunk.source_document.id in self._id_to_boost_map
-                else DEFAULT_BOOST
-            ),
-            tenant_id=self._tenant_id,
-            aggregated_chunk_boost_factor=score,
-            ancestor_hierarchy_node_ids=self._doc_id_to_ancestor_ids[
-                chunk.source_document.id
-            ],
-        )

backend/onyx/indexing/adapters/user_file_indexing_adapter.py

@@ -1,9 +1,6 @@
-from __future__ import annotations
-
 import contextlib
 import datetime
 import time
-from collections import defaultdict
 from collections.abc import Generator
 from uuid import UUID
 
@@ -27,8 +24,7 @@ from onyx.db.user_file import fetch_persona_ids_for_user_files
 from onyx.db.user_file import fetch_user_project_ids_for_user_files
 from onyx.file_store.utils import store_user_file_plaintext
 from onyx.indexing.indexing_pipeline import DocumentBatchPrepareContext
-from onyx.indexing.models import ChunkEnrichmentContext
-from onyx.indexing.models import DocAwareChunk
+from onyx.indexing.models import BuildMetadataAwareChunksResult
 from onyx.indexing.models import DocMetadataAwareIndexChunk
 from onyx.indexing.models import IndexChunk
 from onyx.indexing.models import UpdatableChunkData
@@ -106,20 +102,13 @@ class UserFileIndexingAdapter:
                 f"Failed to acquire locks after {_NUM_LOCK_ATTEMPTS} attempts for user files: {[doc.id for doc in documents]}"
             )
 
-    def prepare_enrichment(
+    def build_metadata_aware_chunks(
         self,
-        context: DocumentBatchPrepareContext,
+        chunks_with_embeddings: list[IndexChunk],
+        chunk_content_scores: list[float],
         tenant_id: str,
-        chunks: list[DocAwareChunk],
-    ) -> UserFileChunkEnricher:
-        """Do all DB lookups and pre-compute file metadata from chunks."""
-        updatable_ids = [doc.id for doc in context.updatable_docs]
-
-        doc_id_to_new_chunk_cnt: dict[str, int] = defaultdict(int)
-        content_by_file: dict[str, list[str]] = defaultdict(list)
-        for chunk in chunks:
-            doc_id_to_new_chunk_cnt[chunk.source_document.id] += 1
-            content_by_file[chunk.source_document.id].append(chunk.content)
+        context: DocumentBatchPrepareContext,
+    ) -> BuildMetadataAwareChunksResult:
 
         no_access = DocumentAccess.build(
             user_emails=[],
@@ -129,6 +118,7 @@ class UserFileIndexingAdapter:
             is_public=False,
         )
 
+        updatable_ids = [doc.id for doc in context.updatable_docs]
         user_file_id_to_project_ids = fetch_user_project_ids_for_user_files(
             user_file_ids=updatable_ids,
             db_session=self.db_session,
@@ -149,6 +139,17 @@ class UserFileIndexingAdapter:
             )
         }
 
+        user_file_id_to_new_chunk_cnt: dict[str, int] = {
+            user_file_id: len(
+                [
+                    chunk
+                    for chunk in chunks_with_embeddings
+                    if chunk.source_document.id == user_file_id
+                ]
+            )
+            for user_file_id in updatable_ids
+        }
+
         # Initialize tokenizer used for token count calculation
         try:
             llm = get_default_llm()
@@ -163,9 +164,15 @@ class UserFileIndexingAdapter:
         user_file_id_to_raw_text: dict[str, str] = {}
         user_file_id_to_token_count: dict[str, int | None] = {}
         for user_file_id in updatable_ids:
-            contents = content_by_file.get(user_file_id)
-            if contents:
-                combined_content = " ".join(contents)
+            user_file_chunks = [
+                chunk
+                for chunk in chunks_with_embeddings
+                if chunk.source_document.id == user_file_id
+            ]
+            if user_file_chunks:
+                combined_content = " ".join(
+                    [chunk.content for chunk in user_file_chunks]
+                )
                 user_file_id_to_raw_text[str(user_file_id)] = combined_content
                 token_count: int = (
                     count_tokens(combined_content, llm_tokenizer)
@@ -177,16 +184,28 @@ class UserFileIndexingAdapter:
                 user_file_id_to_raw_text[str(user_file_id)] = ""
                 user_file_id_to_token_count[str(user_file_id)] = None
 
-        return UserFileChunkEnricher(
-            user_file_id_to_access=user_file_id_to_access,
-            user_file_id_to_project_ids=user_file_id_to_project_ids,
-            user_file_id_to_persona_ids=user_file_id_to_persona_ids,
+        access_aware_chunks = [
+            DocMetadataAwareIndexChunk.from_index_chunk(
+                index_chunk=chunk,
+                access=user_file_id_to_access.get(chunk.source_document.id, no_access),
+                document_sets=set(),
+                user_project=user_file_id_to_project_ids.get(
+                    chunk.source_document.id, []
+                ),
+                personas=user_file_id_to_persona_ids.get(chunk.source_document.id, []),
+                boost=DEFAULT_BOOST,
+                tenant_id=tenant_id,
+                aggregated_chunk_boost_factor=chunk_content_scores[chunk_num],
+            )
+            for chunk_num, chunk in enumerate(chunks_with_embeddings)
+        ]
+
+        return BuildMetadataAwareChunksResult(
+            chunks=access_aware_chunks,
             doc_id_to_previous_chunk_cnt=user_file_id_to_previous_chunk_cnt,
-            doc_id_to_new_chunk_cnt=dict(doc_id_to_new_chunk_cnt),
+            doc_id_to_new_chunk_cnt=user_file_id_to_new_chunk_cnt,
             user_file_id_to_raw_text=user_file_id_to_raw_text,
             user_file_id_to_token_count=user_file_id_to_token_count,
-            no_access=no_access,
-            tenant_id=tenant_id,
         )
 
     def _notify_assistant_owners_if_files_ready(
@@ -230,9 +249,8 @@ class UserFileIndexingAdapter:
         context: DocumentBatchPrepareContext,
         updatable_chunk_data: list[UpdatableChunkData],  # noqa: ARG002
         filtered_documents: list[Document],  # noqa: ARG002
-        enrichment: ChunkEnrichmentContext,
+        result: BuildMetadataAwareChunksResult,
     ) -> None:
-        assert isinstance(enrichment, UserFileChunkEnricher)
         user_file_ids = [doc.id for doc in context.updatable_docs]
 
         user_files = (
@@ -248,10 +266,8 @@ class UserFileIndexingAdapter:
             user_file.last_project_sync_at = datetime.datetime.now(
                 datetime.timezone.utc
             )
-            user_file.chunk_count = enrichment.doc_id_to_new_chunk_cnt.get(
-                str(user_file.id), 0
-            )
-            user_file.token_count = enrichment.user_file_id_to_token_count[
+            user_file.chunk_count = result.doc_id_to_new_chunk_cnt[str(user_file.id)]
+            user_file.token_count = result.user_file_id_to_token_count[
                 str(user_file.id)
             ]
 
@@ -263,54 +279,8 @@ class UserFileIndexingAdapter:
         # Store the plaintext in the file store for faster retrieval
         # NOTE: this creates its own session to avoid committing the overall
         # transaction.
-        for user_file_id, raw_text in enrichment.user_file_id_to_raw_text.items():
+        for user_file_id, raw_text in result.user_file_id_to_raw_text.items():
             store_user_file_plaintext(
                 user_file_id=UUID(user_file_id),
                 plaintext_content=raw_text,
             )
-
-
-class UserFileChunkEnricher:
-    """Pre-computed metadata for per-chunk enrichment of user-uploaded files."""
-
-    def __init__(
-        self,
-        user_file_id_to_access: dict[str, DocumentAccess],
-        user_file_id_to_project_ids: dict[str, list[int]],
-        user_file_id_to_persona_ids: dict[str, list[int]],
-        doc_id_to_previous_chunk_cnt: dict[str, int],
-        doc_id_to_new_chunk_cnt: dict[str, int],
-        user_file_id_to_raw_text: dict[str, str],
-        user_file_id_to_token_count: dict[str, int | None],
-        no_access: DocumentAccess,
-        tenant_id: str,
-    ) -> None:
-        self._user_file_id_to_access = user_file_id_to_access
-        self._user_file_id_to_project_ids = user_file_id_to_project_ids
-        self._user_file_id_to_persona_ids = user_file_id_to_persona_ids
-        self._no_access = no_access
-        self._tenant_id = tenant_id
-        self.doc_id_to_previous_chunk_cnt = doc_id_to_previous_chunk_cnt
-        self.doc_id_to_new_chunk_cnt = doc_id_to_new_chunk_cnt
-        self.user_file_id_to_raw_text = user_file_id_to_raw_text
-        self.user_file_id_to_token_count = user_file_id_to_token_count
-
-    def enrich_chunk(
-        self, chunk: IndexChunk, score: float
-    ) -> DocMetadataAwareIndexChunk:
-        return DocMetadataAwareIndexChunk.from_index_chunk(
-            index_chunk=chunk,
-            access=self._user_file_id_to_access.get(
-                chunk.source_document.id, self._no_access
-            ),
-            document_sets=set(),
-            user_project=self._user_file_id_to_project_ids.get(
-                chunk.source_document.id, []
-            ),
-            personas=self._user_file_id_to_persona_ids.get(
-                chunk.source_document.id, []
-            ),
-            boost=DEFAULT_BOOST,
-            tenant_id=self._tenant_id,
-            aggregated_chunk_boost_factor=score,
-        )

backend/onyx/indexing/chunk_batch_store.py

@@ -1,89 +0,0 @@
-import pickle
-import shutil
-import tempfile
-from collections.abc import Iterator
-from pathlib import Path
-
-from onyx.indexing.models import IndexChunk
-
-
-class ChunkBatchStore:
-    """Manages serialization of embedded chunks to a temporary directory.
-
-    Owns the temp directory lifetime and provides save/load/stream/scrub
-    operations.
-
-    Use as a context manager to ensure cleanup::
-
-        with ChunkBatchStore() as store:
-            store.save(chunks, batch_idx=0)
-            for chunk in store.stream():
-                ...
-    """
-
-    _EXT = ".pkl"
-
-    def __init__(self) -> None:
-        self._tmpdir: Path | None = None
-
-    # -- context manager -----------------------------------------------------
-
-    def __enter__(self) -> "ChunkBatchStore":
-        self._tmpdir = Path(tempfile.mkdtemp(prefix="onyx_embeddings_"))
-        return self
-
-    def __exit__(self, *_exc: object) -> None:
-        if self._tmpdir is not None:
-            shutil.rmtree(self._tmpdir, ignore_errors=True)
-            self._tmpdir = None
-
-    @property
-    def _dir(self) -> Path:
-        assert self._tmpdir is not None, "ChunkBatchStore used outside context manager"
-        return self._tmpdir
-
-    # -- storage primitives --------------------------------------------------
-
-    def save(self, chunks: list[IndexChunk], batch_idx: int) -> None:
-        """Serialize a batch of embedded chunks to disk."""
-        with open(self._dir / f"batch_{batch_idx}{self._EXT}", "wb") as f:
-            pickle.dump(chunks, f)
-
-    def _load(self, batch_file: Path) -> list[IndexChunk]:
-        """Deserialize a batch of embedded chunks from a file."""
-        with open(batch_file, "rb") as f:
-            return pickle.load(f)
-
-    def _batch_files(self) -> list[Path]:
-        """Return batch files sorted by numeric index."""
-        return sorted(
-            self._dir.glob(f"batch_*{self._EXT}"),
-            key=lambda p: int(p.stem.removeprefix("batch_")),
-        )
-
-    # -- higher-level operations ---------------------------------------------
-
-    def stream(self) -> Iterator[IndexChunk]:
-        """Yield all chunks across all batch files.
-
-        Each call returns a fresh generator, so the data can be iterated
-        multiple times (e.g. once per document index).
-        """
-        for batch_file in self._batch_files():
-            yield from self._load(batch_file)
-
-    def scrub_failed_docs(self, failed_doc_ids: set[str]) -> None:
-        """Remove chunks belonging to *failed_doc_ids* from all batch files.
-
-        When a document fails embedding in batch N, earlier batches may
-        already contain successfully embedded chunks for that document.
-        This ensures the output is all-or-nothing per document.
-        """
-        for batch_file in self._batch_files():
-            batch_chunks = self._load(batch_file)
-            cleaned = [
-                c for c in batch_chunks if c.source_document.id not in failed_doc_ids
-            ]
-            if len(cleaned) != len(batch_chunks):
-                with open(batch_file, "wb") as f:
-                    pickle.dump(cleaned, f)

backend/onyx/indexing/indexing_pipeline.py

@@ -1,8 +1,5 @@
 from collections import defaultdict
 from collections.abc import Callable
-from collections.abc import Generator
-from collections.abc import Iterator
-from contextlib import contextmanager
 from typing import Protocol
 
 from pydantic import BaseModel
@@ -12,7 +9,6 @@ from sqlalchemy.orm import Session
 from onyx.configs.app_configs import DEFAULT_CONTEXTUAL_RAG_LLM_NAME
 from onyx.configs.app_configs import DEFAULT_CONTEXTUAL_RAG_LLM_PROVIDER
 from onyx.configs.app_configs import ENABLE_CONTEXTUAL_RAG
-from onyx.configs.app_configs import MAX_CHUNKS_PER_DOC_BATCH
 from onyx.configs.app_configs import MAX_DOCUMENT_CHARS
 from onyx.configs.app_configs import MAX_TOKENS_FOR_FULL_INCLUSION
 from onyx.configs.app_configs import USE_CHUNK_SUMMARY
@@ -47,12 +43,10 @@ from onyx.document_index.interfaces import DocumentMetadata
 from onyx.document_index.interfaces import IndexBatchParams
 from onyx.file_processing.image_summarization import summarize_image_with_error_handling
 from onyx.file_store.file_store import get_default_file_store
-from onyx.indexing.chunk_batch_store import ChunkBatchStore
 from onyx.indexing.chunker import Chunker
 from onyx.indexing.embedder import embed_chunks_with_failure_handling
 from onyx.indexing.embedder import IndexingEmbedder
 from onyx.indexing.models import DocAwareChunk
-from onyx.indexing.models import DocMetadataAwareIndexChunk
 from onyx.indexing.models import IndexingBatchAdapter
 from onyx.indexing.models import UpdatableChunkData
 from onyx.indexing.vector_db_insertion import write_chunks_to_vector_db_with_backoff
@@ -69,7 +63,6 @@ from onyx.natural_language_processing.utils import tokenizer_trim_middle
 from onyx.prompts.contextual_retrieval import CONTEXTUAL_RAG_PROMPT1
 from onyx.prompts.contextual_retrieval import CONTEXTUAL_RAG_PROMPT2
 from onyx.prompts.contextual_retrieval import DOCUMENT_SUMMARY_PROMPT
-from onyx.utils.batching import batch_generator
 from onyx.utils.logger import setup_logger
 from onyx.utils.postgres_sanitization import sanitize_documents_for_postgres
 from onyx.utils.threadpool_concurrency import run_functions_tuples_in_parallel
@@ -98,20 +91,6 @@ class IndexingPipelineResult(BaseModel):
 
     failures: list[ConnectorFailure]
 
-    @classmethod
-    def empty(cls, total_docs: int) -> "IndexingPipelineResult":
-        return cls(
-            new_docs=0,
-            total_docs=total_docs,
-            total_chunks=0,
-            failures=[],
-        )
-
-
-class ChunkEmbeddingResult(BaseModel):
-    successful_chunk_ids: list[tuple[int, str]]  # (chunk_id, document_id)
-    connector_failures: list[ConnectorFailure]
-
 
 class IndexingPipelineProtocol(Protocol):
     def __call__(
@@ -160,110 +139,6 @@ def _upsert_documents_in_db(
         )
 
 
-def _get_failed_doc_ids(failures: list[ConnectorFailure]) -> set[str]:
-    """Extract document IDs from a list of connector failures."""
-    return {f.failed_document.document_id for f in failures if f.failed_document}
-
-
-def _embed_chunks_to_store(
-    chunks: list[DocAwareChunk],
-    embedder: IndexingEmbedder,
-    tenant_id: str,
-    request_id: str | None,
-    store: ChunkBatchStore,
-) -> ChunkEmbeddingResult:
-    """Embed chunks in batches, spilling each batch to *store*.
-
-    If a document fails embedding in any batch, its chunks are excluded from
-    all batches (including earlier ones already written) so that the output
-    is all-or-nothing per document.
-    """
-    successful_chunk_ids: list[tuple[int, str]] = []
-    all_embedding_failures: list[ConnectorFailure] = []
-    # Track failed doc IDs across all batches so that a failure in batch N
-    # causes chunks for that doc to be skipped in batch N+1 and stripped
-    # from earlier batches.
-    all_failed_doc_ids: set[str] = set()
-
-    for batch_idx, chunk_batch in enumerate(
-        batch_generator(chunks, MAX_CHUNKS_PER_DOC_BATCH)
-    ):
-        # Skip chunks belonging to documents that failed in earlier batches.
-        chunk_batch = [
-            c for c in chunk_batch if c.source_document.id not in all_failed_doc_ids
-        ]
-        if not chunk_batch:
-            continue
-
-        logger.debug(f"Embedding batch {batch_idx}: {len(chunk_batch)} chunks")
-
-        chunks_with_embeddings, embedding_failures = embed_chunks_with_failure_handling(
-            chunks=chunk_batch,
-            embedder=embedder,
-            tenant_id=tenant_id,
-            request_id=request_id,
-        )
-        all_embedding_failures.extend(embedding_failures)
-        all_failed_doc_ids.update(_get_failed_doc_ids(embedding_failures))
-
-        # Only keep successfully embedded chunks for non-failed docs.
-        chunks_with_embeddings = [
-            c
-            for c in chunks_with_embeddings
-            if c.source_document.id not in all_failed_doc_ids
-        ]
-
-        successful_chunk_ids.extend(
-            (c.chunk_id, c.source_document.id) for c in chunks_with_embeddings
-        )
-
-        store.save(chunks_with_embeddings, batch_idx)
-        del chunks_with_embeddings
-
-    # Scrub earlier batches for docs that failed in later batches.
-    if all_failed_doc_ids:
-        store.scrub_failed_docs(all_failed_doc_ids)
-        successful_chunk_ids = [
-            (chunk_id, doc_id)
-            for chunk_id, doc_id in successful_chunk_ids
-            if doc_id not in all_failed_doc_ids
-        ]
-
-    return ChunkEmbeddingResult(
-        successful_chunk_ids=successful_chunk_ids,
-        connector_failures=all_embedding_failures,
-    )
-
-
-@contextmanager
-def embed_and_stream(
-    chunks: list[DocAwareChunk],
-    embedder: IndexingEmbedder,
-    tenant_id: str,
-    request_id: str | None,
-) -> Generator[tuple[ChunkEmbeddingResult, ChunkBatchStore], None, None]:
-    """Embed chunks to disk and yield a ``(result, store)`` pair.
-
-    The store owns the temp directory — files are cleaned up when the context
-    manager exits.
-
-    Usage::
-
-        with embed_and_stream(chunks, embedder, tenant_id, req_id) as (result, store):
-            for chunk in store.stream():
-                ...
-    """
-    with ChunkBatchStore() as store:
-        result = _embed_chunks_to_store(
-            chunks=chunks,
-            embedder=embedder,
-            tenant_id=tenant_id,
-            request_id=request_id,
-            store=store,
-        )
-        yield result, store
-
-
 def get_doc_ids_to_update(
     documents: list[Document], db_docs: list[DBDocument]
 ) -> list[Document]:
@@ -762,29 +637,6 @@ def add_contextual_summaries(
     return chunks
 
 
-def _verify_indexing_completeness(
-    insertion_records: list[DocumentInsertionRecord],
-    write_failures: list[ConnectorFailure],
-    embedding_failed_doc_ids: set[str],
-    updatable_ids: list[str],
-    document_index_name: str,
-) -> None:
-    """Verify that every updatable document was either indexed or reported as failed."""
-    all_returned_doc_ids = (
-        {r.document_id for r in insertion_records}
-        | {f.failed_document.document_id for f in write_failures if f.failed_document}
-        | embedding_failed_doc_ids
-    )
-    if all_returned_doc_ids != set(updatable_ids):
-        raise RuntimeError(
-            f"Some documents were not successfully indexed. "
-            f"Updatable IDs: {updatable_ids}, "
-            f"Returned IDs: {all_returned_doc_ids}. "
-            f"This should never happen. "
-            f"This occured for document index {document_index_name}"
-        )
-
-
 @log_function_time(debug_only=True)
 def index_doc_batch(
     *,
@@ -820,7 +672,12 @@ def index_doc_batch(
     filtered_documents = filter_fnc(document_batch)
     context = adapter.prepare(filtered_documents, ignore_time_skip)
     if not context:
-        return IndexingPipelineResult.empty(len(filtered_documents))
+        return IndexingPipelineResult(
+            new_docs=0,
+            total_docs=len(filtered_documents),
+            total_chunks=0,
+            failures=[],
+        )
 
     # Convert documents to IndexingDocument objects with processed section
     # logger.debug("Processing image sections")
@@ -859,99 +716,117 @@ def index_doc_batch(
         )
 
     logger.debug("Starting embedding")
-    with embed_and_stream(chunks, embedder, tenant_id, request_id) as (
-        embedding_result,
-        chunk_store,
-    ):
-        updatable_ids = [doc.id for doc in context.updatable_docs]
-        updatable_chunk_data = [
-            UpdatableChunkData(
-                chunk_id=chunk_id,
-                document_id=document_id,
-                boost_score=1.0,
-            )
-            for chunk_id, document_id in embedding_result.successful_chunk_ids
-        ]
+    chunks_with_embeddings, embedding_failures = (
+        embed_chunks_with_failure_handling(
+            chunks=chunks,
+            embedder=embedder,
+            tenant_id=tenant_id,
+            request_id=request_id,
+        )
+        if chunks
+        else ([], [])
+    )
 
-        embedding_failed_doc_ids = _get_failed_doc_ids(
-            embedding_result.connector_failures
+    chunk_content_scores = [1.0] * len(chunks_with_embeddings)
+
+    updatable_ids = [doc.id for doc in context.updatable_docs]
+    updatable_chunk_data = [
+        UpdatableChunkData(
+            chunk_id=chunk.chunk_id,
+            document_id=chunk.source_document.id,
+            boost_score=score,
+        )
+        for chunk, score in zip(chunks_with_embeddings, chunk_content_scores)
+    ]
+
+    # Acquires a lock on the documents so that no other process can modify them
+    # NOTE: don't need to acquire till here, since this is when the actual race condition
+    # with Vespa can occur.
+    with adapter.lock_context(context.updatable_docs):
+        # we're concerned about race conditions where multiple simultaneous indexings might result
+        # in one set of metadata overwriting another one in vespa.
+        # we still write data here for the immediate and most likely correct sync, but
+        # to resolve this, an update of the last modified field at the end of this loop
+        # always triggers a final metadata sync via the celery queue
+        result = adapter.build_metadata_aware_chunks(
+            chunks_with_embeddings=chunks_with_embeddings,
+            chunk_content_scores=chunk_content_scores,
+            tenant_id=tenant_id,
+            context=context,
         )
 
-        # Filter to only successfully embedded chunks so
-        # doc_id_to_new_chunk_cnt reflects what's actually written to Vespa.
-        embedded_chunks = [
-            c for c in chunks if c.source_document.id not in embedding_failed_doc_ids
-        ]
+        short_descriptor_list = [chunk.to_short_descriptor() for chunk in result.chunks]
+        short_descriptor_log = str(short_descriptor_list)[:1024]
+        logger.debug(f"Indexing the following chunks: {short_descriptor_log}")
 
-        # Acquires a lock on the documents so that no other process can modify
-        # them.  Not needed until here, since this is when the actual race
-        # condition with vector db can occur.
-        with adapter.lock_context(context.updatable_docs):
-            enricher = adapter.prepare_enrichment(
-                context=context,
-                tenant_id=tenant_id,
-                chunks=embedded_chunks,
+        primary_doc_idx_insertion_records: list[DocumentInsertionRecord] | None = None
+        primary_doc_idx_vector_db_write_failures: list[ConnectorFailure] | None = None
+        for document_index in document_indices:
+            # A document will not be spread across different batches, so all the
+            # documents with chunks in this set, are fully represented by the chunks
+            # in this set
+            (
+                insertion_records,
+                vector_db_write_failures,
+            ) = write_chunks_to_vector_db_with_backoff(
+                document_index=document_index,
+                chunks=result.chunks,
+                index_batch_params=IndexBatchParams(
+                    doc_id_to_previous_chunk_cnt=result.doc_id_to_previous_chunk_cnt,
+                    doc_id_to_new_chunk_cnt=result.doc_id_to_new_chunk_cnt,
+                    tenant_id=tenant_id,
+                    large_chunks_enabled=chunker.enable_large_chunks,
+                ),
             )
 
-            index_batch_params = IndexBatchParams(
-                doc_id_to_previous_chunk_cnt=enricher.doc_id_to_previous_chunk_cnt,
-                doc_id_to_new_chunk_cnt=enricher.doc_id_to_new_chunk_cnt,
-                tenant_id=tenant_id,
-                large_chunks_enabled=chunker.enable_large_chunks,
-            )
-
-            primary_doc_idx_insertion_records: list[DocumentInsertionRecord] | None = (
-                None
-            )
-            primary_doc_idx_vector_db_write_failures: list[ConnectorFailure] | None = (
-                None
-            )
-
-            for document_index in document_indices:
-
-                def _enriched_stream() -> Iterator[DocMetadataAwareIndexChunk]:
-                    for chunk in chunk_store.stream():
-                        yield enricher.enrich_chunk(chunk, 1.0)
-
-                insertion_records, write_failures = (
-                    write_chunks_to_vector_db_with_backoff(
-                        document_index=document_index,
-                        make_chunks=_enriched_stream,
-                        index_batch_params=index_batch_params,
-                    )
+            all_returned_doc_ids: set[str] = (
+                {record.document_id for record in insertion_records}
+                .union(
+                    {
+                        record.failed_document.document_id
+                        for record in vector_db_write_failures
+                        if record.failed_document
+                    }
                 )
-
-                _verify_indexing_completeness(
-                    insertion_records=insertion_records,
-                    write_failures=write_failures,
-                    embedding_failed_doc_ids=embedding_failed_doc_ids,
-                    updatable_ids=updatable_ids,
-                    document_index_name=document_index.__class__.__name__,
+                .union(
+                    {
+                        record.failed_document.document_id
+                        for record in embedding_failures
+                        if record.failed_document
+                    }
                 )
-                # We treat the first document index we got as the primary one used
-                # for reporting the state of indexing.
-                if primary_doc_idx_insertion_records is None:
-                    primary_doc_idx_insertion_records = insertion_records
-                if primary_doc_idx_vector_db_write_failures is None:
-                    primary_doc_idx_vector_db_write_failures = write_failures
-
-            adapter.post_index(
-                context=context,
-                updatable_chunk_data=updatable_chunk_data,
-                filtered_documents=filtered_documents,
-                enrichment=enricher,
             )
+            if all_returned_doc_ids != set(updatable_ids):
+                raise RuntimeError(
+                    f"Some documents were not successfully indexed. "
+                    f"Updatable IDs: {updatable_ids}, "
+                    f"Returned IDs: {all_returned_doc_ids}. "
+                    "This should never happen."
+                    f"This occured for document index {document_index.__class__.__name__}"
+                )
+            # We treat the first document index we got as the primary one used
+            # for reporting the state of indexing.
+            if primary_doc_idx_insertion_records is None:
+                primary_doc_idx_insertion_records = insertion_records
+            if primary_doc_idx_vector_db_write_failures is None:
+                primary_doc_idx_vector_db_write_failures = vector_db_write_failures
+
+        adapter.post_index(
+            context=context,
+            updatable_chunk_data=updatable_chunk_data,
+            filtered_documents=filtered_documents,
+            result=result,
+        )
 
     assert primary_doc_idx_insertion_records is not None
     assert primary_doc_idx_vector_db_write_failures is not None
     return IndexingPipelineResult(
-        new_docs=sum(
-            1 for r in primary_doc_idx_insertion_records if not r.already_existed
+        new_docs=len(
+            [r for r in primary_doc_idx_insertion_records if not r.already_existed]
         ),
         total_docs=len(filtered_documents),
-        total_chunks=len(embedding_result.successful_chunk_ids),
-        failures=primary_doc_idx_vector_db_write_failures
-        + embedding_result.connector_failures,
+        total_chunks=len(chunks_with_embeddings),
+        failures=primary_doc_idx_vector_db_write_failures + embedding_failures,
     )
 
 

backend/onyx/indexing/models.py

@@ -235,16 +235,12 @@ class UpdatableChunkData(BaseModel):
     boost_score: float
 
 
-class ChunkEnrichmentContext(Protocol):
-    """Returned by prepare_enrichment. Holds pre-computed metadata lookups
-    and provides per-chunk enrichment."""
-
+class BuildMetadataAwareChunksResult(BaseModel):
+    chunks: list[DocMetadataAwareIndexChunk]
     doc_id_to_previous_chunk_cnt: dict[str, int]
     doc_id_to_new_chunk_cnt: dict[str, int]
-
-    def enrich_chunk(
-        self, chunk: IndexChunk, score: float
-    ) -> DocMetadataAwareIndexChunk: ...
+    user_file_id_to_raw_text: dict[str, str]
+    user_file_id_to_token_count: dict[str, int | None]
 
 
 class IndexingBatchAdapter(Protocol):
@@ -258,24 +254,18 @@ class IndexingBatchAdapter(Protocol):
     ) -> Generator[TransactionalContext, None, None]:
         """Provide a transaction/row-lock context for critical updates."""
 
-    def prepare_enrichment(
+    def build_metadata_aware_chunks(
         self,
-        context: "DocumentBatchPrepareContext",
+        chunks_with_embeddings: list[IndexChunk],
+        chunk_content_scores: list[float],
         tenant_id: str,
-        chunks: list[DocAwareChunk],
-    ) -> ChunkEnrichmentContext:
-        """Prepare per-chunk enrichment data (access, document sets, boost, etc.).
-
-        Precondition: ``chunks`` have already been through the embedding step
-        (i.e. they are ``IndexChunk`` instances with populated embeddings,
-        passed here as the base ``DocAwareChunk`` type).
-        """
-        ...
+        context: "DocumentBatchPrepareContext",
+    ) -> BuildMetadataAwareChunksResult: ...
 
     def post_index(
         self,
         context: "DocumentBatchPrepareContext",
         updatable_chunk_data: list[UpdatableChunkData],
         filtered_documents: list[Document],
-        enrichment: ChunkEnrichmentContext,
+        result: BuildMetadataAwareChunksResult,
     ) -> None: ...

backend/onyx/indexing/vector_db_insertion.py

@@ -1,9 +1,6 @@
 import time
-from collections.abc import Callable
-from collections.abc import Iterable
+from collections import defaultdict
 from http import HTTPStatus
-from itertools import chain
-from itertools import groupby
 
 import httpx
 
@@ -31,22 +28,22 @@ def _log_insufficient_storage_error(e: Exception) -> None:
 
 def write_chunks_to_vector_db_with_backoff(
     document_index: DocumentIndex,
-    make_chunks: Callable[[], Iterable[DocMetadataAwareIndexChunk]],
+    chunks: list[DocMetadataAwareIndexChunk],
     index_batch_params: IndexBatchParams,
 ) -> tuple[list[DocumentInsertionRecord], list[ConnectorFailure]]:
     """Tries to insert all chunks in one large batch. If that batch fails for any reason,
     goes document by document to isolate the failure(s).
 
     IMPORTANT: must pass in whole documents at a time not individual chunks, since the
-    vector DB interface assumes that all chunks for a single document are present. The
-    chunks must also be in contiguous batches
+    vector DB interface assumes that all chunks for a single document are present.
     """
+
     # first try to write the chunks to the vector db
     try:
         return (
             list(
                 document_index.index(
-                    chunks=make_chunks(),
+                    chunks=chunks,
                     index_batch_params=index_batch_params,
                 )
             ),
@@ -63,23 +60,14 @@ def write_chunks_to_vector_db_with_backoff(
         # wait a couple seconds just to give the vector db a chance to recover
         time.sleep(2)
 
+    # try writing each doc one by one
+    chunks_for_docs: dict[str, list[DocMetadataAwareIndexChunk]] = defaultdict(list)
+    for chunk in chunks:
+        chunks_for_docs[chunk.source_document.id].append(chunk)
+
     insertion_records: list[DocumentInsertionRecord] = []
     failures: list[ConnectorFailure] = []
-
-    def key(chunk: DocMetadataAwareIndexChunk) -> str:
-        return chunk.source_document.id
-
-    seen_doc_ids: set[str] = set()
-    for doc_id, chunks_for_doc in groupby(make_chunks(), key=key):
-        if doc_id in seen_doc_ids:
-            raise RuntimeError(
-                f"Doc chunks are not arriving in order. Current doc_id={doc_id}, seen_doc_ids={list(seen_doc_ids)}"
-            )
-        seen_doc_ids.add(doc_id)
-
-        first_chunk = next(chunks_for_doc)
-        chunks_for_doc = chain([first_chunk], chunks_for_doc)
-
+    for doc_id, chunks_for_doc in chunks_for_docs.items():
         try:
             insertion_records.extend(
                 document_index.index(
@@ -99,7 +87,9 @@ def write_chunks_to_vector_db_with_backoff(
                 ConnectorFailure(
                     failed_document=DocumentFailure(
                         document_id=doc_id,
-                        document_link=first_chunk.get_link(),
+                        document_link=(
+                            chunks_for_doc[0].get_link() if chunks_for_doc else None
+                        ),
                     ),
                     failure_message=str(e),
                     exception=e,

backend/onyx/llm/multi_llm.py

@@ -185,21 +185,6 @@ def _messages_contain_tool_content(messages: list[dict[str, Any]]) -> bool:
     return False
 
 
-def _prompt_contains_tool_call_history(prompt: LanguageModelInput) -> bool:
-    """Check if the prompt contains any assistant messages with tool_calls.
-
-    When Anthropic's extended thinking is enabled, the API requires every
-    assistant message to start with a thinking block before any tool_use
-    blocks.  Since we don't preserve thinking_blocks (they carry
-    cryptographic signatures that can't be reconstructed), we must skip
-    the thinking param whenever history contains prior tool-calling turns.
-    """
-    from onyx.llm.models import AssistantMessage
-
-    msgs = prompt if isinstance(prompt, list) else [prompt]
-    return any(isinstance(msg, AssistantMessage) and msg.tool_calls for msg in msgs)
-
-
 def _is_vertex_model_rejecting_output_config(model_name: str) -> bool:
     normalized_model_name = model_name.lower()
     return any(
@@ -481,20 +466,7 @@ class LitellmLLM(LLM):
                     reasoning_effort
                 )
 
-                # Anthropic requires every assistant message with tool_use
-                # blocks to start with a thinking block that carries a
-                # cryptographic signature.  We don't preserve those blocks
-                # across turns, so skip thinking when the history already
-                # contains tool-calling assistant messages.  LiteLLM's
-                # modify_params workaround doesn't cover all providers
-                # (notably Bedrock).
-                can_enable_thinking = (
-                    budget_tokens is not None
-                    and not _prompt_contains_tool_call_history(prompt)
-                )
-
-                if can_enable_thinking:
-                    assert budget_tokens is not None  # mypy
+                if budget_tokens is not None:
                     if max_tokens is not None:
                         # Anthropic has a weird rule where max token has to be at least as much as budget tokens if set
                         # and the minimum budget tokens is 1024

backend/onyx/server/features/build/sandbox/kubernetes/docker/templates/outputs/web/package-lock.json

@@ -3844,9 +3844,9 @@
       }
     },
     "node_modules/@ts-morph/common/node_modules/brace-expansion": {
-      "version": "5.0.5",
-      "resolved": "https://registry.npmjs.org/brace-expansion/-/brace-expansion-5.0.5.tgz",
-      "integrity": "sha512-VZznLgtwhn+Mact9tfiwx64fA9erHH/MCXEUfB/0bX/6Fz6ny5EGTXYltMocqg4xFAQZtnO3DHWWXi8RiuN7cQ==",
+      "version": "5.0.3",
+      "resolved": "https://registry.npmjs.org/brace-expansion/-/brace-expansion-5.0.3.tgz",
+      "integrity": "sha512-fy6KJm2RawA5RcHkLa1z/ScpBeA762UF9KmZQxwIbDtRJrgLzM10depAiEQ+CXYcoiqW1/m96OAAoke2nE9EeA==",
       "license": "MIT",
       "dependencies": {
         "balanced-match": "^4.0.2"
@@ -4224,9 +4224,9 @@
       }
     },
     "node_modules/@typescript-eslint/typescript-estree/node_modules/brace-expansion": {
-      "version": "2.0.3",
-      "resolved": "https://registry.npmjs.org/brace-expansion/-/brace-expansion-2.0.3.tgz",
-      "integrity": "sha512-MCV/fYJEbqx68aE58kv2cA/kiky1G8vux3OR6/jbS+jIMe/6fJWa0DTzJU7dqijOWYwHi1t29FlfYI9uytqlpA==",
+      "version": "2.0.2",
+      "resolved": "https://registry.npmjs.org/brace-expansion/-/brace-expansion-2.0.2.tgz",
+      "integrity": "sha512-Jt0vHyM+jmUBqojB7E1NIYadt0vI0Qxjxd2TErW94wDz+E2LAm5vKMXXwg6ZZBTHPuUlDgQHKXvjGBdfcF1ZDQ==",
       "dev": true,
       "license": "MIT",
       "dependencies": {
@@ -5007,9 +5007,9 @@
       }
     },
     "node_modules/brace-expansion": {
-      "version": "1.1.13",
-      "resolved": "https://registry.npmjs.org/brace-expansion/-/brace-expansion-1.1.13.tgz",
-      "integrity": "sha512-9ZLprWS6EENmhEOpjCYW2c8VkmOvckIJZfkr7rBW6dObmfgJ/L1GpSYW5Hpo9lDz4D1+n0Ckz8rU7FwHDQiG/w==",
+      "version": "1.1.12",
+      "resolved": "https://registry.npmjs.org/brace-expansion/-/brace-expansion-1.1.12.tgz",
+      "integrity": "sha512-9T9UjW3r0UW5c1Q7GTwllptXwhvYmEzFhzMfZ9H7FQWt+uZePjZPjBP/W1ZEyZ1twGWom5/56TF4lPcqjnDHcg==",
       "dev": true,
       "license": "MIT",
       "dependencies": {

backend/onyx/server/features/hooks/api.py

@@ -123,8 +123,9 @@ def _validate_endpoint(
     (not reachable — indicates the api_key is invalid).
 
     Timeout handling:
-    - Any httpx.TimeoutException (ConnectTimeout, ReadTimeout, WriteTimeout, PoolTimeout) →
-      timeout (operator should consider increasing timeout_seconds).
+    - ConnectTimeout: TCP handshake never completed → cannot_connect.
+    - ReadTimeout / WriteTimeout: TCP was established, server responded slowly → timeout
+      (operator should consider increasing timeout_seconds).
     - All other exceptions → cannot_connect.
     """
     _check_ssrf_safety(endpoint_url)

backend/onyx/server/features/projects/projects_file_utils.py

@@ -76,11 +76,26 @@ class CategorizedFiles(BaseModel):
     acceptable: list[UploadFile] = Field(default_factory=list)
     rejected: list[RejectedFile] = Field(default_factory=list)
     acceptable_file_to_token_count: dict[str, int] = Field(default_factory=dict)
+    # Filenames within `acceptable` that should be stored but not indexed.
+    skip_indexing: set[str] = Field(default_factory=set)
 
     # Allow FastAPI UploadFile instances
     model_config = ConfigDict(arbitrary_types_allowed=True)
 
 
+# Extensions that bypass the token-count threshold on upload.
+_TOKEN_THRESHOLD_EXEMPT_EXTENSIONS: set[str] = {
+    ".csv",
+    ".tsv",
+    ".xlsx",
+}
+
+
+def _skip_token_threshold(extension: str) -> bool:
+    """Return True if this file extension should bypass the token limit."""
+    return extension.lower() in _TOKEN_THRESHOLD_EXEMPT_EXTENSIONS
+
+
 def _apply_long_side_cap(width: int, height: int, cap: int) -> tuple[int, int]:
     if max(width, height) <= cap:
         return width, height
@@ -264,7 +279,17 @@ def categorize_uploaded_files(
                 token_count = count_tokens(
                     text_content, tokenizer, token_limit=token_threshold
                 )
-                if token_threshold is not None and token_count > token_threshold:
+                exceeds_threshold = (
+                    token_threshold is not None and token_count > token_threshold
+                )
+                if exceeds_threshold and _skip_token_threshold(extension):
+                    # Exempt extensions (e.g. spreadsheets) are accepted
+                    # but flagged to skip indexing — only metadata is
+                    # injected into the LLM context.
+                    results.acceptable.append(upload)
+                    results.acceptable_file_to_token_count[filename] = token_count
+                    results.skip_indexing.add(filename)
+                elif exceeds_threshold:
                     results.rejected.append(
                         RejectedFile(
                             filename=filename,

backend/onyx/server/metrics/indexing_pipeline.py

@@ -12,6 +12,7 @@ stale, which is fine for monitoring dashboards.
 import json
 import threading
 import time
+from collections.abc import Callable
 from datetime import datetime
 from datetime import timezone
 from typing import Any
@@ -103,23 +104,25 @@ class _CachedCollector(Collector):
 
 
 class QueueDepthCollector(_CachedCollector):
-    """Reads Celery queue lengths from the broker Redis on each scrape."""
+    """Reads Celery queue lengths from the broker Redis on each scrape.
+
+    Uses a Redis client factory (callable) rather than a stored client
+    reference so the connection is always fresh from Celery's pool.
+    """
 
     def __init__(self, cache_ttl: float = _DEFAULT_CACHE_TTL) -> None:
         super().__init__(cache_ttl)
-        self._celery_app: Any | None = None
+        self._get_redis: Callable[[], Redis] | None = None
 
-    def set_celery_app(self, app: Any) -> None:
-        """Set the Celery app for broker Redis access."""
-        self._celery_app = app
+    def set_redis_factory(self, factory: Callable[[], Redis]) -> None:
+        """Set a callable that returns a broker Redis client on demand."""
+        self._get_redis = factory
 
     def _collect_fresh(self) -> list[GaugeMetricFamily]:
-        if self._celery_app is None:
+        if self._get_redis is None:
             return []
 
-        from onyx.background.celery.celery_redis import celery_get_broker_client
-
-        redis_client = celery_get_broker_client(self._celery_app)
+        redis_client = self._get_redis()
 
         depth = GaugeMetricFamily(
             "onyx_queue_depth",
@@ -401,19 +404,17 @@ class RedisHealthCollector(_CachedCollector):
 
     def __init__(self, cache_ttl: float = _DEFAULT_CACHE_TTL) -> None:
         super().__init__(cache_ttl)
-        self._celery_app: Any | None = None
+        self._get_redis: Callable[[], Redis] | None = None
 
-    def set_celery_app(self, app: Any) -> None:
-        """Set the Celery app for broker Redis access."""
-        self._celery_app = app
+    def set_redis_factory(self, factory: Callable[[], Redis]) -> None:
+        """Set a callable that returns a broker Redis client on demand."""
+        self._get_redis = factory
 
     def _collect_fresh(self) -> list[GaugeMetricFamily]:
-        if self._celery_app is None:
+        if self._get_redis is None:
             return []
 
-        from onyx.background.celery.celery_redis import celery_get_broker_client
-
-        redis_client = celery_get_broker_client(self._celery_app)
+        redis_client = self._get_redis()
 
         memory_used = GaugeMetricFamily(
             "onyx_redis_memory_used_bytes",

backend/onyx/server/metrics/indexing_pipeline_setup.py

@@ -3,8 +3,12 @@
 Called once by the monitoring celery worker after Redis and DB are ready.
 """
 
+from collections.abc import Callable
+from typing import Any
+
 from celery import Celery
 from prometheus_client.registry import REGISTRY
+from redis import Redis
 
 from onyx.server.metrics.indexing_pipeline import ConnectorHealthCollector
 from onyx.server.metrics.indexing_pipeline import IndexAttemptCollector
@@ -17,7 +21,7 @@ from onyx.utils.logger import setup_logger
 logger = setup_logger()
 
 # Module-level singletons — these are lightweight objects (no connections or DB
-# state) until configure() / set_celery_app() is called. Keeping them at
+# state) until configure() / set_redis_factory() is called. Keeping them at
 # module level ensures they survive the lifetime of the worker process and are
 # only registered with the Prometheus registry once.
 _queue_collector = QueueDepthCollector()
@@ -28,15 +32,72 @@ _worker_health_collector = WorkerHealthCollector()
 _heartbeat_monitor: WorkerHeartbeatMonitor | None = None
 
 
+def _make_broker_redis_factory(celery_app: Celery) -> Callable[[], Redis]:
+    """Create a factory that returns a cached broker Redis client.
+
+    Reuses a single connection across scrapes to avoid leaking connections.
+    Reconnects automatically if the cached connection becomes stale.
+    """
+    _cached_client: list[Redis | None] = [None]
+    # Keep a reference to the Kombu Connection so we can close it on
+    # reconnect (the raw Redis client outlives the Kombu wrapper).
+    _cached_kombu_conn: list[Any] = [None]
+
+    def _close_client(client: Redis) -> None:
+        """Best-effort close of a Redis client."""
+        try:
+            client.close()
+        except Exception:
+            logger.debug("Failed to close stale Redis client", exc_info=True)
+
+    def _close_kombu_conn() -> None:
+        """Best-effort close of the cached Kombu Connection."""
+        conn = _cached_kombu_conn[0]
+        if conn is not None:
+            try:
+                conn.close()
+            except Exception:
+                logger.debug("Failed to close Kombu connection", exc_info=True)
+            _cached_kombu_conn[0] = None
+
+    def _get_broker_redis() -> Redis:
+        client = _cached_client[0]
+        if client is not None:
+            try:
+                client.ping()
+                return client
+            except Exception:
+                logger.debug("Cached Redis client stale, reconnecting")
+                _close_client(client)
+                _cached_client[0] = None
+                _close_kombu_conn()
+
+        # Get a fresh Redis client from the broker connection.
+        # We hold this client long-term (cached above) rather than using a
+        # context manager, because we need it to persist across scrapes.
+        # The caching logic above ensures we only ever hold one connection,
+        # and we close it explicitly on reconnect.
+        conn = celery_app.broker_connection()
+        # kombu's Channel exposes .client at runtime (the underlying Redis
+        # client) but the type stubs don't declare it.
+        new_client: Redis = conn.channel().client  # type: ignore[attr-defined]
+        _cached_client[0] = new_client
+        _cached_kombu_conn[0] = conn
+        return new_client
+
+    return _get_broker_redis
+
+
 def setup_indexing_pipeline_metrics(celery_app: Celery) -> None:
     """Register all indexing pipeline collectors with the default registry.
 
     Args:
-        celery_app: The Celery application instance. Used to obtain a
+        celery_app: The Celery application instance. Used to obtain a fresh
             broker Redis client on each scrape for queue depth metrics.
     """
-    _queue_collector.set_celery_app(celery_app)
-    _redis_health_collector.set_celery_app(celery_app)
+    redis_factory = _make_broker_redis_factory(celery_app)
+    _queue_collector.set_redis_factory(redis_factory)
+    _redis_health_collector.set_redis_factory(redis_factory)
 
     # Start the heartbeat monitor daemon thread — uses a single persistent
     # connection to receive worker-heartbeat events.

backend/onyx/server/query_and_chat/chat_utils.py

@@ -9,8 +9,8 @@ def mime_type_to_chat_file_type(mime_type: str | None) -> ChatFileType:
     if mime_type in OnyxMimeTypes.IMAGE_MIME_TYPES:
         return ChatFileType.IMAGE
 
-    if mime_type in OnyxMimeTypes.CSV_MIME_TYPES:
-        return ChatFileType.CSV
+    if mime_type in OnyxMimeTypes.TABULAR_MIME_TYPES:
+        return ChatFileType.TABULAR
 
     if mime_type in OnyxMimeTypes.DOCUMENT_MIME_TYPES:
         return ChatFileType.DOC

backend/onyx/tools/tool_implementations/file_reader/file_reader_tool.py

@@ -169,10 +169,10 @@ class FileReaderTool(Tool[FileReaderToolOverrideKwargs]):
 
         chat_file = self._load_file(file_id)
 
-        # Only PLAIN_TEXT and CSV are guaranteed to contain actual text bytes.
+        # Only PLAIN_TEXT and TABULAR are guaranteed to contain actual text bytes.
         # DOC type in a loaded file means plaintext extraction failed and the
         # content is the original binary (e.g. raw PDF/DOCX bytes).
-        if chat_file.file_type not in (ChatFileType.PLAIN_TEXT, ChatFileType.CSV):
+        if chat_file.file_type not in (ChatFileType.PLAIN_TEXT, ChatFileType.TABULAR):
             raise ToolCallException(
                 message=f"File {file_id} is not a text file (type={chat_file.file_type})",
                 llm_facing_message=(

backend/requirements/default.txt

@@ -187,7 +187,7 @@ coloredlogs==15.0.1
     # via onnxruntime
 courlan==1.3.2
     # via trafilatura
-cryptography==46.0.6
+cryptography==46.0.5
     # via
     #   authlib
     #   google-auth
@@ -449,7 +449,7 @@ kombu==5.5.4
     # via celery
 kubernetes==31.0.0
     # via onyx
-langchain-core==1.2.22
+langchain-core==1.2.11
     # via onyx
 langdetect==1.0.9
     # via unstructured

backend/requirements/dev.txt

@@ -97,7 +97,7 @@ comm==0.2.3
     # via ipykernel
 contourpy==1.3.3
     # via matplotlib
-cryptography==46.0.6
+cryptography==46.0.5
     # via
     #   google-auth
     #   pyjwt
@@ -263,7 +263,7 @@ oauthlib==3.2.2
     # via
     #   kubernetes
     #   requests-oauthlib
-onyx-devtools==0.7.2
+onyx-devtools==0.7.1
     # via onyx
 openai==2.14.0
     # via

backend/requirements/ee.txt

@@ -76,7 +76,7 @@ colorama==0.4.6 ; sys_platform == 'win32'
     # via
     #   click
     #   tqdm
-cryptography==46.0.6
+cryptography==46.0.5
     # via
     #   google-auth
     #   pyjwt

backend/requirements/model_server.txt

@@ -92,7 +92,7 @@ colorama==0.4.6 ; sys_platform == 'win32'
     # via
     #   click
     #   tqdm
-cryptography==46.0.6
+cryptography==46.0.5
     # via
     #   google-auth
     #   pyjwt

backend/tests/external_dependency_unit/celery/test_persona_file_sync.py

@@ -129,10 +129,6 @@ def _patch_task_app(task: Any, mock_app: MagicMock) -> Generator[None, None, Non
             return_value=mock_app,
         ),
         patch(_PATCH_QUEUE_DEPTH, return_value=0),
-        patch(
-            "onyx.background.celery.tasks.user_file_processing.tasks.celery_get_broker_client",
-            return_value=MagicMock(),
-        ),
     ):
         yield
 

backend/tests/external_dependency_unit/celery/test_user_file_delete_queue.py

@@ -88,22 +88,10 @@ def _patch_task_app(task: Any, mock_app: MagicMock) -> Generator[None, None, Non
     the actual task instance.  We patch ``app`` on that instance's class
     (a unique Celery-generated Task subclass) so the mock is scoped to this
     task only.
-
-    Also patches ``celery_get_broker_client`` so the mock app doesn't need
-    a real broker URL.
     """
     task_instance = task.run.__self__
-    with (
-        patch.object(
-            type(task_instance),
-            "app",
-            new_callable=PropertyMock,
-            return_value=mock_app,
-        ),
-        patch(
-            "onyx.background.celery.tasks.user_file_processing.tasks.celery_get_broker_client",
-            return_value=MagicMock(),
-        ),
+    with patch.object(
+        type(task_instance), "app", new_callable=PropertyMock, return_value=mock_app
     ):
         yield
 

backend/tests/external_dependency_unit/celery/test_user_file_indexing_adapter.py

@@ -1,7 +1,7 @@
 """
 External dependency unit tests for UserFileIndexingAdapter metadata writing.
 
-Validates that prepare_enrichment produces DocMetadataAwareIndexChunk
+Validates that build_metadata_aware_chunks produces DocMetadataAwareIndexChunk
 objects with both `user_project` and `personas` fields populated correctly
 based on actual DB associations.
 
@@ -127,7 +127,7 @@ def _make_index_chunk(user_file: UserFile) -> IndexChunk:
 
 
 class TestAdapterWritesBothMetadataFields:
-    """prepare_enrichment must populate user_project AND personas."""
+    """build_metadata_aware_chunks must populate user_project AND personas."""
 
     @patch(
         "onyx.indexing.adapters.user_file_indexing_adapter.get_default_llm",
@@ -153,13 +153,15 @@ class TestAdapterWritesBothMetadataFields:
         doc = chunk.source_document
         context = DocumentBatchPrepareContext(updatable_docs=[doc], id_to_boost_map={})
 
-        enricher = adapter.prepare_enrichment(
-            context=context,
+        result = adapter.build_metadata_aware_chunks(
+            chunks_with_embeddings=[chunk],
+            chunk_content_scores=[1.0],
             tenant_id=TEST_TENANT_ID,
-            chunks=[chunk],
+            context=context,
         )
-        aware_chunk = enricher.enrich_chunk(chunk, 1.0)
 
+        assert len(result.chunks) == 1
+        aware_chunk = result.chunks[0]
         assert persona.id in aware_chunk.personas
         assert aware_chunk.user_project == []
 
@@ -188,13 +190,15 @@ class TestAdapterWritesBothMetadataFields:
             updatable_docs=[chunk.source_document], id_to_boost_map={}
         )
 
-        enricher = adapter.prepare_enrichment(
-            context=context,
+        result = adapter.build_metadata_aware_chunks(
+            chunks_with_embeddings=[chunk],
+            chunk_content_scores=[1.0],
             tenant_id=TEST_TENANT_ID,
-            chunks=[chunk],
+            context=context,
         )
-        aware_chunk = enricher.enrich_chunk(chunk, 1.0)
 
+        assert len(result.chunks) == 1
+        aware_chunk = result.chunks[0]
         assert project.id in aware_chunk.user_project
         assert aware_chunk.personas == []
 
@@ -225,13 +229,14 @@ class TestAdapterWritesBothMetadataFields:
             updatable_docs=[chunk.source_document], id_to_boost_map={}
         )
 
-        enricher = adapter.prepare_enrichment(
-            context=context,
+        result = adapter.build_metadata_aware_chunks(
+            chunks_with_embeddings=[chunk],
+            chunk_content_scores=[1.0],
             tenant_id=TEST_TENANT_ID,
-            chunks=[chunk],
+            context=context,
         )
-        aware_chunk = enricher.enrich_chunk(chunk, 1.0)
 
+        aware_chunk = result.chunks[0]
         assert persona.id in aware_chunk.personas
         assert project.id in aware_chunk.user_project
 
@@ -256,13 +261,14 @@ class TestAdapterWritesBothMetadataFields:
             updatable_docs=[chunk.source_document], id_to_boost_map={}
         )
 
-        enricher = adapter.prepare_enrichment(
-            context=context,
+        result = adapter.build_metadata_aware_chunks(
+            chunks_with_embeddings=[chunk],
+            chunk_content_scores=[1.0],
             tenant_id=TEST_TENANT_ID,
-            chunks=[chunk],
+            context=context,
         )
-        aware_chunk = enricher.enrich_chunk(chunk, 1.0)
 
+        aware_chunk = result.chunks[0]
         assert aware_chunk.personas == []
         assert aware_chunk.user_project == []
 
@@ -294,11 +300,12 @@ class TestAdapterWritesBothMetadataFields:
             updatable_docs=[chunk.source_document], id_to_boost_map={}
         )
 
-        enricher = adapter.prepare_enrichment(
-            context=context,
+        result = adapter.build_metadata_aware_chunks(
+            chunks_with_embeddings=[chunk],
+            chunk_content_scores=[1.0],
             tenant_id=TEST_TENANT_ID,
-            chunks=[chunk],
+            context=context,
         )
-        aware_chunk = enricher.enrich_chunk(chunk, 1.0)
 
+        aware_chunk = result.chunks[0]
         assert set(aware_chunk.personas) == {persona_a.id, persona_b.id}

backend/tests/external_dependency_unit/celery/test_user_file_processing_queue.py

@@ -90,17 +90,8 @@ def _patch_task_app(task: Any, mock_app: MagicMock) -> Generator[None, None, Non
     task only.
     """
     task_instance = task.run.__self__
-    with (
-        patch.object(
-            type(task_instance),
-            "app",
-            new_callable=PropertyMock,
-            return_value=mock_app,
-        ),
-        patch(
-            "onyx.background.celery.tasks.user_file_processing.tasks.celery_get_broker_client",
-            return_value=MagicMock(),
-        ),
+    with patch.object(
+        type(task_instance), "app", new_callable=PropertyMock, return_value=mock_app
     ):
         yield
 

backend/tests/external_dependency_unit/document_index/test_document_index.py

@@ -6,7 +6,6 @@ These tests assume Vespa and OpenSearch are running.
 import time
 import uuid
 from collections.abc import Generator
-from collections.abc import Iterator
 
 import httpx
 import pytest
@@ -22,7 +21,6 @@ from onyx.document_index.opensearch.opensearch_document_index import (
 )
 from onyx.document_index.vespa.index import VespaIndex
 from onyx.document_index.vespa.vespa_document_index import VespaDocumentIndex
-from onyx.indexing.models import DocMetadataAwareIndexChunk
 from tests.external_dependency_unit.constants import TEST_TENANT_ID
 from tests.external_dependency_unit.document_index.conftest import EMBEDDING_DIM
 from tests.external_dependency_unit.document_index.conftest import make_chunk
@@ -203,25 +201,3 @@ class TestDocumentIndexNew:
             assert len(result_map) == 2
             assert result_map[existing_doc] is True
             assert result_map[new_doc] is False
-
-    def test_index_accepts_generator(
-        self,
-        document_indices: list[DocumentIndexNew],
-        tenant_context: None,  # noqa: ARG002
-    ) -> None:
-        """index() accepts a generator (any iterable), not just a list."""
-        for document_index in document_indices:
-            doc_id = f"test_gen_{uuid.uuid4().hex[:8]}"
-            metadata = make_indexing_metadata([doc_id], old_counts=[0], new_counts=[3])
-
-            def chunk_gen() -> Iterator[DocMetadataAwareIndexChunk]:
-                for i in range(3):
-                    yield make_chunk(doc_id, chunk_id=i)
-
-            results = document_index.index(
-                chunks=chunk_gen(), indexing_metadata=metadata
-            )
-
-            assert len(results) == 1
-            assert results[0].document_id == doc_id
-            assert results[0].already_existed is False

backend/tests/external_dependency_unit/document_index/test_document_index_old.py

@@ -5,7 +5,6 @@ These tests assume Vespa and OpenSearch are running.
 
 import time
 from collections.abc import Generator
-from collections.abc import Iterator
 
 import pytest
 
@@ -167,29 +166,3 @@ class TestDocumentIndexOld:
                 batch_retrieval=True,
             )
             assert len(inference_chunks) == 0
-
-    def test_index_accepts_generator(
-        self,
-        document_indices: list[DocumentIndex],
-        tenant_context: None,  # noqa: ARG002
-    ) -> None:
-        """index() accepts a generator (any iterable), not just a list."""
-        for document_index in document_indices:
-
-            def chunk_gen() -> Iterator[DocMetadataAwareIndexChunk]:
-                for i in range(3):
-                    yield make_chunk("test_doc_gen", chunk_id=i)
-
-            index_batch_params = IndexBatchParams(
-                doc_id_to_previous_chunk_cnt={"test_doc_gen": 0},
-                doc_id_to_new_chunk_cnt={"test_doc_gen": 3},
-                tenant_id=get_current_tenant_id(),
-                large_chunks_enabled=False,
-            )
-
-            results = document_index.index(chunk_gen(), index_batch_params)
-
-            assert len(results) == 1
-            record = results.pop()
-            assert record.document_id == "test_doc_gen"
-            assert record.already_existed is False

backend/tests/external_dependency_unit/tools/test_python_tool.py

@@ -1175,7 +1175,7 @@ def test_code_interpreter_receives_chat_files(
 
     file_descriptor: FileDescriptor = {
         "id": user_file.file_id,
-        "type": ChatFileType.CSV,
+        "type": ChatFileType.TABULAR,
         "name": "data.csv",
         "user_file_id": str(user_file.id),
     }

backend/tests/unit/onyx/background/celery/test_celery_redis.py

@@ -1,87 +0,0 @@
-"""Tests for celery_get_broker_client singleton."""
-
-from collections.abc import Iterator
-from unittest.mock import MagicMock
-from unittest.mock import patch
-
-import pytest
-
-from onyx.background.celery import celery_redis
-
-
-@pytest.fixture(autouse=True)
-def reset_singleton() -> Iterator[None]:
-    """Reset the module-level singleton between tests."""
-    celery_redis._broker_client = None
-    celery_redis._broker_url = None
-    yield
-    celery_redis._broker_client = None
-    celery_redis._broker_url = None
-
-
-def _make_mock_app(broker_url: str = "redis://localhost:6379/15") -> MagicMock:
-    app = MagicMock()
-    app.conf.broker_url = broker_url
-    return app
-
-
-class TestCeleryGetBrokerClient:
-    @patch("onyx.background.celery.celery_redis.Redis")
-    def test_creates_client_on_first_call(self, mock_redis_cls: MagicMock) -> None:
-        mock_client = MagicMock()
-        mock_redis_cls.from_url.return_value = mock_client
-
-        app = _make_mock_app()
-        result = celery_redis.celery_get_broker_client(app)
-
-        assert result is mock_client
-        call_args = mock_redis_cls.from_url.call_args
-        assert call_args[0][0] == "redis://localhost:6379/15"
-        assert call_args[1]["decode_responses"] is False
-        assert call_args[1]["socket_keepalive"] is True
-        assert call_args[1]["retry_on_timeout"] is True
-
-    @patch("onyx.background.celery.celery_redis.Redis")
-    def test_reuses_cached_client(self, mock_redis_cls: MagicMock) -> None:
-        mock_client = MagicMock()
-        mock_client.ping.return_value = True
-        mock_redis_cls.from_url.return_value = mock_client
-
-        app = _make_mock_app()
-        client1 = celery_redis.celery_get_broker_client(app)
-        client2 = celery_redis.celery_get_broker_client(app)
-
-        assert client1 is client2
-        # from_url called only once
-        assert mock_redis_cls.from_url.call_count == 1
-
-    @patch("onyx.background.celery.celery_redis.Redis")
-    def test_reconnects_on_ping_failure(self, mock_redis_cls: MagicMock) -> None:
-        stale_client = MagicMock()
-        stale_client.ping.side_effect = ConnectionError("disconnected")
-
-        fresh_client = MagicMock()
-        fresh_client.ping.return_value = True
-
-        mock_redis_cls.from_url.side_effect = [stale_client, fresh_client]
-
-        app = _make_mock_app()
-
-        # First call creates stale_client
-        client1 = celery_redis.celery_get_broker_client(app)
-        assert client1 is stale_client
-
-        # Second call: ping fails, creates fresh_client
-        client2 = celery_redis.celery_get_broker_client(app)
-        assert client2 is fresh_client
-        assert mock_redis_cls.from_url.call_count == 2
-
-    @patch("onyx.background.celery.celery_redis.Redis")
-    def test_uses_broker_url_from_app_config(self, mock_redis_cls: MagicMock) -> None:
-        mock_redis_cls.from_url.return_value = MagicMock()
-
-        app = _make_mock_app("redis://custom-host:6380/3")
-        celery_redis.celery_get_broker_client(app)
-
-        call_args = mock_redis_cls.from_url.call_args
-        assert call_args[0][0] == "redis://custom-host:6380/3"

backend/tests/unit/onyx/chat/test_save_chat.py

@@ -139,7 +139,7 @@ def test_csv_file_type() -> None:
     result = _extract_referenced_file_descriptors([tool_call], message)
 
     assert len(result) == 1
-    assert result[0]["type"] == ChatFileType.CSV
+    assert result[0]["type"] == ChatFileType.TABULAR
 
 
 def test_unknown_extension_defaults_to_plain_text() -> None:

backend/tests/unit/onyx/document_index/opensearch/test_opensearch_batch_flush.py

@@ -1,223 +0,0 @@
-from unittest.mock import MagicMock
-from unittest.mock import patch
-
-from onyx.access.models import DocumentAccess
-from onyx.configs.constants import DocumentSource
-from onyx.connectors.models import Document
-from onyx.connectors.models import TextSection
-from onyx.document_index.interfaces_new import IndexingMetadata
-from onyx.document_index.interfaces_new import TenantState
-from onyx.document_index.opensearch.opensearch_document_index import (
-    OpenSearchDocumentIndex,
-)
-from onyx.indexing.models import ChunkEmbedding
-from onyx.indexing.models import DocMetadataAwareIndexChunk
-
-
-def _make_chunk(
-    doc_id: str,
-    chunk_id: int,
-) -> DocMetadataAwareIndexChunk:
-    """Creates a minimal DocMetadataAwareIndexChunk for testing."""
-    doc = Document(
-        id=doc_id,
-        sections=[TextSection(text="test", link="http://test.com")],
-        source=DocumentSource.FILE,
-        semantic_identifier="test_doc",
-        metadata={},
-    )
-    access = DocumentAccess.build(
-        user_emails=[],
-        user_groups=[],
-        external_user_emails=[],
-        external_user_group_ids=[],
-        is_public=True,
-    )
-    return DocMetadataAwareIndexChunk(
-        chunk_id=chunk_id,
-        blurb="test",
-        content="test content",
-        source_links={0: "http://test.com"},
-        image_file_id=None,
-        section_continuation=False,
-        source_document=doc,
-        title_prefix="",
-        metadata_suffix_semantic="",
-        metadata_suffix_keyword="",
-        mini_chunk_texts=None,
-        large_chunk_id=None,
-        doc_summary="",
-        chunk_context="",
-        contextual_rag_reserved_tokens=0,
-        embeddings=ChunkEmbedding(full_embedding=[0.1] * 10, mini_chunk_embeddings=[]),
-        title_embedding=[0.1] * 10,
-        tenant_id="test_tenant",
-        access=access,
-        document_sets=set(),
-        user_project=[],
-        personas=[],
-        boost=0,
-        aggregated_chunk_boost_factor=1.0,
-        ancestor_hierarchy_node_ids=[],
-    )
-
-
-def _make_index() -> tuple[OpenSearchDocumentIndex, MagicMock]:
-    """Creates an OpenSearchDocumentIndex with a mocked client.
-    Returns the index and the mock for bulk_index_documents."""
-    mock_client = MagicMock()
-    mock_bulk = MagicMock()
-    mock_client.bulk_index_documents = mock_bulk
-
-    tenant_state = TenantState(tenant_id="test_tenant", multitenant=False)
-
-    index = OpenSearchDocumentIndex.__new__(OpenSearchDocumentIndex)
-    index._index_name = "test_index"
-    index._client = mock_client
-    index._tenant_state = tenant_state
-
-    return index, mock_bulk
-
-
-def _make_metadata(doc_id: str, chunk_count: int) -> IndexingMetadata:
-    return IndexingMetadata(
-        doc_id_to_chunk_cnt_diff={
-            doc_id: IndexingMetadata.ChunkCounts(
-                old_chunk_cnt=0,
-                new_chunk_cnt=chunk_count,
-            ),
-        },
-    )
-
-
-@patch(
-    "onyx.document_index.opensearch.opensearch_document_index.MAX_CHUNKS_PER_DOC_BATCH",
-    100,
-)
-def test_single_doc_under_batch_limit_flushes_once() -> None:
-    """A document with fewer chunks than MAX_CHUNKS_PER_DOC_BATCH should flush once."""
-    index, mock_bulk = _make_index()
-    doc_id = "doc_1"
-    num_chunks = 50
-    chunks = [_make_chunk(doc_id, i) for i in range(num_chunks)]
-    metadata = _make_metadata(doc_id, num_chunks)
-
-    with patch.object(index, "delete", return_value=0):
-        index.index(chunks, metadata)
-
-    assert mock_bulk.call_count == 1
-    batch_arg = mock_bulk.call_args_list[0]
-    assert len(batch_arg.kwargs["documents"]) == num_chunks
-
-
-@patch(
-    "onyx.document_index.opensearch.opensearch_document_index.MAX_CHUNKS_PER_DOC_BATCH",
-    100,
-)
-def test_single_doc_over_batch_limit_flushes_multiple_times() -> None:
-    """A document with more chunks than MAX_CHUNKS_PER_DOC_BATCH should flush multiple times."""
-    index, mock_bulk = _make_index()
-    doc_id = "doc_1"
-    num_chunks = 250
-    chunks = [_make_chunk(doc_id, i) for i in range(num_chunks)]
-    metadata = _make_metadata(doc_id, num_chunks)
-
-    with patch.object(index, "delete", return_value=0):
-        index.index(chunks, metadata)
-
-    # 250 chunks / 100 per batch = 3 flushes (100 + 100 + 50)
-    assert mock_bulk.call_count == 3
-    batch_sizes = [len(call.kwargs["documents"]) for call in mock_bulk.call_args_list]
-    assert batch_sizes == [100, 100, 50]
-
-
-@patch(
-    "onyx.document_index.opensearch.opensearch_document_index.MAX_CHUNKS_PER_DOC_BATCH",
-    100,
-)
-def test_single_doc_exactly_at_batch_limit() -> None:
-    """A document with exactly MAX_CHUNKS_PER_DOC_BATCH chunks should flush once
-    (the flush happens on the next chunk, not at the boundary)."""
-    index, mock_bulk = _make_index()
-    doc_id = "doc_1"
-    num_chunks = 100
-    chunks = [_make_chunk(doc_id, i) for i in range(num_chunks)]
-    metadata = _make_metadata(doc_id, num_chunks)
-
-    with patch.object(index, "delete", return_value=0):
-        index.index(chunks, metadata)
-
-    # 100 chunks hit the >= check on chunk 101 which doesn't exist,
-    # so final flush handles all 100
-    # Actually: the elif fires when len(current_chunks) >= 100, which happens
-    # when current_chunks has 100 items and the 101st chunk arrives.
-    # With exactly 100 chunks, the 100th chunk makes len == 99, then appended -> 100.
-    # No 101st chunk arrives, so the final flush handles all 100.
-    assert mock_bulk.call_count == 1
-
-
-@patch(
-    "onyx.document_index.opensearch.opensearch_document_index.MAX_CHUNKS_PER_DOC_BATCH",
-    100,
-)
-def test_single_doc_one_over_batch_limit() -> None:
-    """101 chunks for one doc: first 100 flushed when the 101st arrives, then
-    the 101st is flushed at the end."""
-    index, mock_bulk = _make_index()
-    doc_id = "doc_1"
-    num_chunks = 101
-    chunks = [_make_chunk(doc_id, i) for i in range(num_chunks)]
-    metadata = _make_metadata(doc_id, num_chunks)
-
-    with patch.object(index, "delete", return_value=0):
-        index.index(chunks, metadata)
-
-    assert mock_bulk.call_count == 2
-    batch_sizes = [len(call.kwargs["documents"]) for call in mock_bulk.call_args_list]
-    assert batch_sizes == [100, 1]
-
-
-@patch(
-    "onyx.document_index.opensearch.opensearch_document_index.MAX_CHUNKS_PER_DOC_BATCH",
-    100,
-)
-def test_multiple_docs_each_under_limit_flush_per_doc() -> None:
-    """Multiple documents each under the batch limit should flush once per document."""
-    index, mock_bulk = _make_index()
-    chunks = []
-    for doc_idx in range(3):
-        doc_id = f"doc_{doc_idx}"
-        for chunk_idx in range(50):
-            chunks.append(_make_chunk(doc_id, chunk_idx))
-
-    metadata = IndexingMetadata(
-        doc_id_to_chunk_cnt_diff={
-            f"doc_{i}": IndexingMetadata.ChunkCounts(old_chunk_cnt=0, new_chunk_cnt=50)
-            for i in range(3)
-        },
-    )
-
-    with patch.object(index, "delete", return_value=0):
-        index.index(chunks, metadata)
-
-    # 3 documents = 3 flushes (one per doc boundary + final)
-    assert mock_bulk.call_count == 3
-
-
-@patch(
-    "onyx.document_index.opensearch.opensearch_document_index.MAX_CHUNKS_PER_DOC_BATCH",
-    100,
-)
-def test_delete_called_once_per_document() -> None:
-    """Even with multiple flushes for a single document, delete should only be
-    called once per document."""
-    index, _mock_bulk = _make_index()
-    doc_id = "doc_1"
-    num_chunks = 250
-    chunks = [_make_chunk(doc_id, i) for i in range(num_chunks)]
-    metadata = _make_metadata(doc_id, num_chunks)
-
-    with patch.object(index, "delete", return_value=0) as mock_delete:
-        index.index(chunks, metadata)
-
-    mock_delete.assert_called_once_with(doc_id, None)

backend/tests/unit/onyx/document_index/vespa/test_vespa_batch_flush.py

@@ -1,152 +0,0 @@
-"""Unit tests for VespaDocumentIndex.index().
-
-These tests mock all external I/O (HTTP calls, thread pools) and verify
-the streaming logic, ID cleaning/mapping, and DocumentInsertionRecord
-construction.
-"""
-
-from unittest.mock import MagicMock
-from unittest.mock import patch
-
-from onyx.access.models import DocumentAccess
-from onyx.configs.constants import DocumentSource
-from onyx.connectors.models import Document
-from onyx.connectors.models import TextSection
-from onyx.document_index.interfaces import EnrichedDocumentIndexingInfo
-from onyx.document_index.interfaces_new import IndexingMetadata
-from onyx.document_index.interfaces_new import TenantState
-from onyx.document_index.vespa.vespa_document_index import VespaDocumentIndex
-from onyx.indexing.models import ChunkEmbedding
-from onyx.indexing.models import DocMetadataAwareIndexChunk
-from onyx.indexing.models import IndexChunk
-
-
-def _make_chunk(
-    doc_id: str,
-    chunk_id: int = 0,
-    content: str = "test content",
-) -> DocMetadataAwareIndexChunk:
-    doc = Document(
-        id=doc_id,
-        semantic_identifier="test_doc",
-        sections=[TextSection(text=content, link=None)],
-        source=DocumentSource.NOT_APPLICABLE,
-        metadata={},
-    )
-    index_chunk = IndexChunk(
-        chunk_id=chunk_id,
-        blurb=content[:50],
-        content=content,
-        source_links=None,
-        image_file_id=None,
-        section_continuation=False,
-        source_document=doc,
-        title_prefix="",
-        metadata_suffix_semantic="",
-        metadata_suffix_keyword="",
-        contextual_rag_reserved_tokens=0,
-        doc_summary="",
-        chunk_context="",
-        mini_chunk_texts=None,
-        large_chunk_id=None,
-        embeddings=ChunkEmbedding(
-            full_embedding=[0.1] * 10,
-            mini_chunk_embeddings=[],
-        ),
-        title_embedding=None,
-    )
-    access = DocumentAccess.build(
-        user_emails=[],
-        user_groups=[],
-        external_user_emails=[],
-        external_user_group_ids=[],
-        is_public=True,
-    )
-    return DocMetadataAwareIndexChunk.from_index_chunk(
-        index_chunk=index_chunk,
-        access=access,
-        document_sets=set(),
-        user_project=[],
-        personas=[],
-        boost=0,
-        aggregated_chunk_boost_factor=1.0,
-        tenant_id="test_tenant",
-    )
-
-
-def _make_indexing_metadata(
-    doc_ids: list[str],
-    old_counts: list[int],
-    new_counts: list[int],
-) -> IndexingMetadata:
-    return IndexingMetadata(
-        doc_id_to_chunk_cnt_diff={
-            doc_id: IndexingMetadata.ChunkCounts(
-                old_chunk_cnt=old,
-                new_chunk_cnt=new,
-            )
-            for doc_id, old, new in zip(doc_ids, old_counts, new_counts)
-        }
-    )
-
-
-def _stub_enrich(
-    doc_id: str,
-    old_chunk_cnt: int,
-) -> EnrichedDocumentIndexingInfo:
-    """Build an EnrichedDocumentIndexingInfo that says 'no chunks to delete'
-    when old_chunk_cnt == 0, or 'has existing chunks' otherwise."""
-    return EnrichedDocumentIndexingInfo(
-        doc_id=doc_id,
-        chunk_start_index=0,
-        old_version=False,
-        chunk_end_index=old_chunk_cnt,
-    )
-
-
-@patch("onyx.document_index.vespa.vespa_document_index.batch_index_vespa_chunks")
-@patch("onyx.document_index.vespa.vespa_document_index.delete_vespa_chunks")
-@patch(
-    "onyx.document_index.vespa.vespa_document_index.get_document_chunk_ids",
-    return_value=[],
-)
-@patch("onyx.document_index.vespa.vespa_document_index._enrich_basic_chunk_info")
-@patch(
-    "onyx.document_index.vespa.vespa_document_index.BATCH_SIZE",
-    3,
-)
-def test_index_respects_batch_size(
-    mock_enrich: MagicMock,
-    mock_get_chunk_ids: MagicMock,  # noqa: ARG001
-    mock_delete: MagicMock,  # noqa: ARG001
-    mock_batch_index: MagicMock,
-) -> None:
-    """When chunks exceed BATCH_SIZE, batch_index_vespa_chunks is called
-    multiple times with correctly sized batches."""
-    mock_enrich.return_value = _stub_enrich("doc1", old_chunk_cnt=0)
-
-    index = VespaDocumentIndex(
-        index_name="test_index",
-        tenant_state=TenantState(tenant_id="test_tenant", multitenant=False),
-        large_chunks_enabled=False,
-        httpx_client=MagicMock(),
-    )
-
-    chunks = [_make_chunk("doc1", chunk_id=i) for i in range(7)]
-    metadata = _make_indexing_metadata(["doc1"], old_counts=[0], new_counts=[7])
-
-    results = index.index(chunks=chunks, indexing_metadata=metadata)
-
-    assert len(results) == 1
-
-    # With BATCH_SIZE=3 and 7 chunks: batches of 3, 3, 1
-    assert mock_batch_index.call_count == 3
-    batch_sizes = [len(c.kwargs["chunks"]) for c in mock_batch_index.call_args_list]
-    assert batch_sizes == [3, 3, 1]
-
-    # Verify all chunks are accounted for and in order
-    all_indexed = [
-        chunk for c in mock_batch_index.call_args_list for chunk in c.kwargs["chunks"]
-    ]
-    assert len(all_indexed) == 7
-    assert [c.chunk_id for c in all_indexed] == list(range(7))

backend/tests/unit/onyx/indexing/test_embed_chunks_in_batches.py

@@ -1,391 +0,0 @@
-"""Unit tests for _embed_chunks_to_store.
-
-Tests cover:
-  - Single batch, no failures
-  - Multiple batches, no failures
-  - Failure in a single batch
-  - Cross-batch document failure scrubbing
-  - Later batches skip already-failed docs
-  - Empty input
-  - All chunks fail
-"""
-
-from collections.abc import Callable
-from unittest.mock import MagicMock
-from unittest.mock import patch
-
-from onyx.connectors.models import ConnectorFailure
-from onyx.connectors.models import Document
-from onyx.connectors.models import DocumentFailure
-from onyx.connectors.models import DocumentSource
-from onyx.connectors.models import TextSection
-from onyx.indexing.chunk_batch_store import ChunkBatchStore
-from onyx.indexing.indexing_pipeline import _embed_chunks_to_store
-from onyx.indexing.models import ChunkEmbedding
-from onyx.indexing.models import DocAwareChunk
-from onyx.indexing.models import IndexChunk
-
-
-def _make_doc(doc_id: str) -> Document:
-    return Document(
-        id=doc_id,
-        semantic_identifier="test",
-        source=DocumentSource.FILE,
-        sections=[TextSection(text="test", link=None)],
-        metadata={},
-    )
-
-
-def _make_chunk(doc_id: str, chunk_id: int) -> DocAwareChunk:
-    return DocAwareChunk(
-        chunk_id=chunk_id,
-        blurb="test",
-        content="test content",
-        source_links=None,
-        image_file_id=None,
-        section_continuation=False,
-        source_document=_make_doc(doc_id),
-        title_prefix="",
-        metadata_suffix_semantic="",
-        metadata_suffix_keyword="",
-        mini_chunk_texts=None,
-        large_chunk_id=None,
-        doc_summary="",
-        chunk_context="",
-        contextual_rag_reserved_tokens=0,
-    )
-
-
-def _make_index_chunk(doc_id: str, chunk_id: int) -> IndexChunk:
-    """Create an IndexChunk (a DocAwareChunk with embeddings)."""
-    return IndexChunk(
-        chunk_id=chunk_id,
-        blurb="test",
-        content="test content",
-        source_links=None,
-        image_file_id=None,
-        section_continuation=False,
-        source_document=_make_doc(doc_id),
-        title_prefix="",
-        metadata_suffix_semantic="",
-        metadata_suffix_keyword="",
-        mini_chunk_texts=None,
-        large_chunk_id=None,
-        doc_summary="",
-        chunk_context="",
-        contextual_rag_reserved_tokens=0,
-        embeddings=ChunkEmbedding(
-            full_embedding=[0.1] * 10,
-            mini_chunk_embeddings=[],
-        ),
-        title_embedding=None,
-    )
-
-
-def _make_failure(doc_id: str) -> ConnectorFailure:
-    return ConnectorFailure(
-        failed_document=DocumentFailure(document_id=doc_id, document_link=None),
-        failure_message="embedding failed",
-        exception=RuntimeError("embedding failed"),
-    )
-
-
-def _mock_embed_success(
-    chunks: list[DocAwareChunk], **_kwargs: object
-) -> tuple[list[IndexChunk], list[ConnectorFailure]]:
-    """Simulate successful embedding of all chunks."""
-    return (
-        [_make_index_chunk(c.source_document.id, c.chunk_id) for c in chunks],
-        [],
-    )
-
-
-def _mock_embed_fail_doc(
-    fail_doc_id: str,
-) -> Callable[..., tuple[list[IndexChunk], list[ConnectorFailure]]]:
-    """Return an embed mock that fails all chunks for a specific doc."""
-
-    def _embed(
-        chunks: list[DocAwareChunk], **_kwargs: object
-    ) -> tuple[list[IndexChunk], list[ConnectorFailure]]:
-        successes = [
-            _make_index_chunk(c.source_document.id, c.chunk_id)
-            for c in chunks
-            if c.source_document.id != fail_doc_id
-        ]
-        failures = (
-            [_make_failure(fail_doc_id)]
-            if any(c.source_document.id == fail_doc_id for c in chunks)
-            else []
-        )
-        return successes, failures
-
-    return _embed
-
-
-class TestEmbedChunksInBatches:
-    @patch(
-        "onyx.indexing.indexing_pipeline.embed_chunks_with_failure_handling",
-    )
-    @patch("onyx.indexing.indexing_pipeline.MAX_CHUNKS_PER_DOC_BATCH", 100)
-    def test_single_batch_no_failures(self, mock_embed: MagicMock) -> None:
-        """All chunks fit in one batch and embed successfully."""
-        mock_embed.side_effect = _mock_embed_success
-
-        with ChunkBatchStore() as store:
-            chunks = [_make_chunk("doc1", i) for i in range(3)]
-            result = _embed_chunks_to_store(
-                chunks=chunks,
-                embedder=MagicMock(),
-                tenant_id="test",
-                request_id=None,
-                store=store,
-            )
-
-            assert len(result.successful_chunk_ids) == 3
-            assert len(result.connector_failures) == 0
-
-            # Verify stored contents
-            assert len(store._batch_files()) == 1
-            stored = list(store.stream())
-            assert len(stored) == 3
-
-    @patch(
-        "onyx.indexing.indexing_pipeline.embed_chunks_with_failure_handling",
-    )
-    @patch("onyx.indexing.indexing_pipeline.MAX_CHUNKS_PER_DOC_BATCH", 3)
-    def test_multiple_batches_no_failures(self, mock_embed: MagicMock) -> None:
-        """Chunks are split across multiple batches, all succeed."""
-        mock_embed.side_effect = _mock_embed_success
-
-        with ChunkBatchStore() as store:
-            chunks = [_make_chunk("doc1", i) for i in range(7)]
-            result = _embed_chunks_to_store(
-                chunks=chunks,
-                embedder=MagicMock(),
-                tenant_id="test",
-                request_id=None,
-                store=store,
-            )
-
-            assert len(result.successful_chunk_ids) == 7
-            assert len(result.connector_failures) == 0
-            assert len(store._batch_files()) == 3  # 3 + 3 + 1
-
-    @patch(
-        "onyx.indexing.indexing_pipeline.embed_chunks_with_failure_handling",
-    )
-    @patch("onyx.indexing.indexing_pipeline.MAX_CHUNKS_PER_DOC_BATCH", 100)
-    def test_single_batch_with_failure(self, mock_embed: MagicMock) -> None:
-        """One doc fails embedding, its chunks are excluded from results."""
-        mock_embed.side_effect = _mock_embed_fail_doc("doc2")
-
-        with ChunkBatchStore() as store:
-            chunks = [
-                _make_chunk("doc1", 0),
-                _make_chunk("doc2", 1),
-                _make_chunk("doc1", 2),
-            ]
-            result = _embed_chunks_to_store(
-                chunks=chunks,
-                embedder=MagicMock(),
-                tenant_id="test",
-                request_id=None,
-                store=store,
-            )
-
-            assert len(result.connector_failures) == 1
-            successful_doc_ids = {doc_id for _, doc_id in result.successful_chunk_ids}
-            assert "doc2" not in successful_doc_ids
-            assert "doc1" in successful_doc_ids
-
-    @patch(
-        "onyx.indexing.indexing_pipeline.embed_chunks_with_failure_handling",
-    )
-    @patch("onyx.indexing.indexing_pipeline.MAX_CHUNKS_PER_DOC_BATCH", 3)
-    def test_cross_batch_failure_scrubs_earlier_batch(
-        self, mock_embed: MagicMock
-    ) -> None:
-        """Doc A spans batches 0 and 1.  It succeeds in batch 0 but fails in
-        batch 1.  Its chunks should be scrubbed from batch 0's batch file."""
-        call_count = 0
-
-        def _embed(
-            chunks: list[DocAwareChunk], **_kwargs: object
-        ) -> tuple[list[IndexChunk], list[ConnectorFailure]]:
-            nonlocal call_count
-            call_count += 1
-            if call_count == 1:
-                return _mock_embed_success(chunks)
-            else:
-                return _mock_embed_fail_doc("docA")(chunks)
-
-        mock_embed.side_effect = _embed
-
-        with ChunkBatchStore() as store:
-            chunks = [
-                _make_chunk("docA", 0),
-                _make_chunk("docA", 1),
-                _make_chunk("docA", 2),
-                _make_chunk("docA", 3),
-                _make_chunk("docB", 0),
-                _make_chunk("docB", 1),
-            ]
-            result = _embed_chunks_to_store(
-                chunks=chunks,
-                embedder=MagicMock(),
-                tenant_id="test",
-                request_id=None,
-                store=store,
-            )
-
-            # docA should be fully excluded from results
-            successful_doc_ids = {doc_id for _, doc_id in result.successful_chunk_ids}
-            assert "docA" not in successful_doc_ids
-            assert "docB" in successful_doc_ids
-            assert len(result.connector_failures) == 1
-
-            # Verify batch 0 was scrubbed of docA chunks
-            all_stored = list(store.stream())
-            stored_doc_ids = {c.source_document.id for c in all_stored}
-            assert "docA" not in stored_doc_ids
-            assert "docB" in stored_doc_ids
-
-    @patch(
-        "onyx.indexing.indexing_pipeline.embed_chunks_with_failure_handling",
-    )
-    @patch("onyx.indexing.indexing_pipeline.MAX_CHUNKS_PER_DOC_BATCH", 3)
-    def test_later_batch_skips_already_failed_doc(self, mock_embed: MagicMock) -> None:
-        """If docA fails in batch 0, its chunks in batch 1 are skipped
-        entirely (never sent to the embedder)."""
-        embedded_doc_ids: list[str] = []
-
-        def _embed(
-            chunks: list[DocAwareChunk], **_kwargs: object
-        ) -> tuple[list[IndexChunk], list[ConnectorFailure]]:
-            for c in chunks:
-                embedded_doc_ids.append(c.source_document.id)
-            return _mock_embed_fail_doc("docA")(chunks)
-
-        mock_embed.side_effect = _embed
-
-        with ChunkBatchStore() as store:
-            chunks = [
-                _make_chunk("docA", 0),
-                _make_chunk("docA", 1),
-                _make_chunk("docA", 2),
-                _make_chunk("docA", 3),
-                _make_chunk("docB", 0),
-                _make_chunk("docB", 1),
-            ]
-            _embed_chunks_to_store(
-                chunks=chunks,
-                embedder=MagicMock(),
-                tenant_id="test",
-                request_id=None,
-                store=store,
-            )
-
-        # docA should only appear in batch 0, not batch 1
-        batch_1_doc_ids = embedded_doc_ids[3:]
-        assert "docA" not in batch_1_doc_ids
-
-    @patch(
-        "onyx.indexing.indexing_pipeline.embed_chunks_with_failure_handling",
-    )
-    @patch("onyx.indexing.indexing_pipeline.MAX_CHUNKS_PER_DOC_BATCH", 3)
-    def test_failed_doc_skipped_in_later_batch_while_other_doc_succeeds(
-        self, mock_embed: MagicMock
-    ) -> None:
-        """doc1 spans batches 0 and 1, doc2 only in batch 1.  Batch 0 fails
-        doc1.  In batch 1, doc1 chunks should be skipped but doc2 chunks
-        should still be embedded successfully."""
-        embedded_chunks: list[list[str]] = []
-
-        def _embed(
-            chunks: list[DocAwareChunk], **_kwargs: object
-        ) -> tuple[list[IndexChunk], list[ConnectorFailure]]:
-            embedded_chunks.append([c.source_document.id for c in chunks])
-            return _mock_embed_fail_doc("doc1")(chunks)
-
-        mock_embed.side_effect = _embed
-
-        with ChunkBatchStore() as store:
-            chunks = [
-                _make_chunk("doc1", 0),
-                _make_chunk("doc1", 1),
-                _make_chunk("doc1", 2),
-                _make_chunk("doc1", 3),
-                _make_chunk("doc2", 0),
-                _make_chunk("doc2", 1),
-            ]
-            result = _embed_chunks_to_store(
-                chunks=chunks,
-                embedder=MagicMock(),
-                tenant_id="test",
-                request_id=None,
-                store=store,
-            )
-
-            # doc1 should be fully excluded, doc2 fully included
-            successful_doc_ids = {doc_id for _, doc_id in result.successful_chunk_ids}
-            assert "doc1" not in successful_doc_ids
-            assert "doc2" in successful_doc_ids
-            assert len(result.successful_chunk_ids) == 2  # doc2's 2 chunks
-
-            # Batch 1 should only contain doc2 (doc1 was filtered before embedding)
-            assert len(embedded_chunks) == 2
-            assert "doc1" not in embedded_chunks[1]
-            assert embedded_chunks[1] == ["doc2", "doc2"]
-
-            # Verify on-disk state has no doc1 chunks
-            all_stored = list(store.stream())
-            assert all(c.source_document.id == "doc2" for c in all_stored)
-
-    @patch(
-        "onyx.indexing.indexing_pipeline.embed_chunks_with_failure_handling",
-    )
-    def test_empty_input(self, mock_embed: MagicMock) -> None:
-        """Empty chunk list produces empty results."""
-        mock_embed.side_effect = _mock_embed_success
-
-        with ChunkBatchStore() as store:
-            result = _embed_chunks_to_store(
-                chunks=[],
-                embedder=MagicMock(),
-                tenant_id="test",
-                request_id=None,
-                store=store,
-            )
-
-            assert len(result.successful_chunk_ids) == 0
-            assert len(result.connector_failures) == 0
-            mock_embed.assert_not_called()
-
-    @patch(
-        "onyx.indexing.indexing_pipeline.embed_chunks_with_failure_handling",
-    )
-    @patch("onyx.indexing.indexing_pipeline.MAX_CHUNKS_PER_DOC_BATCH", 100)
-    def test_all_chunks_fail(self, mock_embed: MagicMock) -> None:
-        """When all documents fail, results have no successful chunks."""
-
-        def _fail_all(
-            chunks: list[DocAwareChunk], **_kwargs: object
-        ) -> tuple[list[IndexChunk], list[ConnectorFailure]]:
-            doc_ids = {c.source_document.id for c in chunks}
-            return [], [_make_failure(doc_id) for doc_id in doc_ids]
-
-        mock_embed.side_effect = _fail_all
-
-        with ChunkBatchStore() as store:
-            chunks = [_make_chunk("doc1", 0), _make_chunk("doc2", 1)]
-            result = _embed_chunks_to_store(
-                chunks=chunks,
-                embedder=MagicMock(),
-                tenant_id="test",
-                request_id=None,
-                store=store,
-            )
-
-            assert len(result.successful_chunk_ids) == 0
-            assert len(result.connector_failures) == 2

backend/tests/unit/onyx/indexing/test_personas_in_chunks.py

@@ -116,7 +116,7 @@ def _run_adapter_build(
     project_ids_map: dict[str, list[int]],
     persona_ids_map: dict[str, list[int]],
 ) -> list[DocMetadataAwareIndexChunk]:
-    """Helper that runs UserFileIndexingAdapter.prepare_enrichment + enrich_chunk
+    """Helper that runs UserFileIndexingAdapter.build_metadata_aware_chunks
     with all external dependencies mocked."""
     from onyx.indexing.adapters.user_file_indexing_adapter import (
         UserFileIndexingAdapter,
@@ -155,16 +155,18 @@ def _run_adapter_build(
             side_effect=Exception("no LLM in tests"),
         ),
     ):
-        enricher = adapter.prepare_enrichment(
-            context=context,
+        result = adapter.build_metadata_aware_chunks(
+            chunks_with_embeddings=[chunk],
+            chunk_content_scores=[1.0],
             tenant_id="test_tenant",
-            chunks=[chunk],
+            context=context,
         )
-        return [enricher.enrich_chunk(chunk, 1.0)]
+
+    return result.chunks
 
 
-def test_prepare_enrichment_includes_persona_ids() -> None:
-    """UserFileIndexingAdapter.prepare_enrichment writes persona IDs
+def test_build_metadata_aware_chunks_includes_persona_ids() -> None:
+    """UserFileIndexingAdapter.build_metadata_aware_chunks writes persona IDs
     fetched from the DB into each chunk's metadata."""
     file_id = str(uuid4())
     persona_ids = [5, 12]
@@ -181,7 +183,7 @@ def test_prepare_enrichment_includes_persona_ids() -> None:
     assert chunks[0].user_project == project_ids
 
 
-def test_prepare_enrichment_missing_file_defaults_to_empty() -> None:
+def test_build_metadata_aware_chunks_missing_file_defaults_to_empty() -> None:
     """When a file has no persona or project associations in the DB, the
     adapter should default to empty lists (not KeyError or None)."""
     file_id = str(uuid4())

backend/tests/unit/onyx/llm/test_multi_llm.py

@@ -11,7 +11,6 @@ from litellm.types.utils import ChatCompletionDeltaToolCall
 from litellm.types.utils import Delta
 from litellm.types.utils import Function as LiteLLMFunction
 
-import onyx.llm.models
 from onyx.configs.app_configs import MOCK_LLM_RESPONSE
 from onyx.llm.constants import LlmProviderNames
 from onyx.llm.interfaces import LLMUserIdentity
@@ -1480,147 +1479,6 @@ def test_bifrost_normalizes_api_base_in_model_kwargs() -> None:
     assert llm._model_kwargs["api_base"] == "https://bifrost.example.com/v1"
 
 
-def test_prompt_contains_tool_call_history_true() -> None:
-    from onyx.llm.multi_llm import _prompt_contains_tool_call_history
-
-    messages: LanguageModelInput = [
-        UserMessage(content="What's the weather?"),
-        AssistantMessage(
-            content=None,
-            tool_calls=[
-                ToolCall(
-                    id="tc_1",
-                    function=FunctionCall(name="get_weather", arguments="{}"),
-                )
-            ],
-        ),
-    ]
-    assert _prompt_contains_tool_call_history(messages) is True
-
-
-def test_prompt_contains_tool_call_history_false_no_tools() -> None:
-    from onyx.llm.multi_llm import _prompt_contains_tool_call_history
-
-    messages: LanguageModelInput = [
-        UserMessage(content="Hello"),
-        AssistantMessage(content="Hi there!"),
-    ]
-    assert _prompt_contains_tool_call_history(messages) is False
-
-
-def test_prompt_contains_tool_call_history_false_user_only() -> None:
-    from onyx.llm.multi_llm import _prompt_contains_tool_call_history
-
-    messages: LanguageModelInput = [UserMessage(content="Hello")]
-    assert _prompt_contains_tool_call_history(messages) is False
-
-
-def test_bedrock_claude_drops_thinking_when_thinking_blocks_missing() -> None:
-    """When thinking is enabled but assistant messages with tool_calls lack
-    thinking_blocks, the thinking param must be dropped to avoid the Bedrock
-    BadRequestError about missing thinking blocks."""
-    llm = LitellmLLM(
-        api_key=None,
-        timeout=30,
-        model_provider=LlmProviderNames.BEDROCK,
-        model_name="anthropic.claude-sonnet-4-20250514-v1:0",
-        max_input_tokens=200000,
-    )
-
-    messages: LanguageModelInput = [
-        UserMessage(content="What's the weather?"),
-        AssistantMessage(
-            content=None,
-            tool_calls=[
-                ToolCall(
-                    id="tc_1",
-                    function=FunctionCall(
-                        name="get_weather",
-                        arguments='{"city": "Paris"}',
-                    ),
-                )
-            ],
-        ),
-        onyx.llm.models.ToolMessage(
-            content="22°C sunny",
-            tool_call_id="tc_1",
-        ),
-    ]
-
-    tools = [
-        {
-            "type": "function",
-            "function": {
-                "name": "get_weather",
-                "description": "Get the weather",
-                "parameters": {
-                    "type": "object",
-                    "properties": {"city": {"type": "string"}},
-                },
-            },
-        }
-    ]
-
-    with (
-        patch("litellm.completion") as mock_completion,
-        patch("onyx.llm.multi_llm.model_is_reasoning_model", return_value=True),
-    ):
-        mock_completion.return_value = []
-
-        list(llm.stream(messages, tools=tools, reasoning_effort=ReasoningEffort.HIGH))
-
-        kwargs = mock_completion.call_args.kwargs
-        assert "thinking" not in kwargs, (
-            "thinking param should be dropped when thinking_blocks are missing "
-            "from assistant messages with tool_calls"
-        )
-
-
-def test_bedrock_claude_keeps_thinking_when_no_tool_history() -> None:
-    """When thinking is enabled and there are no historical assistant messages
-    with tool_calls, the thinking param should be preserved."""
-    llm = LitellmLLM(
-        api_key=None,
-        timeout=30,
-        model_provider=LlmProviderNames.BEDROCK,
-        model_name="anthropic.claude-sonnet-4-20250514-v1:0",
-        max_input_tokens=200000,
-    )
-
-    messages: LanguageModelInput = [
-        UserMessage(content="What's the weather?"),
-    ]
-
-    tools = [
-        {
-            "type": "function",
-            "function": {
-                "name": "get_weather",
-                "description": "Get the weather",
-                "parameters": {
-                    "type": "object",
-                    "properties": {"city": {"type": "string"}},
-                },
-            },
-        }
-    ]
-
-    with (
-        patch("litellm.completion") as mock_completion,
-        patch("onyx.llm.multi_llm.model_is_reasoning_model", return_value=True),
-    ):
-        mock_completion.return_value = []
-
-        list(llm.stream(messages, tools=tools, reasoning_effort=ReasoningEffort.HIGH))
-
-        kwargs = mock_completion.call_args.kwargs
-        assert "thinking" in kwargs, (
-            "thinking param should be preserved when no assistant messages "
-            "with tool_calls exist in history"
-        )
-        assert kwargs["thinking"]["type"] == "enabled"
-
-
 def test_bifrost_claude_includes_allowed_openai_params() -> None:
     llm = LitellmLLM(
         api_key="test_key",

backend/tests/unit/onyx/tools/test_tool_runner_chat_files.py

@@ -82,7 +82,7 @@ class TestChatFileConversion:
             ChatLoadedFile(
                 file_id="file-2",
                 content=b"csv,data\n1,2",
-                file_type=ChatFileType.CSV,
+                file_type=ChatFileType.TABULAR,
                 filename="data.csv",
                 content_text="csv,data\n1,2",
                 token_count=5,

backend/tests/unit/server/metrics/test_indexing_pipeline_collectors.py

@@ -1,6 +1,5 @@
 """Tests for indexing pipeline Prometheus collectors."""
 
-from collections.abc import Iterator
 from datetime import datetime
 from datetime import timedelta
 from datetime import timezone
@@ -14,16 +13,6 @@ from onyx.server.metrics.indexing_pipeline import IndexAttemptCollector
 from onyx.server.metrics.indexing_pipeline import QueueDepthCollector
 
 
-@pytest.fixture(autouse=True)
-def _mock_broker_client() -> Iterator[None]:
-    """Patch celery_get_broker_client for all collector tests."""
-    with patch(
-        "onyx.background.celery.celery_redis.celery_get_broker_client",
-        return_value=MagicMock(),
-    ):
-        yield
-
-
 class TestQueueDepthCollector:
     def test_returns_empty_when_factory_not_set(self) -> None:
         collector = QueueDepthCollector()
@@ -35,7 +24,8 @@ class TestQueueDepthCollector:
 
     def test_collects_queue_depths(self) -> None:
         collector = QueueDepthCollector(cache_ttl=0)
-        collector.set_celery_app(MagicMock())
+        mock_redis = MagicMock()
+        collector.set_redis_factory(lambda: mock_redis)
 
         with (
             patch(
@@ -70,8 +60,8 @@ class TestQueueDepthCollector:
 
     def test_handles_redis_error_gracefully(self) -> None:
         collector = QueueDepthCollector(cache_ttl=0)
-        MagicMock()
-        collector.set_celery_app(MagicMock())
+        mock_redis = MagicMock()
+        collector.set_redis_factory(lambda: mock_redis)
 
         with patch(
             "onyx.server.metrics.indexing_pipeline.celery_get_queue_length",
@@ -84,8 +74,8 @@ class TestQueueDepthCollector:
 
     def test_caching_returns_stale_within_ttl(self) -> None:
         collector = QueueDepthCollector(cache_ttl=60)
-        MagicMock()
-        collector.set_celery_app(MagicMock())
+        mock_redis = MagicMock()
+        collector.set_redis_factory(lambda: mock_redis)
 
         with (
             patch(
@@ -108,10 +98,31 @@ class TestQueueDepthCollector:
 
         assert first is second  # Same object, from cache
 
+    def test_factory_called_each_scrape(self) -> None:
+        """Verify the Redis factory is called on each fresh collect, not cached."""
+        collector = QueueDepthCollector(cache_ttl=0)
+        factory = MagicMock(return_value=MagicMock())
+        collector.set_redis_factory(factory)
+
+        with (
+            patch(
+                "onyx.server.metrics.indexing_pipeline.celery_get_queue_length",
+                return_value=0,
+            ),
+            patch(
+                "onyx.server.metrics.indexing_pipeline.celery_get_unacked_task_ids",
+                return_value=set(),
+            ),
+        ):
+            collector.collect()
+            collector.collect()
+
+        assert factory.call_count == 2
+
     def test_error_returns_stale_cache(self) -> None:
         collector = QueueDepthCollector(cache_ttl=0)
-        MagicMock()
-        collector.set_celery_app(MagicMock())
+        mock_redis = MagicMock()
+        collector.set_redis_factory(lambda: mock_redis)
 
         # First call succeeds
         with (

backend/tests/unit/server/metrics/test_indexing_pipeline_setup.py

@@ -1,22 +1,96 @@
-"""Tests for indexing pipeline setup."""
+"""Tests for indexing pipeline setup (Redis factory caching)."""
 
 from unittest.mock import MagicMock
 
-from onyx.server.metrics.indexing_pipeline import QueueDepthCollector
-from onyx.server.metrics.indexing_pipeline import RedisHealthCollector
+from onyx.server.metrics.indexing_pipeline_setup import _make_broker_redis_factory
 
 
-class TestCollectorCeleryAppSetup:
-    def test_queue_depth_collector_uses_celery_app(self) -> None:
-        """QueueDepthCollector.set_celery_app stores the app for broker access."""
-        collector = QueueDepthCollector()
-        mock_app = MagicMock()
-        collector.set_celery_app(mock_app)
-        assert collector._celery_app is mock_app
+def _make_mock_app(client: MagicMock) -> MagicMock:
+    """Create a mock Celery app whose broker_connection().channel().client
+    returns the given client."""
+    mock_app = MagicMock()
+    mock_conn = MagicMock()
+    mock_conn.channel.return_value.client = client
 
-    def test_redis_health_collector_uses_celery_app(self) -> None:
-        """RedisHealthCollector.set_celery_app stores the app for broker access."""
-        collector = RedisHealthCollector()
-        mock_app = MagicMock()
-        collector.set_celery_app(mock_app)
-        assert collector._celery_app is mock_app
+    mock_app.broker_connection.return_value = mock_conn
+
+    return mock_app
+
+
+class TestMakeBrokerRedisFactory:
+    def test_caches_redis_client_across_calls(self) -> None:
+        """Factory should reuse the same client on subsequent calls."""
+        mock_client = MagicMock()
+        mock_client.ping.return_value = True
+        mock_app = _make_mock_app(mock_client)
+
+        factory = _make_broker_redis_factory(mock_app)
+
+        client1 = factory()
+        client2 = factory()
+
+        assert client1 is client2
+        # broker_connection should only be called once
+        assert mock_app.broker_connection.call_count == 1
+
+    def test_reconnects_when_ping_fails(self) -> None:
+        """Factory should create a new client if ping fails (stale connection)."""
+        mock_client_stale = MagicMock()
+        mock_client_stale.ping.side_effect = ConnectionError("disconnected")
+
+        mock_client_fresh = MagicMock()
+        mock_client_fresh.ping.return_value = True
+
+        mock_app = _make_mock_app(mock_client_stale)
+
+        factory = _make_broker_redis_factory(mock_app)
+
+        # First call — creates and caches
+        client1 = factory()
+        assert client1 is mock_client_stale
+        assert mock_app.broker_connection.call_count == 1
+
+        # Switch to fresh client for next connection
+        mock_conn_fresh = MagicMock()
+        mock_conn_fresh.channel.return_value.client = mock_client_fresh
+        mock_app.broker_connection.return_value = mock_conn_fresh
+
+        # Second call — ping fails on stale, reconnects
+        client2 = factory()
+        assert client2 is mock_client_fresh
+        assert mock_app.broker_connection.call_count == 2
+
+    def test_reconnect_closes_stale_client(self) -> None:
+        """When ping fails, the old client should be closed before reconnecting."""
+        mock_client_stale = MagicMock()
+        mock_client_stale.ping.side_effect = ConnectionError("disconnected")
+
+        mock_client_fresh = MagicMock()
+        mock_client_fresh.ping.return_value = True
+
+        mock_app = _make_mock_app(mock_client_stale)
+
+        factory = _make_broker_redis_factory(mock_app)
+
+        # First call — creates and caches
+        factory()
+
+        # Switch to fresh client
+        mock_conn_fresh = MagicMock()
+        mock_conn_fresh.channel.return_value.client = mock_client_fresh
+        mock_app.broker_connection.return_value = mock_conn_fresh
+
+        # Second call — ping fails, should close stale client
+        factory()
+        mock_client_stale.close.assert_called_once()
+
+    def test_first_call_creates_connection(self) -> None:
+        """First call should always create a new connection."""
+        mock_client = MagicMock()
+        mock_app = _make_mock_app(mock_client)
+
+        factory = _make_broker_redis_factory(mock_app)
+        client = factory()
+
+        assert client is mock_client
+        mock_app.broker_connection.assert_called_once()

cli/.gitignore

@@ -1,4 +1,3 @@
 onyx-cli
 cli
 onyx.cli
-__pycache__

cli/README.md

@@ -63,31 +63,6 @@ onyx-cli agents
 onyx-cli agents --json
 ```
 
-### Serve over SSH
-
-```shell
-# Start a public SSH endpoint for the CLI TUI
-onyx-cli serve --host 0.0.0.0 --port 2222
-
-# Connect as a client
-ssh your-host -p 2222
-```
-
-Clients can either:
-- paste an API key at the login prompt, or
-- skip the prompt by sending `ONYX_API_KEY` over SSH:
-
-```shell
-export ONYX_API_KEY=your-key
-ssh -o SendEnv=ONYX_API_KEY your-host -p 2222
-```
-
-Useful hardening flags:
-- `--idle-timeout` (default `15m`)
-- `--max-session-timeout` (default `8h`)
-- `--rate-limit-per-minute` (default `20`)
-- `--rate-limit-burst` (default `40`)
-
 ## Commands
 
 | Command | Description |
@@ -95,7 +70,6 @@ Useful hardening flags:
 | `chat` | Launch the interactive chat TUI (default) |
 | `ask` | Ask a one-shot question (non-interactive) |
 | `agents` | List available agents |
-| `serve` | Serve the interactive chat TUI over SSH |
 | `configure` | Configure server URL and API key |
 | `validate-config` | Validate configuration and test connection |
 

cli/cmd/root.go

@@ -1,17 +1,7 @@
 // Package cmd implements Cobra CLI commands for the Onyx CLI.
 package cmd
 
-import (
-	"context"
-	"fmt"
-	"time"
-
-	"github.com/onyx-dot-app/onyx/cli/internal/api"
-	"github.com/onyx-dot-app/onyx/cli/internal/config"
-	"github.com/onyx-dot-app/onyx/cli/internal/version"
-	log "github.com/sirupsen/logrus"
-	"github.com/spf13/cobra"
-)
+import "github.com/spf13/cobra"
 
 // Version and Commit are set via ldflags at build time.
 var (
@@ -26,69 +16,15 @@ func fullVersion() string {
 	return Version
 }
 
-func printVersion(cmd *cobra.Command) {
-	_, _ = fmt.Fprintf(cmd.OutOrStdout(), "Client version: %s\n", fullVersion())
-
-	cfg := config.Load()
-	if !cfg.IsConfigured() {
-		_, _ = fmt.Fprintf(cmd.OutOrStdout(), "Server version: unknown (not configured)\n")
-		return
-	}
-
-	client := api.NewClient(cfg)
-	ctx, cancel := context.WithTimeout(cmd.Context(), 5*time.Second)
-	defer cancel()
-
-	log.Debug("fetching backend version from /api/version")
-	backendVersion, err := client.GetBackendVersion(ctx)
-	if err != nil {
-		log.WithError(err).Debug("could not fetch backend version")
-		_, _ = fmt.Fprintf(cmd.OutOrStdout(), "Server version: unknown (could not reach server)\n")
-		return
-	}
-
-	if backendVersion == "" {
-		_, _ = fmt.Fprintf(cmd.OutOrStdout(), "Server version: unknown (empty response)\n")
-		return
-	}
-
-	_, _ = fmt.Fprintf(cmd.OutOrStdout(), "Server version: %s\n", backendVersion)
-
-	min := version.MinServer()
-	if sv, ok := version.Parse(backendVersion); ok && sv.LessThan(min) {
-		log.Warnf("Server version %s is below minimum required %d.%d, please upgrade",
-			backendVersion, min.Major, min.Minor)
-	}
-}
-
 // Execute creates and runs the root command.
 func Execute() error {
-	opts := struct {
-		Debug bool
-	}{}
-
 	rootCmd := &cobra.Command{
-		Use:   "onyx-cli",
-		Short: "Terminal UI for chatting with Onyx",
-		Long:  "Onyx CLI — a terminal interface for chatting with your Onyx agent.",
-		PersistentPreRun: func(cmd *cobra.Command, args []string) {
-			if opts.Debug {
-				log.SetLevel(log.DebugLevel)
-			} else {
-				log.SetLevel(log.InfoLevel)
-			}
-			log.SetFormatter(&log.TextFormatter{
-				DisableTimestamp: true,
-			})
-		},
+		Use:     "onyx-cli",
+		Short:   "Terminal UI for chatting with Onyx",
+		Long:    "Onyx CLI — a terminal interface for chatting with your Onyx agent.",
+		Version: fullVersion(),
 	}
 
-	rootCmd.PersistentFlags().BoolVar(&opts.Debug, "debug", false, "run in debug mode")
-
-	// Custom --version flag instead of Cobra's built-in (which only shows one version string)
-	var showVersion bool
-	rootCmd.Flags().BoolVarP(&showVersion, "version", "v", false, "Print client and server version information")
-
 	// Register subcommands
 	chatCmd := newChatCmd()
 	rootCmd.AddCommand(chatCmd)
@@ -96,16 +32,9 @@ func Execute() error {
 	rootCmd.AddCommand(newAgentsCmd())
 	rootCmd.AddCommand(newConfigureCmd())
 	rootCmd.AddCommand(newValidateConfigCmd())
-	rootCmd.AddCommand(newServeCmd())
 
-	// Default command is chat, but intercept --version first
-	rootCmd.RunE = func(cmd *cobra.Command, args []string) error {
-		if showVersion {
-			printVersion(cmd)
-			return nil
-		}
-		return chatCmd.RunE(cmd, args)
-	}
+	// Default command is chat
+	rootCmd.RunE = chatCmd.RunE
 
 	return rootCmd.Execute()
 }

cli/cmd/serve.go

@@ -1,450 +0,0 @@
-package cmd
-
-import (
-	"context"
-	"errors"
-	"fmt"
-	"net"
-	"os"
-	"os/signal"
-	"path/filepath"
-	"strings"
-	"syscall"
-	"time"
-
-	"github.com/charmbracelet/bubbles/textinput"
-	tea "github.com/charmbracelet/bubbletea"
-	"github.com/charmbracelet/log"
-	"github.com/charmbracelet/ssh"
-	"github.com/charmbracelet/wish"
-	"github.com/charmbracelet/wish/activeterm"
-	"github.com/charmbracelet/wish/bubbletea"
-	"github.com/charmbracelet/wish/logging"
-	"github.com/charmbracelet/wish/ratelimiter"
-	"github.com/onyx-dot-app/onyx/cli/internal/api"
-	"github.com/onyx-dot-app/onyx/cli/internal/config"
-	"github.com/onyx-dot-app/onyx/cli/internal/tui"
-	"github.com/spf13/cobra"
-	"golang.org/x/time/rate"
-)
-
-const (
-	defaultServeIdleTimeout        = 15 * time.Minute
-	defaultServeMaxSessionTimeout  = 8 * time.Hour
-	defaultServeRateLimitPerMinute = 20
-	defaultServeRateLimitBurst     = 40
-	defaultServeRateLimitCacheSize = 4096
-	maxAPIKeyLength                = 512
-	apiKeyValidationTimeout        = 15 * time.Second
-	maxAPIKeyRetries               = 5
-)
-
-func sessionEnv(s ssh.Session, key string) string {
-	prefix := key + "="
-	for _, env := range s.Environ() {
-		if strings.HasPrefix(env, prefix) {
-			return env[len(prefix):]
-		}
-	}
-	return ""
-}
-
-func validateAPIKey(serverURL string, apiKey string) error {
-	trimmedKey := strings.TrimSpace(apiKey)
-	if len(trimmedKey) > maxAPIKeyLength {
-		return fmt.Errorf("API key is too long (max %d characters)", maxAPIKeyLength)
-	}
-
-	cfg := config.OnyxCliConfig{
-		ServerURL: serverURL,
-		APIKey:    trimmedKey,
-	}
-	client := api.NewClient(cfg)
-	ctx, cancel := context.WithTimeout(context.Background(), apiKeyValidationTimeout)
-	defer cancel()
-	return client.TestConnection(ctx)
-}
-
-// --- auth prompt (bubbletea model) ---
-
-type authState int
-
-const (
-	authInput authState = iota
-	authValidating
-	authDone
-)
-
-type authValidatedMsg struct {
-	key string
-	err error
-}
-
-type authModel struct {
-	input     textinput.Model
-	serverURL string
-	state     authState
-	apiKey    string // set on successful validation
-	errMsg    string
-	retries   int
-	aborted   bool
-}
-
-func newAuthModel(serverURL, initialErr string) authModel {
-	ti := textinput.New()
-	ti.Prompt = "  API Key: "
-	ti.EchoMode = textinput.EchoPassword
-	ti.EchoCharacter = '•'
-	ti.CharLimit = maxAPIKeyLength
-	ti.Width = 80
-	ti.Focus()
-
-	return authModel{
-		input:     ti,
-		serverURL: serverURL,
-		errMsg:    initialErr,
-	}
-}
-
-func (m authModel) Update(msg tea.Msg) (authModel, tea.Cmd) {
-	switch msg := msg.(type) {
-	case tea.WindowSizeMsg:
-		m.input.Width = max(msg.Width-14, 20) // account for prompt width
-		return m, nil
-	case tea.KeyMsg:
-		switch msg.Type {
-		case tea.KeyCtrlC, tea.KeyCtrlD:
-			m.aborted = true
-			return m, nil
-		default:
-			if m.state == authValidating {
-				return m, nil
-			}
-		}
-		switch msg.Type {
-		case tea.KeyEnter:
-			key := strings.TrimSpace(m.input.Value())
-			if key == "" {
-				m.errMsg = "No key entered."
-				m.retries++
-				if m.retries >= maxAPIKeyRetries {
-					m.errMsg = "Too many failed attempts. Disconnecting."
-					m.aborted = true
-					return m, nil
-				}
-				m.input.SetValue("")
-				return m, nil
-			}
-			m.state = authValidating
-			m.errMsg = ""
-			serverURL := m.serverURL
-			return m, func() tea.Msg {
-				return authValidatedMsg{key: key, err: validateAPIKey(serverURL, key)}
-			}
-		}
-
-	case authValidatedMsg:
-		if msg.err != nil {
-			m.state = authInput
-			m.errMsg = msg.err.Error()
-			m.retries++
-			if m.retries >= maxAPIKeyRetries {
-				m.errMsg = "Too many failed attempts. Disconnecting."
-				m.aborted = true
-				return m, nil
-			}
-			m.input.SetValue("")
-			return m, m.input.Focus()
-		}
-		m.apiKey = msg.key
-		m.state = authDone
-		return m, nil
-	}
-
-	if m.state == authInput {
-		var cmd tea.Cmd
-		m.input, cmd = m.input.Update(msg)
-		return m, cmd
-	}
-	return m, nil
-}
-
-func (m authModel) View() string {
-	settingsURL := strings.TrimRight(m.serverURL, "/") + "/app/settings/accounts-access"
-
-	var b strings.Builder
-	b.WriteString("\n")
-	b.WriteString("  \x1b[1;35mOnyx CLI\x1b[0m\n")
-	b.WriteString("  \x1b[90m" + m.serverURL + "\x1b[0m\n")
-	b.WriteString("\n")
-	b.WriteString("  Generate an API key at:\n")
-	b.WriteString("  \x1b[4;34m" + settingsURL + "\x1b[0m\n")
-	b.WriteString("\n")
-	b.WriteString("  \x1b[90mTip: skip this prompt by passing your key via SSH:\x1b[0m\n")
-	b.WriteString("  \x1b[90m  export ONYX_API_KEY=<key>\x1b[0m\n")
-	b.WriteString("  \x1b[90m  ssh -o SendEnv=ONYX_API_KEY <host> -p <port>\x1b[0m\n")
-	b.WriteString("\n")
-
-	if m.errMsg != "" {
-		b.WriteString("  \x1b[1;31m" + m.errMsg + "\x1b[0m\n\n")
-	}
-
-	switch m.state {
-	case authDone:
-		b.WriteString("  \x1b[32mAuthenticated.\x1b[0m\n")
-	case authValidating:
-		b.WriteString("  \x1b[90mValidating…\x1b[0m\n")
-	default:
-		b.WriteString(m.input.View() + "\n")
-	}
-
-	return b.String()
-}
-
-// --- serve model (wraps auth → TUI in a single bubbletea program) ---
-
-type serveModel struct {
-	auth      authModel
-	tui       tea.Model
-	authed    bool
-	serverCfg config.OnyxCliConfig
-	width     int
-	height    int
-}
-
-func newServeModel(serverCfg config.OnyxCliConfig, initialErr string) serveModel {
-	return serveModel{
-		auth:      newAuthModel(serverCfg.ServerURL, initialErr),
-		serverCfg: serverCfg,
-	}
-}
-
-func (m serveModel) Init() tea.Cmd {
-	return textinput.Blink
-}
-
-func (m serveModel) Update(msg tea.Msg) (tea.Model, tea.Cmd) {
-	if !m.authed {
-		if ws, ok := msg.(tea.WindowSizeMsg); ok {
-			m.width = ws.Width
-			m.height = ws.Height
-		}
-
-		var cmd tea.Cmd
-		m.auth, cmd = m.auth.Update(msg)
-
-		if m.auth.aborted {
-			return m, tea.Quit
-		}
-		if m.auth.apiKey != "" {
-			cfg := config.OnyxCliConfig{
-				ServerURL:      m.serverCfg.ServerURL,
-				APIKey:         m.auth.apiKey,
-				DefaultAgentID: m.serverCfg.DefaultAgentID,
-			}
-			m.tui = tui.NewModel(cfg)
-			m.authed = true
-			w, h := m.width, m.height
-			return m, tea.Batch(
-				tea.EnterAltScreen,
-				tea.EnableMouseCellMotion,
-				m.tui.Init(),
-				func() tea.Msg { return tea.WindowSizeMsg{Width: w, Height: h} },
-			)
-		}
-		return m, cmd
-	}
-
-	var cmd tea.Cmd
-	m.tui, cmd = m.tui.Update(msg)
-	return m, cmd
-}
-
-func (m serveModel) View() string {
-	if !m.authed {
-		return m.auth.View()
-	}
-	return m.tui.View()
-}
-
-// --- serve command ---
-
-func newServeCmd() *cobra.Command {
-	var (
-		host              string
-		port              int
-		keyPath           string
-		idleTimeout       time.Duration
-		maxSessionTimeout time.Duration
-		rateLimitPerMin   int
-		rateLimitBurst    int
-		rateLimitCache    int
-	)
-
-	cmd := &cobra.Command{
-		Use:   "serve",
-		Short: "Serve the Onyx TUI over SSH",
-		Long: `Start an SSH server that presents the interactive Onyx chat TUI to
-connecting clients. Each SSH session gets its own independent TUI instance.
-
-Clients are prompted for their Onyx API key on connect. The key can also be
-provided via the ONYX_API_KEY environment variable to skip the prompt:
-
-  ssh -o SendEnv=ONYX_API_KEY host -p port
-
-The server URL is taken from the server operator's config. The server
-auto-generates an Ed25519 host key on first run if the key file does not
-already exist. The host key path can also be set via the ONYX_SSH_HOST_KEY
-environment variable (the --host-key flag takes precedence).
-
-Example:
-  onyx-cli serve --port 2222
-  ssh localhost -p 2222`,
-		RunE: func(cmd *cobra.Command, args []string) error {
-			serverCfg := config.Load()
-			if serverCfg.ServerURL == "" {
-				return fmt.Errorf("server URL is not configured; run 'onyx-cli configure' first")
-			}
-			if !cmd.Flags().Changed("host-key") {
-				if v := os.Getenv(config.EnvSSHHostKey); v != "" {
-					keyPath = v
-				}
-			}
-			if rateLimitPerMin <= 0 {
-				return fmt.Errorf("--rate-limit-per-minute must be > 0")
-			}
-			if rateLimitBurst <= 0 {
-				return fmt.Errorf("--rate-limit-burst must be > 0")
-			}
-			if rateLimitCache <= 0 {
-				return fmt.Errorf("--rate-limit-cache must be > 0")
-			}
-
-			addr := net.JoinHostPort(host, fmt.Sprintf("%d", port))
-			connectionLimiter := ratelimiter.NewRateLimiter(
-				rate.Limit(float64(rateLimitPerMin)/60.0),
-				rateLimitBurst,
-				rateLimitCache,
-			)
-
-			handler := func(s ssh.Session) (tea.Model, []tea.ProgramOption) {
-				apiKey := strings.TrimSpace(sessionEnv(s, config.EnvAPIKey))
-				var envErr string
-
-				if apiKey != "" {
-					if err := validateAPIKey(serverCfg.ServerURL, apiKey); err != nil {
-						envErr = fmt.Sprintf("ONYX_API_KEY from SSH environment is invalid: %s", err.Error())
-						apiKey = ""
-					}
-				}
-
-				if apiKey != "" {
-					// Env key is valid — go straight to the TUI.
-					cfg := config.OnyxCliConfig{
-						ServerURL:      serverCfg.ServerURL,
-						APIKey:         apiKey,
-						DefaultAgentID: serverCfg.DefaultAgentID,
-					}
-					return tui.NewModel(cfg), []tea.ProgramOption{
-						tea.WithAltScreen(),
-						tea.WithMouseCellMotion(),
-					}
-				}
-
-				// No valid env key — show auth prompt, then transition
-				// to the TUI within the same bubbletea program.
-				return newServeModel(serverCfg, envErr), []tea.ProgramOption{
-					tea.WithMouseCellMotion(),
-				}
-			}
-
-			serverOptions := []ssh.Option{
-				wish.WithAddress(addr),
-				wish.WithHostKeyPath(keyPath),
-				wish.WithMiddleware(
-					bubbletea.Middleware(handler),
-					activeterm.Middleware(),
-					ratelimiter.Middleware(connectionLimiter),
-					logging.Middleware(),
-				),
-			}
-			if idleTimeout > 0 {
-				serverOptions = append(serverOptions, wish.WithIdleTimeout(idleTimeout))
-			}
-			if maxSessionTimeout > 0 {
-				serverOptions = append(serverOptions, wish.WithMaxTimeout(maxSessionTimeout))
-			}
-
-			s, err := wish.NewServer(serverOptions...)
-			if err != nil {
-				return fmt.Errorf("could not create SSH server: %w", err)
-			}
-
-			done := make(chan os.Signal, 1)
-			signal.Notify(done, os.Interrupt, syscall.SIGTERM)
-
-			log.Info("Starting Onyx SSH server", "addr", addr)
-			log.Info("Connect with", "cmd", fmt.Sprintf("ssh %s -p %d", host, port))
-
-			errCh := make(chan error, 1)
-			go func() {
-				if err := s.ListenAndServe(); err != nil && !errors.Is(err, ssh.ErrServerClosed) {
-					log.Error("SSH server failed", "error", err)
-					errCh <- err
-				}
-			}()
-
-			var serverErr error
-			select {
-			case <-done:
-			case serverErr = <-errCh:
-			}
-
-			signal.Stop(done)
-			log.Info("Shutting down SSH server")
-			ctx, cancel := context.WithTimeout(context.Background(), 10*time.Second)
-			defer cancel()
-			if shutdownErr := s.Shutdown(ctx); shutdownErr != nil {
-				return errors.Join(serverErr, shutdownErr)
-			}
-			return serverErr
-		},
-	}
-
-	cmd.Flags().StringVar(&host, "host", "localhost", "Host address to bind to")
-	cmd.Flags().IntVarP(&port, "port", "p", 2222, "Port to listen on")
-	cmd.Flags().StringVar(&keyPath, "host-key", filepath.Join(config.ConfigDir(), "host_ed25519"),
-		"Path to SSH host key (auto-generated if missing)")
-	cmd.Flags().DurationVar(
-		&idleTimeout,
-		"idle-timeout",
-		defaultServeIdleTimeout,
-		"Disconnect idle clients after this duration (set 0 to disable)",
-	)
-	cmd.Flags().DurationVar(
-		&maxSessionTimeout,
-		"max-session-timeout",
-		defaultServeMaxSessionTimeout,
-		"Maximum lifetime of a client session (set 0 to disable)",
-	)
-	cmd.Flags().IntVar(
-		&rateLimitPerMin,
-		"rate-limit-per-minute",
-		defaultServeRateLimitPerMinute,
-		"Per-IP connection rate limit (new sessions per minute)",
-	)
-	cmd.Flags().IntVar(
-		&rateLimitBurst,
-		"rate-limit-burst",
-		defaultServeRateLimitBurst,
-		"Per-IP burst limit for connection attempts",
-	)
-	cmd.Flags().IntVar(
-		&rateLimitCache,
-		"rate-limit-cache",
-		defaultServeRateLimitCacheSize,
-		"Maximum number of IP limiter entries tracked in memory",
-	)
-
-	return cmd
-}

cli/cmd/validate.go

@@ -1,14 +1,10 @@
 package cmd
 
 import (
-	"context"
 	"fmt"
-	"time"
 
 	"github.com/onyx-dot-app/onyx/cli/internal/api"
 	"github.com/onyx-dot-app/onyx/cli/internal/config"
-	"github.com/onyx-dot-app/onyx/cli/internal/version"
-	log "github.com/sirupsen/logrus"
 	"github.com/spf13/cobra"
 )
 
@@ -39,25 +35,6 @@ func newValidateConfigCmd() *cobra.Command {
 			}
 
 			_, _ = fmt.Fprintln(cmd.OutOrStdout(), "Status:  connected and authenticated")
-
-			// Check backend version compatibility
-			vCtx, vCancel := context.WithTimeout(cmd.Context(), 5*time.Second)
-			defer vCancel()
-
-			backendVersion, err := client.GetBackendVersion(vCtx)
-			if err != nil {
-				log.WithError(err).Debug("could not fetch backend version")
-			} else if backendVersion == "" {
-				log.Debug("server returned empty version string")
-			} else {
-				_, _ = fmt.Fprintf(cmd.OutOrStdout(), "Version: %s\n", backendVersion)
-				min := version.MinServer()
-				if sv, ok := version.Parse(backendVersion); ok && sv.LessThan(min) {
-					log.Warnf("Server version %s is below minimum required %d.%d, please upgrade",
-						backendVersion, min.Major, min.Minor)
-				}
-			}
-
 			return nil
 		},
 	}

cli/go.mod

@@ -1,63 +1,45 @@
 module github.com/onyx-dot-app/onyx/cli
 
-go 1.26.1
+go 1.26.0
 
 require (
-	github.com/charmbracelet/bubbles v1.0.0
-	github.com/charmbracelet/bubbletea v1.3.10
-	github.com/charmbracelet/glamour v1.0.0
-	github.com/charmbracelet/lipgloss v1.1.1-0.20250404203927-76690c660834
-	github.com/charmbracelet/log v1.0.0
-	github.com/charmbracelet/ssh v0.0.0-20250826160808-ebfa259c7309
-	github.com/charmbracelet/wish v1.4.7
-	github.com/sirupsen/logrus v1.9.4
-	github.com/spf13/cobra v1.10.2
-	golang.org/x/term v0.41.0
-	golang.org/x/text v0.35.0
-	golang.org/x/time v0.15.0
+	github.com/charmbracelet/bubbles v0.20.0
+	github.com/charmbracelet/bubbletea v1.3.4
+	github.com/charmbracelet/glamour v0.8.0
+	github.com/charmbracelet/lipgloss v1.1.0
+	github.com/spf13/cobra v1.9.1
+	golang.org/x/term v0.30.0
+	golang.org/x/text v0.34.0
 )
 
 require (
-	github.com/alecthomas/chroma/v2 v2.23.1 // indirect
-	github.com/anmitsu/go-shlex v0.0.0-20200514113438-38f4b401e2be // indirect
+	github.com/alecthomas/chroma/v2 v2.14.0 // indirect
 	github.com/atotto/clipboard v0.1.4 // indirect
 	github.com/aymanbagabas/go-osc52/v2 v2.0.1 // indirect
 	github.com/aymerick/douceur v0.2.0 // indirect
-	github.com/charmbracelet/colorprofile v0.4.3 // indirect
-	github.com/charmbracelet/keygen v0.5.4 // indirect
-	github.com/charmbracelet/x/ansi v0.11.6 // indirect
-	github.com/charmbracelet/x/cellbuf v0.0.15 // indirect
-	github.com/charmbracelet/x/conpty v0.2.0 // indirect
-	github.com/charmbracelet/x/exp/slice v0.0.0-20260323091123-df7b1bcffcca // indirect
-	github.com/charmbracelet/x/input v0.3.7 // indirect
-	github.com/charmbracelet/x/term v0.2.2 // indirect
-	github.com/charmbracelet/x/termios v0.1.1 // indirect
-	github.com/charmbracelet/x/windows v0.2.2 // indirect
-	github.com/clipperhouse/displaywidth v0.11.0 // indirect
-	github.com/clipperhouse/uax29/v2 v2.7.0 // indirect
-	github.com/creack/pty v1.1.24 // indirect
-	github.com/dlclark/regexp2 v1.11.5 // indirect
+	github.com/charmbracelet/colorprofile v0.2.3-0.20250311203215-f60798e515dc // indirect
+	github.com/charmbracelet/x/ansi v0.8.0 // indirect
+	github.com/charmbracelet/x/cellbuf v0.0.13-0.20250311204145-2c3ea96c31dd // indirect
+	github.com/charmbracelet/x/term v0.2.1 // indirect
+	github.com/dlclark/regexp2 v1.11.0 // indirect
 	github.com/erikgeiser/coninput v0.0.0-20211004153227-1c3628e74d0f // indirect
-	github.com/go-logfmt/logfmt v0.6.1 // indirect
 	github.com/gorilla/css v1.0.1 // indirect
-	github.com/hashicorp/golang-lru/v2 v2.0.7 // indirect
 	github.com/inconshreveable/mousetrap v1.1.0 // indirect
-	github.com/lucasb-eyer/go-colorful v1.3.0 // indirect
+	github.com/lucasb-eyer/go-colorful v1.2.0 // indirect
 	github.com/mattn/go-isatty v0.0.20 // indirect
 	github.com/mattn/go-localereader v0.0.1 // indirect
-	github.com/mattn/go-runewidth v0.0.21 // indirect
+	github.com/mattn/go-runewidth v0.0.16 // indirect
 	github.com/microcosm-cc/bluemonday v1.0.27 // indirect
 	github.com/muesli/ansi v0.0.0-20230316100256-276c6243b2f6 // indirect
 	github.com/muesli/cancelreader v0.2.2 // indirect
 	github.com/muesli/reflow v0.3.0 // indirect
 	github.com/muesli/termenv v0.16.0 // indirect
 	github.com/rivo/uniseg v0.4.7 // indirect
-	github.com/spf13/pflag v1.0.10 // indirect
+	github.com/spf13/pflag v1.0.6 // indirect
 	github.com/xo/terminfo v0.0.0-20220910002029-abceb7e1c41e // indirect
-	github.com/yuin/goldmark v1.8.2 // indirect
-	github.com/yuin/goldmark-emoji v1.0.6 // indirect
-	golang.org/x/crypto v0.49.0 // indirect
-	golang.org/x/exp v0.0.0-20260312153236-7ab1446f8b90 // indirect
-	golang.org/x/net v0.52.0 // indirect
-	golang.org/x/sys v0.42.0 // indirect
+	github.com/yuin/goldmark v1.7.4 // indirect
+	github.com/yuin/goldmark-emoji v1.0.3 // indirect
+	golang.org/x/net v0.38.0 // indirect
+	golang.org/x/sync v0.19.0 // indirect
+	golang.org/x/sys v0.31.0 // indirect
 )

cli/go.sum

@@ -1,89 +1,55 @@
-github.com/alecthomas/assert/v2 v2.11.0 h1:2Q9r3ki8+JYXvGsDyBXwH3LcJ+WK5D0gc5E8vS6K3D0=
-github.com/alecthomas/assert/v2 v2.11.0/go.mod h1:Bze95FyfUr7x34QZrjL+XP+0qgp/zg8yS+TtBj1WA3k=
-github.com/alecthomas/chroma/v2 v2.23.1 h1:nv2AVZdTyClGbVQkIzlDm/rnhk1E9bU9nXwmZ/Vk/iY=
-github.com/alecthomas/chroma/v2 v2.23.1/go.mod h1:NqVhfBR0lte5Ouh3DcthuUCTUpDC9cxBOfyMbMQPs3o=
-github.com/alecthomas/repr v0.5.2 h1:SU73FTI9D1P5UNtvseffFSGmdNci/O6RsqzeXJtP0Qs=
-github.com/alecthomas/repr v0.5.2/go.mod h1:Fr0507jx4eOXV7AlPV6AVZLYrLIuIeSOWtW57eE/O/4=
-github.com/anmitsu/go-shlex v0.0.0-20200514113438-38f4b401e2be h1:9AeTilPcZAjCFIImctFaOjnTIavg87rW78vTPkQqLI8=
-github.com/anmitsu/go-shlex v0.0.0-20200514113438-38f4b401e2be/go.mod h1:ySMOLuWl6zY27l47sB3qLNK6tF2fkHG55UZxx8oIVo4=
+github.com/alecthomas/assert/v2 v2.7.0 h1:QtqSACNS3tF7oasA8CU6A6sXZSBDqnm7RfpLl9bZqbE=
+github.com/alecthomas/assert/v2 v2.7.0/go.mod h1:Bze95FyfUr7x34QZrjL+XP+0qgp/zg8yS+TtBj1WA3k=
+github.com/alecthomas/chroma/v2 v2.14.0 h1:R3+wzpnUArGcQz7fCETQBzO5n9IMNi13iIs46aU4V9E=
+github.com/alecthomas/chroma/v2 v2.14.0/go.mod h1:QolEbTfmUHIMVpBqxeDnNBj2uoeI4EbYP4i6n68SG4I=
+github.com/alecthomas/repr v0.4.0 h1:GhI2A8MACjfegCPVq9f1FLvIBS+DrQ2KQBFZP1iFzXc=
+github.com/alecthomas/repr v0.4.0/go.mod h1:Fr0507jx4eOXV7AlPV6AVZLYrLIuIeSOWtW57eE/O/4=
 github.com/atotto/clipboard v0.1.4 h1:EH0zSVneZPSuFR11BlR9YppQTVDbh5+16AmcJi4g1z4=
 github.com/atotto/clipboard v0.1.4/go.mod h1:ZY9tmq7sm5xIbd9bOK4onWV4S6X0u6GY7Vn0Yu86PYI=
 github.com/aymanbagabas/go-osc52/v2 v2.0.1 h1:HwpRHbFMcZLEVr42D4p7XBqjyuxQH5SMiErDT4WkJ2k=
 github.com/aymanbagabas/go-osc52/v2 v2.0.1/go.mod h1:uYgXzlJ7ZpABp8OJ+exZzJJhRNQ2ASbcXHWsFqH8hp8=
-github.com/aymanbagabas/go-udiff v0.3.1 h1:LV+qyBQ2pqe0u42ZsUEtPiCaUoqgA9gYRDs3vj1nolY=
-github.com/aymanbagabas/go-udiff v0.3.1/go.mod h1:G0fsKmG+P6ylD0r6N/KgQD/nWzgfnl8ZBcNLgcbrw8E=
+github.com/aymanbagabas/go-udiff v0.2.0 h1:TK0fH4MteXUDspT88n8CKzvK0X9O2xu9yQjWpi6yML8=
+github.com/aymanbagabas/go-udiff v0.2.0/go.mod h1:RE4Ex0qsGkTAJoQdQQCA0uG+nAzJO/pI/QwceO5fgrA=
 github.com/aymerick/douceur v0.2.0 h1:Mv+mAeH1Q+n9Fr+oyamOlAkUNPWPlA8PPGR0QAaYuPk=
 github.com/aymerick/douceur v0.2.0/go.mod h1:wlT5vV2O3h55X9m7iVYN0TBM0NH/MmbLnd30/FjWUq4=
-github.com/charmbracelet/bubbles v1.0.0 h1:12J8/ak/uCZEMQ6KU7pcfwceyjLlWsDLAxB5fXonfvc=
-github.com/charmbracelet/bubbles v1.0.0/go.mod h1:9d/Zd5GdnauMI5ivUIVisuEm3ave1XwXtD1ckyV6r3E=
-github.com/charmbracelet/bubbletea v1.3.10 h1:otUDHWMMzQSB0Pkc87rm691KZ3SWa4KUlvF9nRvCICw=
-github.com/charmbracelet/bubbletea v1.3.10/go.mod h1:ORQfo0fk8U+po9VaNvnV95UPWA1BitP1E0N6xJPlHr4=
-github.com/charmbracelet/colorprofile v0.4.3 h1:QPa1IWkYI+AOB+fE+mg/5/4HRMZcaXex9t5KX76i20Q=
-github.com/charmbracelet/colorprofile v0.4.3/go.mod h1:/zT4BhpD5aGFpqQQqw7a+VtHCzu+zrQtt1zhMt9mR4Q=
-github.com/charmbracelet/glamour v1.0.0 h1:AWMLOVFHTsysl4WV8T8QgkQ0s/ZNZo7CiE4WKhk8l08=
-github.com/charmbracelet/glamour v1.0.0/go.mod h1:DSdohgOBkMr2ZQNhw4LZxSGpx3SvpeujNoXrQyH2hxo=
-github.com/charmbracelet/keygen v0.5.4 h1:XQYgf6UEaTGgQSSmiPpIQ78WfseNQp4Pz8N/c1OsrdA=
-github.com/charmbracelet/keygen v0.5.4/go.mod h1:t4oBRr41bvK7FaJsAaAQhhkUuHslzFXVjOBwA55CZNM=
-github.com/charmbracelet/lipgloss v1.1.1-0.20250404203927-76690c660834 h1:ZR7e0ro+SZZiIZD7msJyA+NjkCNNavuiPBLgerbOziE=
-github.com/charmbracelet/lipgloss v1.1.1-0.20250404203927-76690c660834/go.mod h1:aKC/t2arECF6rNOnaKaVU6y4t4ZeHQzqfxedE/VkVhA=
-github.com/charmbracelet/log v1.0.0 h1:HVVVMmfOorfj3BA9i8X8UL69Hoz9lI0PYwXfJvOdRc4=
-github.com/charmbracelet/log v1.0.0/go.mod h1:uYgY3SmLpwJWxmlrPwXvzVYujxis1vAKRV/0VQB7yWA=
-github.com/charmbracelet/ssh v0.0.0-20250826160808-ebfa259c7309 h1:dCVbCRRtg9+tsfiTXTp0WupDlHruAXyp+YoxGVofHHc=
-github.com/charmbracelet/ssh v0.0.0-20250826160808-ebfa259c7309/go.mod h1:R9cISUs5kAH4Cq/rguNbSwcR+slE5Dfm8FEs//uoIGE=
-github.com/charmbracelet/wish v1.4.7 h1:O+jdLac3s6GaqkOHHSwezejNK04vl6VjO1A+hl8J8Yc=
-github.com/charmbracelet/wish v1.4.7/go.mod h1:OBZ8vC62JC5cvbxJLh+bIWtG7Ctmct+ewziuUWK+G14=
-github.com/charmbracelet/x/ansi v0.11.6 h1:GhV21SiDz/45W9AnV2R61xZMRri5NlLnl6CVF7ihZW8=
-github.com/charmbracelet/x/ansi v0.11.6/go.mod h1:2JNYLgQUsyqaiLovhU2Rv/pb8r6ydXKS3NIttu3VGZQ=
-github.com/charmbracelet/x/cellbuf v0.0.15 h1:ur3pZy0o6z/R7EylET877CBxaiE1Sp1GMxoFPAIztPI=
-github.com/charmbracelet/x/cellbuf v0.0.15/go.mod h1:J1YVbR7MUuEGIFPCaaZ96KDl5NoS0DAWkskup+mOY+Q=
-github.com/charmbracelet/x/conpty v0.2.0 h1:eKtA2hm34qNfgJCDp/M6Dc0gLy7e07YEK4qAdNGOvVY=
-github.com/charmbracelet/x/conpty v0.2.0/go.mod h1:fexgUnVrZgw8scD49f6VSi0Ggj9GWYIrpedRthAwW/8=
-github.com/charmbracelet/x/exp/golden v0.0.0-20241011142426-46044092ad91 h1:payRxjMjKgx2PaCWLZ4p3ro9y97+TVLZNaRZgJwSVDQ=
-github.com/charmbracelet/x/exp/golden v0.0.0-20241011142426-46044092ad91/go.mod h1:wDlXFlCrmJ8J+swcL/MnGUuYnqgQdW9rhSD61oNMb6U=
-github.com/charmbracelet/x/exp/slice v0.0.0-20260323091123-df7b1bcffcca h1:QQoyQLgUzojMNWHVHToN6d9qTvT0KWtxUKIRPx/Ox5o=
-github.com/charmbracelet/x/exp/slice v0.0.0-20260323091123-df7b1bcffcca/go.mod h1:vqEfX6xzqW1pKKZUUiFOKg0OQ7bCh54Q2vR/tserrRA=
-github.com/charmbracelet/x/input v0.3.7 h1:UzVbkt1vgM9dBQ+K+uRolBlN6IF2oLchmPKKo/aucXo=
-github.com/charmbracelet/x/input v0.3.7/go.mod h1:ZSS9Cia6Cycf2T6ToKIOxeTBTDwl25AGwArJuGaOBH8=
-github.com/charmbracelet/x/term v0.2.2 h1:xVRT/S2ZcKdhhOuSP4t5cLi5o+JxklsoEObBSgfgZRk=
-github.com/charmbracelet/x/term v0.2.2/go.mod h1:kF8CY5RddLWrsgVwpw4kAa6TESp6EB5y3uxGLeCqzAI=
-github.com/charmbracelet/x/termios v0.1.1 h1:o3Q2bT8eqzGnGPOYheoYS8eEleT5ZVNYNy8JawjaNZY=
-github.com/charmbracelet/x/termios v0.1.1/go.mod h1:rB7fnv1TgOPOyyKRJ9o+AsTU/vK5WHJ2ivHeut/Pcwo=
-github.com/charmbracelet/x/windows v0.2.2 h1:IofanmuvaxnKHuV04sC0eBy/smG6kIKrWG2/jYn2GuM=
-github.com/charmbracelet/x/windows v0.2.2/go.mod h1:/8XtdKZzedat74NQFn0NGlGL4soHB0YQZrETF96h75k=
-github.com/clipperhouse/displaywidth v0.11.0 h1:lBc6kY44VFw+TDx4I8opi/EtL9m20WSEFgwIwO+UVM8=
-github.com/clipperhouse/displaywidth v0.11.0/go.mod h1:bkrFNkf81G8HyVqmKGxsPufD3JhNl3dSqnGhOoSD/o0=
-github.com/clipperhouse/uax29/v2 v2.7.0 h1:+gs4oBZ2gPfVrKPthwbMzWZDaAFPGYK72F0NJv2v7Vk=
-github.com/clipperhouse/uax29/v2 v2.7.0/go.mod h1:EFJ2TJMRUaplDxHKj1qAEhCtQPW2tJSwu5BF98AuoVM=
+github.com/charmbracelet/bubbles v0.20.0 h1:jSZu6qD8cRQ6k9OMfR1WlM+ruM8fkPWkHvQWD9LIutE=
+github.com/charmbracelet/bubbles v0.20.0/go.mod h1:39slydyswPy+uVOHZ5x/GjwVAFkCsV8IIVy+4MhzwwU=
+github.com/charmbracelet/bubbletea v1.3.4 h1:kCg7B+jSCFPLYRA52SDZjr51kG/fMUEoPoZrkaDHyoI=
+github.com/charmbracelet/bubbletea v1.3.4/go.mod h1:dtcUCyCGEX3g9tosuYiut3MXgY/Jsv9nKVdibKKRRXo=
+github.com/charmbracelet/colorprofile v0.2.3-0.20250311203215-f60798e515dc h1:4pZI35227imm7yK2bGPcfpFEmuY1gc2YSTShr4iJBfs=
+github.com/charmbracelet/colorprofile v0.2.3-0.20250311203215-f60798e515dc/go.mod h1:X4/0JoqgTIPSFcRA/P6INZzIuyqdFY5rm8tb41s9okk=
+github.com/charmbracelet/glamour v0.8.0 h1:tPrjL3aRcQbn++7t18wOpgLyl8wrOHUEDS7IZ68QtZs=
+github.com/charmbracelet/glamour v0.8.0/go.mod h1:ViRgmKkf3u5S7uakt2czJ272WSg2ZenlYEZXT2x7Bjw=
+github.com/charmbracelet/lipgloss v1.1.0 h1:vYXsiLHVkK7fp74RkV7b2kq9+zDLoEU4MZoFqR/noCY=
+github.com/charmbracelet/lipgloss v1.1.0/go.mod h1:/6Q8FR2o+kj8rz4Dq0zQc3vYf7X+B0binUUBwA0aL30=
+github.com/charmbracelet/x/ansi v0.8.0 h1:9GTq3xq9caJW8ZrBTe0LIe2fvfLR/bYXKTx2llXn7xE=
+github.com/charmbracelet/x/ansi v0.8.0/go.mod h1:wdYl/ONOLHLIVmQaxbIYEC/cRKOQyjTkowiI4blgS9Q=
+github.com/charmbracelet/x/cellbuf v0.0.13-0.20250311204145-2c3ea96c31dd h1:vy0GVL4jeHEwG5YOXDmi86oYw2yuYUGqz6a8sLwg0X8=
+github.com/charmbracelet/x/cellbuf v0.0.13-0.20250311204145-2c3ea96c31dd/go.mod h1:xe0nKWGd3eJgtqZRaN9RjMtK7xUYchjzPr7q6kcvCCs=
+github.com/charmbracelet/x/exp/golden v0.0.0-20240815200342-61de596daa2b h1:MnAMdlwSltxJyULnrYbkZpp4k58Co7Tah3ciKhSNo0Q=
+github.com/charmbracelet/x/exp/golden v0.0.0-20240815200342-61de596daa2b/go.mod h1:wDlXFlCrmJ8J+swcL/MnGUuYnqgQdW9rhSD61oNMb6U=
+github.com/charmbracelet/x/term v0.2.1 h1:AQeHeLZ1OqSXhrAWpYUtZyX1T3zVxfpZuEQMIQaGIAQ=
+github.com/charmbracelet/x/term v0.2.1/go.mod h1:oQ4enTYFV7QN4m0i9mzHrViD7TQKvNEEkHUMCmsxdUg=
 github.com/cpuguy83/go-md2man/v2 v2.0.6/go.mod h1:oOW0eioCTA6cOiMLiUPZOpcVxMig6NIQQ7OS05n1F4g=
-github.com/creack/pty v1.1.24 h1:bJrF4RRfyJnbTJqzRLHzcGaZK1NeM5kTC9jGgovnR1s=
-github.com/creack/pty v1.1.24/go.mod h1:08sCNb52WyoAwi2QDyzUCTgcvVFhUzewun7wtTfvcwE=
-github.com/davecgh/go-spew v1.1.1 h1:vj9j/u1bqnvCEfJOwUhtlOARqs3+rkHYY13jYWTU97c=
-github.com/davecgh/go-spew v1.1.1/go.mod h1:J7Y8YcW2NihsgmVo/mv3lAwl/skON4iLHjSsI+c5H38=
-github.com/dlclark/regexp2 v1.11.5 h1:Q/sSnsKerHeCkc/jSTNq1oCm7KiVgUMZRDUoRu0JQZQ=
-github.com/dlclark/regexp2 v1.11.5/go.mod h1:DHkYz0B9wPfa6wondMfaivmHpzrQ3v9q8cnmRbL6yW8=
+github.com/dlclark/regexp2 v1.11.0 h1:G/nrcoOa7ZXlpoa/91N3X7mM3r8eIlMBBJZvsz/mxKI=
+github.com/dlclark/regexp2 v1.11.0/go.mod h1:DHkYz0B9wPfa6wondMfaivmHpzrQ3v9q8cnmRbL6yW8=
 github.com/erikgeiser/coninput v0.0.0-20211004153227-1c3628e74d0f h1:Y/CXytFA4m6baUTXGLOoWe4PQhGxaX0KpnayAqC48p4=
 github.com/erikgeiser/coninput v0.0.0-20211004153227-1c3628e74d0f/go.mod h1:vw97MGsxSvLiUE2X8qFplwetxpGLQrlU1Q9AUEIzCaM=
-github.com/go-logfmt/logfmt v0.6.1 h1:4hvbpePJKnIzH1B+8OR/JPbTx37NktoI9LE2QZBBkvE=
-github.com/go-logfmt/logfmt v0.6.1/go.mod h1:EV2pOAQoZaT1ZXZbqDl5hrymndi4SY9ED9/z6CO0XAk=
-github.com/google/go-cmp v0.7.0 h1:wk8382ETsv4JYUZwIsn6YpYiWiBsYLSJiTsyBybVuN8=
-github.com/google/go-cmp v0.7.0/go.mod h1:pXiqmnSA92OHEEa9HXL2W4E7lf9JzCmGVUdgjX3N/iU=
 github.com/gorilla/css v1.0.1 h1:ntNaBIghp6JmvWnxbZKANoLyuXTPZ4cAMlo6RyhlbO8=
 github.com/gorilla/css v1.0.1/go.mod h1:BvnYkspnSzMmwRK+b8/xgNPLiIuNZr6vbZBTPQ2A3b0=
-github.com/hashicorp/golang-lru/v2 v2.0.7 h1:a+bsQ5rvGLjzHuww6tVxozPZFVghXaHOwFs4luLUK2k=
-github.com/hashicorp/golang-lru/v2 v2.0.7/go.mod h1:QeFd9opnmA6QUJc5vARoKUSoFhyfM2/ZepoAG6RGpeM=
 github.com/hexops/gotextdiff v1.0.3 h1:gitA9+qJrrTCsiCl7+kh75nPqQt1cx4ZkudSTLoUqJM=
 github.com/hexops/gotextdiff v1.0.3/go.mod h1:pSWU5MAI3yDq+fZBTazCSJysOMbxWL1BSow5/V2vxeg=
 github.com/inconshreveable/mousetrap v1.1.0 h1:wN+x4NVGpMsO7ErUn/mUI3vEoE6Jt13X2s0bqwp9tc8=
 github.com/inconshreveable/mousetrap v1.1.0/go.mod h1:vpF70FUmC8bwa3OWnCshd2FqLfsEA9PFc4w1p2J65bw=
-github.com/lucasb-eyer/go-colorful v1.3.0 h1:2/yBRLdWBZKrf7gB40FoiKfAWYQ0lqNcbuQwVHXptag=
-github.com/lucasb-eyer/go-colorful v1.3.0/go.mod h1:R4dSotOR9KMtayYi1e77YzuveK+i7ruzyGqttikkLy0=
+github.com/lucasb-eyer/go-colorful v1.2.0 h1:1nnpGOrhyZZuNyfu1QjKiUICQ74+3FNCN69Aj6K7nkY=
+github.com/lucasb-eyer/go-colorful v1.2.0/go.mod h1:R4dSotOR9KMtayYi1e77YzuveK+i7ruzyGqttikkLy0=
 github.com/mattn/go-isatty v0.0.20 h1:xfD0iDuEKnDkl03q4limB+vH+GxLEtL/jb4xVJSWWEY=
 github.com/mattn/go-isatty v0.0.20/go.mod h1:W+V8PltTTMOvKvAeJH7IuucS94S2C6jfK/D7dTCTo3Y=
 github.com/mattn/go-localereader v0.0.1 h1:ygSAOl7ZXTx4RdPYinUpg6W99U8jWvWi9Ye2JC/oIi4=
 github.com/mattn/go-localereader v0.0.1/go.mod h1:8fBrzywKY7BI3czFoHkuzRoWE9C+EiG4R1k4Cjx5p88=
 github.com/mattn/go-runewidth v0.0.12/go.mod h1:RAqKPSqVFrSLVXbA8x7dzmKdmGzieGRCM46jaSJTDAk=
-github.com/mattn/go-runewidth v0.0.21 h1:jJKAZiQH+2mIinzCJIaIG9Be1+0NR+5sz/lYEEjdM8w=
-github.com/mattn/go-runewidth v0.0.21/go.mod h1:XBkDxAl56ILZc9knddidhrOlY5R/pDhgLpndooCuJAs=
+github.com/mattn/go-runewidth v0.0.16 h1:E5ScNMtiwvlvB5paMFdw9p4kSQzbXFikJ5SQO6TULQc=
+github.com/mattn/go-runewidth v0.0.16/go.mod h1:Jdepj2loyihRzMpdS35Xk/zdY8IAYHsh153qUoGf23w=
 github.com/microcosm-cc/bluemonday v1.0.27 h1:MpEUotklkwCSLeH+Qdx1VJgNqLlpY2KXwXFM08ygZfk=
 github.com/microcosm-cc/bluemonday v1.0.27/go.mod h1:jFi9vgW+H7c3V0lb6nR74Ib/DIB5OBs92Dimizgw2cA=
 github.com/muesli/ansi v0.0.0-20230316100256-276c6243b2f6 h1:ZK8zHtRHOkbHy6Mmr5D264iyp3TiX5OmNcI5cIARiQI=
@@ -94,47 +60,35 @@ github.com/muesli/reflow v0.3.0 h1:IFsN6K9NfGtjeggFP+68I4chLZV2yIKsXJFNZ+eWh6s=
 github.com/muesli/reflow v0.3.0/go.mod h1:pbwTDkVPibjO2kyvBQRBxTWEEGDGq0FlB1BIKtnHY/8=
 github.com/muesli/termenv v0.16.0 h1:S5AlUN9dENB57rsbnkPyfdGuWIlkmzJjbFf0Tf5FWUc=
 github.com/muesli/termenv v0.16.0/go.mod h1:ZRfOIKPFDYQoDFF4Olj7/QJbW60Ol/kL1pU3VfY/Cnk=
-github.com/pmezard/go-difflib v1.0.0 h1:4DBwDE0NGyQoBHbLQYPwSUPoCMWR5BEzIk/f1lZbAQM=
-github.com/pmezard/go-difflib v1.0.0/go.mod h1:iKH77koFhYxTK1pcRnkKkqfTogsbg7gZNVY4sRDYZ/4=
 github.com/rivo/uniseg v0.1.0/go.mod h1:J6wj4VEh+S6ZtnVlnTBMWIodfgj8LQOQFoIToxlJtxc=
 github.com/rivo/uniseg v0.2.0/go.mod h1:J6wj4VEh+S6ZtnVlnTBMWIodfgj8LQOQFoIToxlJtxc=
 github.com/rivo/uniseg v0.4.7 h1:WUdvkW8uEhrYfLC4ZzdpI2ztxP1I582+49Oc5Mq64VQ=
 github.com/rivo/uniseg v0.4.7/go.mod h1:FN3SvrM+Zdj16jyLfmOkMNblXMcoc8DfTHruCPUcx88=
 github.com/russross/blackfriday/v2 v2.1.0/go.mod h1:+Rmxgy9KzJVeS9/2gXHxylqXiyQDYRxCVz55jmeOWTM=
-github.com/sirupsen/logrus v1.9.4 h1:TsZE7l11zFCLZnZ+teH4Umoq5BhEIfIzfRDZ1Uzql2w=
-github.com/sirupsen/logrus v1.9.4/go.mod h1:ftWc9WdOfJ0a92nsE2jF5u5ZwH8Bv2zdeOC42RjbV2g=
-github.com/spf13/cobra v1.10.2 h1:DMTTonx5m65Ic0GOoRY2c16WCbHxOOw6xxezuLaBpcU=
-github.com/spf13/cobra v1.10.2/go.mod h1:7C1pvHqHw5A4vrJfjNwvOdzYu0Gml16OCs2GRiTUUS4=
-github.com/spf13/pflag v1.0.9/go.mod h1:McXfInJRrz4CZXVZOBLb0bTZqETkiAhM9Iw0y3An2Bg=
-github.com/spf13/pflag v1.0.10 h1:4EBh2KAYBwaONj6b2Ye1GiHfwjqyROoF4RwYO+vPwFk=
-github.com/spf13/pflag v1.0.10/go.mod h1:McXfInJRrz4CZXVZOBLb0bTZqETkiAhM9Iw0y3An2Bg=
-github.com/stretchr/testify v1.11.1 h1:7s2iGBzp5EwR7/aIZr8ao5+dra3wiQyKjjFuvgVKu7U=
-github.com/stretchr/testify v1.11.1/go.mod h1:wZwfW3scLgRK+23gO65QZefKpKQRnfz6sD981Nm4B6U=
+github.com/spf13/cobra v1.9.1 h1:CXSaggrXdbHK9CF+8ywj8Amf7PBRmPCOJugH954Nnlo=
+github.com/spf13/cobra v1.9.1/go.mod h1:nDyEzZ8ogv936Cinf6g1RU9MRY64Ir93oCnqb9wxYW0=
+github.com/spf13/pflag v1.0.6 h1:jFzHGLGAlb3ruxLB8MhbI6A8+AQX/2eW4qeyNZXNp2o=
+github.com/spf13/pflag v1.0.6/go.mod h1:McXfInJRrz4CZXVZOBLb0bTZqETkiAhM9Iw0y3An2Bg=
 github.com/xo/terminfo v0.0.0-20220910002029-abceb7e1c41e h1:JVG44RsyaB9T2KIHavMF/ppJZNG9ZpyihvCd0w101no=
 github.com/xo/terminfo v0.0.0-20220910002029-abceb7e1c41e/go.mod h1:RbqR21r5mrJuqunuUZ/Dhy/avygyECGrLceyNeo4LiM=
-github.com/yuin/goldmark v1.8.2 h1:kEGpgqJXdgbkhcOgBxkC0X0PmoPG1ZyoZ117rDVp4zE=
-github.com/yuin/goldmark v1.8.2/go.mod h1:ip/1k0VRfGynBgxOz0yCqHrbZXhcjxyuS66Brc7iBKg=
-github.com/yuin/goldmark-emoji v1.0.6 h1:QWfF2FYaXwL74tfGOW5izeiZepUDroDJfWubQI9HTHs=
-github.com/yuin/goldmark-emoji v1.0.6/go.mod h1:ukxJDKFpdFb5x0a5HqbdlcKtebh086iJpI31LTKmWuA=
-go.yaml.in/yaml/v3 v3.0.4/go.mod h1:DhzuOOF2ATzADvBadXxruRBLzYTpT36CKvDb3+aBEFg=
-golang.org/x/crypto v0.49.0 h1:+Ng2ULVvLHnJ/ZFEq4KdcDd/cfjrrjjNSXNzxg0Y4U4=
-golang.org/x/crypto v0.49.0/go.mod h1:ErX4dUh2UM+CFYiXZRTcMpEcN8b/1gxEuv3nODoYtCA=
-golang.org/x/exp v0.0.0-20260312153236-7ab1446f8b90 h1:jiDhWWeC7jfWqR9c/uplMOqJ0sbNlNWv0UkzE0vX1MA=
-golang.org/x/exp v0.0.0-20260312153236-7ab1446f8b90/go.mod h1:xE1HEv6b+1SCZ5/uscMRjUBKtIxworgEcEi+/n9NQDQ=
-golang.org/x/net v0.52.0 h1:He/TN1l0e4mmR3QqHMT2Xab3Aj3L9qjbhRm78/6jrW0=
-golang.org/x/net v0.52.0/go.mod h1:R1MAz7uMZxVMualyPXb+VaqGSa3LIaUqk0eEt3w36Sw=
-golang.org/x/sync v0.20.0 h1:e0PTpb7pjO8GAtTs2dQ6jYa5BWYlMuX047Dco/pItO4=
-golang.org/x/sync v0.20.0/go.mod h1:9xrNwdLfx4jkKbNva9FpL6vEN7evnE43NNNJQ2LF3+0=
+github.com/yuin/goldmark v1.7.1/go.mod h1:uzxRWxtg69N339t3louHJ7+O03ezfj6PlliRlaOzY1E=
+github.com/yuin/goldmark v1.7.4 h1:BDXOHExt+A7gwPCJgPIIq7ENvceR7we7rOS9TNoLZeg=
+github.com/yuin/goldmark v1.7.4/go.mod h1:uzxRWxtg69N339t3louHJ7+O03ezfj6PlliRlaOzY1E=
+github.com/yuin/goldmark-emoji v1.0.3 h1:aLRkLHOuBR2czCY4R8olwMjID+tENfhyFDMCRhbIQY4=
+github.com/yuin/goldmark-emoji v1.0.3/go.mod h1:tTkZEbwu5wkPmgTcitqddVxY9osFZiavD+r4AzQrh1U=
+golang.org/x/exp v0.0.0-20220909182711-5c715a9e8561 h1:MDc5xs78ZrZr3HMQugiXOAkSZtfTpbJLDr/lwfgO53E=
+golang.org/x/exp v0.0.0-20220909182711-5c715a9e8561/go.mod h1:cyybsKvd6eL0RnXn6p/Grxp8F5bW7iYuBgsNCOHpMYE=
+golang.org/x/net v0.38.0 h1:vRMAPTMaeGqVhG5QyLJHqNDwecKTomGeqbnfZyKlBI8=
+golang.org/x/net v0.38.0/go.mod h1:ivrbrMbzFq5J41QOQh0siUuly180yBYtLp+CKbEaFx8=
+golang.org/x/sync v0.19.0 h1:vV+1eWNmZ5geRlYjzm2adRgW2/mcpevXNg50YZtPCE4=
+golang.org/x/sync v0.19.0/go.mod h1:9KTHXmSnoGruLpwFjVSX0lNNA75CykiMECbovNTZqGI=
 golang.org/x/sys v0.0.0-20210809222454-d867a43fc93e/go.mod h1:oPkhp1MJrh7nUepCBck5+mAzfO9JrbApNNgaTdGDITg=
 golang.org/x/sys v0.6.0/go.mod h1:oPkhp1MJrh7nUepCBck5+mAzfO9JrbApNNgaTdGDITg=
-golang.org/x/sys v0.42.0 h1:omrd2nAlyT5ESRdCLYdm3+fMfNFE/+Rf4bDIQImRJeo=
-golang.org/x/sys v0.42.0/go.mod h1:4GL1E5IUh+htKOUEOaiffhrAeqysfVGipDYzABqnCmw=
-golang.org/x/term v0.41.0 h1:QCgPso/Q3RTJx2Th4bDLqML4W6iJiaXFq2/ftQF13YU=
-golang.org/x/term v0.41.0/go.mod h1:3pfBgksrReYfZ5lvYM0kSO0LIkAl4Yl2bXOkKP7Ec2A=
-golang.org/x/text v0.35.0 h1:JOVx6vVDFokkpaq1AEptVzLTpDe9KGpj5tR4/X+ybL8=
-golang.org/x/text v0.35.0/go.mod h1:khi/HExzZJ2pGnjenulevKNX1W67CUy0AsXcNubPGCA=
-golang.org/x/time v0.15.0 h1:bbrp8t3bGUeFOx08pvsMYRTCVSMk89u4tKbNOZbp88U=
-golang.org/x/time v0.15.0/go.mod h1:Y4YMaQmXwGQZoFaVFk4YpCt4FLQMYKZe9oeV/f4MSno=
+golang.org/x/sys v0.31.0 h1:ioabZlmFYtWhL+TRYpcnNlLwhyxaM9kWTDEmfnprqik=
+golang.org/x/sys v0.31.0/go.mod h1:BJP2sWEmIv4KK5OTEluFJCKSidICx8ciO85XgH3Ak8k=
+golang.org/x/term v0.30.0 h1:PQ39fJZ+mfadBm0y5WlL4vlM7Sx1Hgf13sMIY2+QS9Y=
+golang.org/x/term v0.30.0/go.mod h1:NYYFdzHoI5wRh/h5tDMdMqCqPJZEuNqVR5xJLd/n67g=
+golang.org/x/text v0.34.0 h1:oL/Qq0Kdaqxa1KbNeMKwQq0reLCCaFtqu2eNuSeNHbk=
+golang.org/x/text v0.34.0/go.mod h1:homfLqTYRFyVYemLBFl5GgL/DWEiH5wcsQ5gSh1yziA=
 gopkg.in/check.v1 v0.0.0-20161208181325-20d25e280405/go.mod h1:Co6ibVJAznAaIkqp8huTwlJQCZ016jof/cbN4VW5Yz0=
-gopkg.in/yaml.v3 v3.0.1 h1:fxVm/GzAzEWqLHuvctI91KS9hhNmmWOoWu0XTYJS7CA=
 gopkg.in/yaml.v3 v3.0.1/go.mod h1:K4uyk7z7BCEPqu6E+C64Yfv1cQ7kz7rIZviUmN+EgEM=

cli/hatch_build.py

@@ -34,7 +34,8 @@ class CustomBuildHook(BuildHookInterface):
         # Build the Go binary (always rebuild to ensure correct version injection)
         if not os.path.exists(binary_name):
             print(f"Building Go binary '{binary_name}'...")
-            ldflags = f"-X main.version={tag} -X main.commit={commit} -s -w"
+            pkg = "github.com/onyx-dot-app/onyx/cli/cmd"
+            ldflags = f"-X {pkg}.version={tag}" f" -X {pkg}.commit={commit}" " -s -w"
             subprocess.check_call(  # noqa: S603
                 ["go", "build", f"-ldflags={ldflags}", "-o", binary_name],
             )

cli/internal/api/client.go

@@ -270,17 +270,6 @@ func (c *Client) UploadFile(ctx context.Context, filePath string) (*models.FileD
 	}, nil
 }
 
-// GetBackendVersion fetches the backend version string from /api/version.
-func (c *Client) GetBackendVersion(ctx context.Context) (string, error) {
-	var resp struct {
-		BackendVersion string `json:"backend_version"`
-	}
-	if err := c.doJSON(ctx, "GET", "/api/version", nil, &resp); err != nil {
-		return "", err
-	}
-	return resp.BackendVersion, nil
-}
-
 // StopChatSession sends a stop signal for a streaming session (best-effort).
 func (c *Client) StopChatSession(ctx context.Context, sessionID string) {
 	req, err := c.newRequest(ctx, "POST", "/api/chat/stop-chat-session/"+sessionID, nil)

cli/internal/config/config.go

@@ -9,10 +9,9 @@ import (
 )
 
 const (
-	EnvServerURL  = "ONYX_SERVER_URL"
-	EnvAPIKey     = "ONYX_API_KEY"
+	EnvServerURL    = "ONYX_SERVER_URL"
+	EnvAPIKey = "ONYX_API_KEY"
 	EnvAgentID    = "ONYX_PERSONA_ID"
-	EnvSSHHostKey = "ONYX_SSH_HOST_KEY"
 )
 
 // OnyxCliConfig holds the CLI configuration.
@@ -36,8 +35,8 @@ func (c OnyxCliConfig) IsConfigured() bool {
 	return c.APIKey != ""
 }
 
-// ConfigDir returns ~/.config/onyx-cli
-func ConfigDir() string {
+// configDir returns ~/.config/onyx-cli
+func configDir() string {
 	if xdg := os.Getenv("XDG_CONFIG_HOME"); xdg != "" {
 		return filepath.Join(xdg, "onyx-cli")
 	}
@@ -50,7 +49,7 @@ func ConfigDir() string {
 
 // ConfigFilePath returns the full path to the config file.
 func ConfigFilePath() string {
-	return filepath.Join(ConfigDir(), "config.json")
+	return filepath.Join(configDir(), "config.json")
 }
 
 // ConfigExists checks if the config file exists on disk.
@@ -88,7 +87,7 @@ func Load() OnyxCliConfig {
 
 // Save writes the config to disk, creating parent directories if needed.
 func Save(cfg OnyxCliConfig) error {
-	dir := ConfigDir()
+	dir := configDir()
 	if err := os.MkdirAll(dir, 0o755); err != nil {
 		return err
 	}

cli/internal/version/version.go

@@ -1,58 +0,0 @@
-// Package version provides semver parsing and compatibility checks.
-package version
-
-import (
-	"strconv"
-	"strings"
-)
-
-// Semver holds parsed semantic version components.
-type Semver struct {
-	Major int
-	Minor int
-	Patch int
-}
-
-// minServer is the minimum backend version required by this CLI.
-var minServer = Semver{Major: 3, Minor: 0, Patch: 0}
-
-// MinServer returns the minimum backend version required by this CLI.
-func MinServer() Semver { return minServer }
-
-// Parse extracts major, minor, patch from a version string like "3.1.2" or "v3.1.2".
-// Returns ok=false if the string is not valid semver.
-func Parse(v string) (Semver, bool) {
-	v = strings.TrimPrefix(v, "v")
-	// Strip any pre-release suffix (e.g. "-beta.1") and build metadata (e.g. "+build.1")
-	if idx := strings.IndexAny(v, "-+"); idx != -1 {
-		v = v[:idx]
-	}
-	parts := strings.SplitN(v, ".", 3)
-	if len(parts) != 3 {
-		return Semver{}, false
-	}
-	major, err := strconv.Atoi(parts[0])
-	if err != nil {
-		return Semver{}, false
-	}
-	minor, err := strconv.Atoi(parts[1])
-	if err != nil {
-		return Semver{}, false
-	}
-	patch, err := strconv.Atoi(parts[2])
-	if err != nil {
-		return Semver{}, false
-	}
-	return Semver{Major: major, Minor: minor, Patch: patch}, true
-}
-
-// LessThan reports whether s is strictly less than other.
-func (s Semver) LessThan(other Semver) bool {
-	if s.Major != other.Major {
-		return s.Major < other.Major
-	}
-	if s.Minor != other.Minor {
-		return s.Minor < other.Minor
-	}
-	return s.Patch < other.Patch
-}

cli/pyproject.toml

@@ -1,5 +1,5 @@
 [build-system]
-requires = ["hatchling==1.29.0", "go-bin~=1.26.1", "manygo==0.2.0"]
+requires = ["hatchling", "go-bin~=1.24.11", "manygo"]
 build-backend = "hatchling.build"
 
 [project]

deployment/helm/charts/onyx/Chart.yaml

@@ -5,7 +5,7 @@ home: https://www.onyx.app/
 sources:
   - "https://github.com/onyx-dot-app/onyx"
 type: application
-version: 0.4.38
+version: 0.4.37
 appVersion: latest
 annotations:
   category: Productivity

deployment/helm/charts/onyx/templates/celery-worker-docfetching-metrics-service.yaml

@@ -1,26 +0,0 @@
-{{- /* Metrics port must match the default in metrics_server.py (_DEFAULT_PORTS).
-       Do NOT use PROMETHEUS_METRICS_PORT env var in Helm — each worker needs its own port. */ -}}
-{{- if and .Values.vectorDB.enabled (gt (int .Values.celery_worker_docfetching.replicaCount) 0) }}
-apiVersion: v1
-kind: Service
-metadata:
-  name: {{ include "onyx.fullname" . }}-celery-worker-docfetching-metrics
-  labels:
-    {{- include "onyx.labels" . | nindent 4 }}
-    {{- if .Values.celery_worker_docfetching.deploymentLabels }}
-    {{- toYaml .Values.celery_worker_docfetching.deploymentLabels | nindent 4 }}
-    {{- end }}
-    metrics: "true"
-spec:
-  type: ClusterIP
-  ports:
-    - port: 9092
-      targetPort: metrics
-      protocol: TCP
-      name: metrics
-  selector:
-    {{- include "onyx.selectorLabels" . | nindent 4 }}
-    {{- if .Values.celery_worker_docfetching.deploymentLabels }}
-    {{- toYaml .Values.celery_worker_docfetching.deploymentLabels | nindent 4 }}
-    {{- end }}
-{{- end }}

deployment/helm/charts/onyx/templates/celery-worker-docfetching.yaml

@@ -73,10 +73,6 @@ spec:
               "-Q",
               "connector_doc_fetching",
             ]
-          ports:
-            - name: metrics
-              containerPort: 9092
-              protocol: TCP
           resources:
             {{- toYaml .Values.celery_worker_docfetching.resources | nindent 12 }}
           envFrom:

deployment/helm/charts/onyx/templates/celery-worker-docprocessing-metrics-service.yaml

@@ -1,26 +0,0 @@
-{{- /* Metrics port must match the default in metrics_server.py (_DEFAULT_PORTS).
-       Do NOT use PROMETHEUS_METRICS_PORT env var in Helm — each worker needs its own port. */ -}}
-{{- if and .Values.vectorDB.enabled (gt (int .Values.celery_worker_docprocessing.replicaCount) 0) }}
-apiVersion: v1
-kind: Service
-metadata:
-  name: {{ include "onyx.fullname" . }}-celery-worker-docprocessing-metrics
-  labels:
-    {{- include "onyx.labels" . | nindent 4 }}
-    {{- if .Values.celery_worker_docprocessing.deploymentLabels }}
-    {{- toYaml .Values.celery_worker_docprocessing.deploymentLabels | nindent 4 }}
-    {{- end }}
-    metrics: "true"
-spec:
-  type: ClusterIP
-  ports:
-    - port: 9093
-      targetPort: metrics
-      protocol: TCP
-      name: metrics
-  selector:
-    {{- include "onyx.selectorLabels" . | nindent 4 }}
-    {{- if .Values.celery_worker_docprocessing.deploymentLabels }}
-    {{- toYaml .Values.celery_worker_docprocessing.deploymentLabels | nindent 4 }}
-    {{- end }}
-{{- end }}

deployment/helm/charts/onyx/templates/celery-worker-docprocessing.yaml

@@ -73,10 +73,6 @@ spec:
               "-Q",
               "docprocessing",
             ]
-          ports:
-            - name: metrics
-              containerPort: 9093
-              protocol: TCP
           resources:
             {{- toYaml .Values.celery_worker_docprocessing.resources | nindent 12 }}
           envFrom:

deployment/helm/charts/onyx/templates/celery-worker-monitoring-metrics-service.yaml

@@ -1,26 +0,0 @@
-{{- /* Metrics port must match the default in metrics_server.py (_DEFAULT_PORTS).
-       Do NOT use PROMETHEUS_METRICS_PORT env var in Helm — each worker needs its own port. */ -}}
-{{- if and .Values.vectorDB.enabled (gt (int .Values.celery_worker_monitoring.replicaCount) 0) }}
-apiVersion: v1
-kind: Service
-metadata:
-  name: {{ include "onyx.fullname" . }}-celery-worker-monitoring-metrics
-  labels:
-    {{- include "onyx.labels" . | nindent 4 }}
-    {{- if .Values.celery_worker_monitoring.deploymentLabels }}
-    {{- toYaml .Values.celery_worker_monitoring.deploymentLabels | nindent 4 }}
-    {{- end }}
-    metrics: "true"
-spec:
-  type: ClusterIP
-  ports:
-    - port: 9096
-      targetPort: metrics
-      protocol: TCP
-      name: metrics
-  selector:
-    {{- include "onyx.selectorLabels" . | nindent 4 }}
-    {{- if .Values.celery_worker_monitoring.deploymentLabels }}
-    {{- toYaml .Values.celery_worker_monitoring.deploymentLabels | nindent 4 }}
-    {{- end }}
-{{- end }}

deployment/helm/charts/onyx/templates/celery-worker-monitoring.yaml

@@ -70,10 +70,6 @@ spec:
               "-Q",
               "monitoring",
             ]
-          ports:
-            - name: metrics
-              containerPort: 9096
-              protocol: TCP
           resources:
             {{- toYaml .Values.celery_worker_monitoring.resources | nindent 12 }}
           envFrom:

pyproject.toml

@@ -66,7 +66,7 @@ backend = [
     "jsonref==1.1.0",
     "kubernetes==31.0.0",
     "trafilatura==1.12.2",
-    "langchain-core==1.2.22",
+    "langchain-core==1.2.11",
     "lazy_imports==1.0.1",
     "lxml==5.3.0",
     "Mako==1.2.4",
@@ -144,7 +144,7 @@ dev = [
     "matplotlib==3.10.8",
     "mypy-extensions==1.0.0",
     "mypy==1.13.0",
-    "onyx-devtools==0.7.2",
+    "onyx-devtools==0.7.1",
     "openapi-generator-cli==7.17.0",
     "pandas-stubs~=2.3.3",
     "pre-commit==3.2.2",

tools/ods/go.mod

@@ -1,6 +1,6 @@
 module github.com/onyx-dot-app/onyx/tools/ods
 
-go 1.26.1
+go 1.26.0
 
 require (
 	github.com/gdamore/tcell/v2 v2.13.8

tools/ods/pyproject.toml

@@ -1,5 +1,5 @@
 [build-system]
-requires = ["hatchling==1.29.0", "go-bin~=1.26.1", "manygo==0.2.0"]
+requires = ["hatchling", "go-bin~=1.26.0", "manygo"]
 build-backend = "hatchling.build"
 
 [project]

uv.lock

@@ -1255,61 +1255,61 @@ wheels = [
 
 [[package]]
 name = "cryptography"
-version = "46.0.6"
+version = "46.0.5"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "cffi", marker = "platform_python_implementation != 'PyPy'" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/a4/ba/04b1bd4218cbc58dc90ce967106d51582371b898690f3ae0402876cc4f34/cryptography-46.0.6.tar.gz", hash = "sha256:27550628a518c5c6c903d84f637fbecf287f6cb9ced3804838a1295dc1fd0759", size = 750542, upload-time = "2026-03-25T23:34:53.396Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/60/04/ee2a9e8542e4fa2773b81771ff8349ff19cdd56b7258a0cc442639052edb/cryptography-46.0.5.tar.gz", hash = "sha256:abace499247268e3757271b2f1e244b36b06f8515cf27c4d49468fc9eb16e93d", size = 750064, upload-time = "2026-02-10T19:18:38.255Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/47/23/9285e15e3bc57325b0a72e592921983a701efc1ee8f91c06c5f0235d86d9/cryptography-46.0.6-cp311-abi3-macosx_10_9_universal2.whl", hash = "sha256:64235194bad039a10bb6d2d930ab3323baaec67e2ce36215fd0952fad0930ca8", size = 7176401, upload-time = "2026-03-25T23:33:22.096Z" },
-    { url = "https://files.pythonhosted.org/packages/60/f8/e61f8f13950ab6195b31913b42d39f0f9afc7d93f76710f299b5ec286ae6/cryptography-46.0.6-cp311-abi3-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:26031f1e5ca62fcb9d1fcb34b2b60b390d1aacaa15dc8b895a9ed00968b97b30", size = 4275275, upload-time = "2026-03-25T23:33:23.844Z" },
-    { url = "https://files.pythonhosted.org/packages/19/69/732a736d12c2631e140be2348b4ad3d226302df63ef64d30dfdb8db7ad1c/cryptography-46.0.6-cp311-abi3-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:9a693028b9cbe51b5a1136232ee8f2bc242e4e19d456ded3fa7c86e43c713b4a", size = 4425320, upload-time = "2026-03-25T23:33:25.703Z" },
-    { url = "https://files.pythonhosted.org/packages/d4/12/123be7292674abf76b21ac1fc0e1af50661f0e5b8f0ec8285faac18eb99e/cryptography-46.0.6-cp311-abi3-manylinux_2_28_aarch64.whl", hash = "sha256:67177e8a9f421aa2d3a170c3e56eca4e0128883cf52a071a7cbf53297f18b175", size = 4278082, upload-time = "2026-03-25T23:33:27.423Z" },
-    { url = "https://files.pythonhosted.org/packages/5b/ba/d5e27f8d68c24951b0a484924a84c7cdaed7502bac9f18601cd357f8b1d2/cryptography-46.0.6-cp311-abi3-manylinux_2_28_ppc64le.whl", hash = "sha256:d9528b535a6c4f8ff37847144b8986a9a143585f0540fbcb1a98115b543aa463", size = 4926514, upload-time = "2026-03-25T23:33:29.206Z" },
-    { url = "https://files.pythonhosted.org/packages/34/71/1ea5a7352ae516d5512d17babe7e1b87d9db5150b21f794b1377eac1edc0/cryptography-46.0.6-cp311-abi3-manylinux_2_28_x86_64.whl", hash = "sha256:22259338084d6ae497a19bae5d4c66b7ca1387d3264d1c2c0e72d9e9b6a77b97", size = 4457766, upload-time = "2026-03-25T23:33:30.834Z" },
-    { url = "https://files.pythonhosted.org/packages/01/59/562be1e653accee4fdad92c7a2e88fced26b3fdfce144047519bbebc299e/cryptography-46.0.6-cp311-abi3-manylinux_2_31_armv7l.whl", hash = "sha256:760997a4b950ff00d418398ad73fbc91aa2894b5c1db7ccb45b4f68b42a63b3c", size = 3986535, upload-time = "2026-03-25T23:33:33.02Z" },
-    { url = "https://files.pythonhosted.org/packages/d6/8b/b1ebfeb788bf4624d36e45ed2662b8bd43a05ff62157093c1539c1288a18/cryptography-46.0.6-cp311-abi3-manylinux_2_34_aarch64.whl", hash = "sha256:3dfa6567f2e9e4c5dceb8ccb5a708158a2a871052fa75c8b78cb0977063f1507", size = 4277618, upload-time = "2026-03-25T23:33:34.567Z" },
-    { url = "https://files.pythonhosted.org/packages/dd/52/a005f8eabdb28df57c20f84c44d397a755782d6ff6d455f05baa2785bd91/cryptography-46.0.6-cp311-abi3-manylinux_2_34_ppc64le.whl", hash = "sha256:cdcd3edcbc5d55757e5f5f3d330dd00007ae463a7e7aa5bf132d1f22a4b62b19", size = 4890802, upload-time = "2026-03-25T23:33:37.034Z" },
-    { url = "https://files.pythonhosted.org/packages/ec/4d/8e7d7245c79c617d08724e2efa397737715ca0ec830ecb3c91e547302555/cryptography-46.0.6-cp311-abi3-manylinux_2_34_x86_64.whl", hash = "sha256:d4e4aadb7fc1f88687f47ca20bb7227981b03afaae69287029da08096853b738", size = 4457425, upload-time = "2026-03-25T23:33:38.904Z" },
-    { url = "https://files.pythonhosted.org/packages/1d/5c/f6c3596a1430cec6f949085f0e1a970638d76f81c3ea56d93d564d04c340/cryptography-46.0.6-cp311-abi3-musllinux_1_2_aarch64.whl", hash = "sha256:2b417edbe8877cda9022dde3a008e2deb50be9c407eef034aeeb3a8b11d9db3c", size = 4405530, upload-time = "2026-03-25T23:33:40.842Z" },
-    { url = "https://files.pythonhosted.org/packages/7e/c9/9f9cea13ee2dbde070424e0c4f621c091a91ffcc504ffea5e74f0e1daeff/cryptography-46.0.6-cp311-abi3-musllinux_1_2_x86_64.whl", hash = "sha256:380343e0653b1c9d7e1f55b52aaa2dbb2fdf2730088d48c43ca1c7c0abb7cc2f", size = 4667896, upload-time = "2026-03-25T23:33:42.781Z" },
-    { url = "https://files.pythonhosted.org/packages/ad/b5/1895bc0821226f129bc74d00eccfc6a5969e2028f8617c09790bf89c185e/cryptography-46.0.6-cp311-abi3-win32.whl", hash = "sha256:bcb87663e1f7b075e48c3be3ecb5f0b46c8fc50b50a97cf264e7f60242dca3f2", size = 3026348, upload-time = "2026-03-25T23:33:45.021Z" },
-    { url = "https://files.pythonhosted.org/packages/c3/f8/c9bcbf0d3e6ad288b9d9aa0b1dee04b063d19e8c4f871855a03ab3a297ab/cryptography-46.0.6-cp311-abi3-win_amd64.whl", hash = "sha256:6739d56300662c468fddb0e5e291f9b4d084bead381667b9e654c7dd81705124", size = 3483896, upload-time = "2026-03-25T23:33:46.649Z" },
-    { url = "https://files.pythonhosted.org/packages/01/41/3a578f7fd5c70611c0aacba52cd13cb364a5dee895a5c1d467208a9380b0/cryptography-46.0.6-cp314-cp314t-macosx_10_9_universal2.whl", hash = "sha256:2ef9e69886cbb137c2aef9772c2e7138dc581fad4fcbcf13cc181eb5a3ab6275", size = 7117147, upload-time = "2026-03-25T23:33:48.249Z" },
-    { url = "https://files.pythonhosted.org/packages/fa/87/887f35a6fca9dde90cad08e0de0c89263a8e59b2d2ff904fd9fcd8025b6f/cryptography-46.0.6-cp314-cp314t-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:7f417f034f91dcec1cb6c5c35b07cdbb2ef262557f701b4ecd803ee8cefed4f4", size = 4266221, upload-time = "2026-03-25T23:33:49.874Z" },
-    { url = "https://files.pythonhosted.org/packages/aa/a8/0a90c4f0b0871e0e3d1ed126aed101328a8a57fd9fd17f00fb67e82a51ca/cryptography-46.0.6-cp314-cp314t-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:d24c13369e856b94892a89ddf70b332e0b70ad4a5c43cf3e9cb71d6d7ffa1f7b", size = 4408952, upload-time = "2026-03-25T23:33:52.128Z" },
-    { url = "https://files.pythonhosted.org/packages/16/0b/b239701eb946523e4e9f329336e4ff32b1247e109cbab32d1a7b61da8ed7/cryptography-46.0.6-cp314-cp314t-manylinux_2_28_aarch64.whl", hash = "sha256:aad75154a7ac9039936d50cf431719a2f8d4ed3d3c277ac03f3339ded1a5e707", size = 4270141, upload-time = "2026-03-25T23:33:54.11Z" },
-    { url = "https://files.pythonhosted.org/packages/0f/a8/976acdd4f0f30df7b25605f4b9d3d89295351665c2091d18224f7ad5cdbf/cryptography-46.0.6-cp314-cp314t-manylinux_2_28_ppc64le.whl", hash = "sha256:3c21d92ed15e9cfc6eb64c1f5a0326db22ca9c2566ca46d845119b45b4400361", size = 4904178, upload-time = "2026-03-25T23:33:55.725Z" },
-    { url = "https://files.pythonhosted.org/packages/b1/1b/bf0e01a88efd0e59679b69f42d4afd5bced8700bb5e80617b2d63a3741af/cryptography-46.0.6-cp314-cp314t-manylinux_2_28_x86_64.whl", hash = "sha256:4668298aef7cddeaf5c6ecc244c2302a2b8e40f384255505c22875eebb47888b", size = 4441812, upload-time = "2026-03-25T23:33:57.364Z" },
-    { url = "https://files.pythonhosted.org/packages/bb/8b/11df86de2ea389c65aa1806f331cae145f2ed18011f30234cc10ca253de8/cryptography-46.0.6-cp314-cp314t-manylinux_2_31_armv7l.whl", hash = "sha256:8ce35b77aaf02f3b59c90b2c8a05c73bac12cea5b4e8f3fbece1f5fddea5f0ca", size = 3963923, upload-time = "2026-03-25T23:33:59.361Z" },
-    { url = "https://files.pythonhosted.org/packages/91/e0/207fb177c3a9ef6a8108f234208c3e9e76a6aa8cf20d51932916bd43bda0/cryptography-46.0.6-cp314-cp314t-manylinux_2_34_aarch64.whl", hash = "sha256:c89eb37fae9216985d8734c1afd172ba4927f5a05cfd9bf0e4863c6d5465b013", size = 4269695, upload-time = "2026-03-25T23:34:00.909Z" },
-    { url = "https://files.pythonhosted.org/packages/21/5e/19f3260ed1e95bced52ace7501fabcd266df67077eeb382b79c81729d2d3/cryptography-46.0.6-cp314-cp314t-manylinux_2_34_ppc64le.whl", hash = "sha256:ed418c37d095aeddf5336898a132fba01091f0ac5844e3e8018506f014b6d2c4", size = 4869785, upload-time = "2026-03-25T23:34:02.796Z" },
-    { url = "https://files.pythonhosted.org/packages/10/38/cd7864d79aa1d92ef6f1a584281433419b955ad5a5ba8d1eb6c872165bcb/cryptography-46.0.6-cp314-cp314t-manylinux_2_34_x86_64.whl", hash = "sha256:69cf0056d6947edc6e6760e5f17afe4bea06b56a9ac8a06de9d2bd6b532d4f3a", size = 4441404, upload-time = "2026-03-25T23:34:04.35Z" },
-    { url = "https://files.pythonhosted.org/packages/09/0a/4fe7a8d25fed74419f91835cf5829ade6408fd1963c9eae9c4bce390ecbb/cryptography-46.0.6-cp314-cp314t-musllinux_1_2_aarch64.whl", hash = "sha256:8e7304c4f4e9490e11efe56af6713983460ee0780f16c63f219984dab3af9d2d", size = 4397549, upload-time = "2026-03-25T23:34:06.342Z" },
-    { url = "https://files.pythonhosted.org/packages/5f/a0/7d738944eac6513cd60a8da98b65951f4a3b279b93479a7e8926d9cd730b/cryptography-46.0.6-cp314-cp314t-musllinux_1_2_x86_64.whl", hash = "sha256:b928a3ca837c77a10e81a814a693f2295200adb3352395fad024559b7be7a736", size = 4651874, upload-time = "2026-03-25T23:34:07.916Z" },
-    { url = "https://files.pythonhosted.org/packages/cb/f1/c2326781ca05208845efca38bf714f76939ae446cd492d7613808badedf1/cryptography-46.0.6-cp314-cp314t-win32.whl", hash = "sha256:97c8115b27e19e592a05c45d0dd89c57f81f841cc9880e353e0d3bf25b2139ed", size = 3001511, upload-time = "2026-03-25T23:34:09.892Z" },
-    { url = "https://files.pythonhosted.org/packages/c9/57/fe4a23eb549ac9d903bd4698ffda13383808ef0876cc912bcb2838799ece/cryptography-46.0.6-cp314-cp314t-win_amd64.whl", hash = "sha256:c797e2517cb7880f8297e2c0f43bb910e91381339336f75d2c1c2cbf811b70b4", size = 3471692, upload-time = "2026-03-25T23:34:11.613Z" },
-    { url = "https://files.pythonhosted.org/packages/c4/cc/f330e982852403da79008552de9906804568ae9230da8432f7496ce02b71/cryptography-46.0.6-cp38-abi3-macosx_10_9_universal2.whl", hash = "sha256:12cae594e9473bca1a7aceb90536060643128bb274fcea0fc459ab90f7d1ae7a", size = 7162776, upload-time = "2026-03-25T23:34:13.308Z" },
-    { url = "https://files.pythonhosted.org/packages/49/b3/dc27efd8dcc4bff583b3f01d4a3943cd8b5821777a58b3a6a5f054d61b79/cryptography-46.0.6-cp38-abi3-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:639301950939d844a9e1c4464d7e07f902fe9a7f6b215bb0d4f28584729935d8", size = 4270529, upload-time = "2026-03-25T23:34:15.019Z" },
-    { url = "https://files.pythonhosted.org/packages/e6/05/e8d0e6eb4f0d83365b3cb0e00eb3c484f7348db0266652ccd84632a3d58d/cryptography-46.0.6-cp38-abi3-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:ed3775295fb91f70b4027aeba878d79b3e55c0b3e97eaa4de71f8f23a9f2eb77", size = 4414827, upload-time = "2026-03-25T23:34:16.604Z" },
-    { url = "https://files.pythonhosted.org/packages/2f/97/daba0f5d2dc6d855e2dcb70733c812558a7977a55dd4a6722756628c44d1/cryptography-46.0.6-cp38-abi3-manylinux_2_28_aarch64.whl", hash = "sha256:8927ccfbe967c7df312ade694f987e7e9e22b2425976ddbf28271d7e58845290", size = 4271265, upload-time = "2026-03-25T23:34:18.586Z" },
-    { url = "https://files.pythonhosted.org/packages/89/06/fe1fce39a37ac452e58d04b43b0855261dac320a2ebf8f5260dd55b201a9/cryptography-46.0.6-cp38-abi3-manylinux_2_28_ppc64le.whl", hash = "sha256:b12c6b1e1651e42ab5de8b1e00dc3b6354fdfd778e7fa60541ddacc27cd21410", size = 4916800, upload-time = "2026-03-25T23:34:20.561Z" },
-    { url = "https://files.pythonhosted.org/packages/ff/8a/b14f3101fe9c3592603339eb5d94046c3ce5f7fc76d6512a2d40efd9724e/cryptography-46.0.6-cp38-abi3-manylinux_2_28_x86_64.whl", hash = "sha256:063b67749f338ca9c5a0b7fe438a52c25f9526b851e24e6c9310e7195aad3b4d", size = 4448771, upload-time = "2026-03-25T23:34:22.406Z" },
-    { url = "https://files.pythonhosted.org/packages/01/b3/0796998056a66d1973fd52ee89dc1bb3b6581960a91ad4ac705f182d398f/cryptography-46.0.6-cp38-abi3-manylinux_2_31_armv7l.whl", hash = "sha256:02fad249cb0e090b574e30b276a3da6a149e04ee2f049725b1f69e7b8351ec70", size = 3978333, upload-time = "2026-03-25T23:34:24.281Z" },
-    { url = "https://files.pythonhosted.org/packages/c5/3d/db200af5a4ffd08918cd55c08399dc6c9c50b0bc72c00a3246e099d3a849/cryptography-46.0.6-cp38-abi3-manylinux_2_34_aarch64.whl", hash = "sha256:7e6142674f2a9291463e5e150090b95a8519b2fb6e6aaec8917dd8d094ce750d", size = 4271069, upload-time = "2026-03-25T23:34:25.895Z" },
-    { url = "https://files.pythonhosted.org/packages/d7/18/61acfd5b414309d74ee838be321c636fe71815436f53c9f0334bf19064fa/cryptography-46.0.6-cp38-abi3-manylinux_2_34_ppc64le.whl", hash = "sha256:456b3215172aeefb9284550b162801d62f5f264a081049a3e94307fe20792cfa", size = 4878358, upload-time = "2026-03-25T23:34:27.67Z" },
-    { url = "https://files.pythonhosted.org/packages/8b/65/5bf43286d566f8171917cae23ac6add941654ccf085d739195a4eacf1674/cryptography-46.0.6-cp38-abi3-manylinux_2_34_x86_64.whl", hash = "sha256:341359d6c9e68834e204ceaf25936dffeafea3829ab80e9503860dcc4f4dac58", size = 4448061, upload-time = "2026-03-25T23:34:29.375Z" },
-    { url = "https://files.pythonhosted.org/packages/e0/25/7e49c0fa7205cf3597e525d156a6bce5b5c9de1fd7e8cb01120e459f205a/cryptography-46.0.6-cp38-abi3-musllinux_1_2_aarch64.whl", hash = "sha256:9a9c42a2723999a710445bc0d974e345c32adfd8d2fac6d8a251fa829ad31cfb", size = 4399103, upload-time = "2026-03-25T23:34:32.036Z" },
-    { url = "https://files.pythonhosted.org/packages/44/46/466269e833f1c4718d6cd496ffe20c56c9c8d013486ff66b4f69c302a68d/cryptography-46.0.6-cp38-abi3-musllinux_1_2_x86_64.whl", hash = "sha256:6617f67b1606dfd9fe4dbfa354a9508d4a6d37afe30306fe6c101b7ce3274b72", size = 4659255, upload-time = "2026-03-25T23:34:33.679Z" },
-    { url = "https://files.pythonhosted.org/packages/0a/09/ddc5f630cc32287d2c953fc5d32705e63ec73e37308e5120955316f53827/cryptography-46.0.6-cp38-abi3-win32.whl", hash = "sha256:7f6690b6c55e9c5332c0b59b9c8a3fb232ebf059094c17f9019a51e9827df91c", size = 3010660, upload-time = "2026-03-25T23:34:35.418Z" },
-    { url = "https://files.pythonhosted.org/packages/1b/82/ca4893968aeb2709aacfb57a30dec6fa2ab25b10fa9f064b8882ce33f599/cryptography-46.0.6-cp38-abi3-win_amd64.whl", hash = "sha256:79e865c642cfc5c0b3eb12af83c35c5aeff4fa5c672dc28c43721c2c9fdd2f0f", size = 3471160, upload-time = "2026-03-25T23:34:37.191Z" },
-    { url = "https://files.pythonhosted.org/packages/2e/84/7ccff00ced5bac74b775ce0beb7d1be4e8637536b522b5df9b73ada42da2/cryptography-46.0.6-pp311-pypy311_pp73-macosx_11_0_arm64.whl", hash = "sha256:2ea0f37e9a9cf0df2952893ad145fd9627d326a59daec9b0802480fa3bcd2ead", size = 3475444, upload-time = "2026-03-25T23:34:38.944Z" },
-    { url = "https://files.pythonhosted.org/packages/bc/1f/4c926f50df7749f000f20eede0c896769509895e2648db5da0ed55db711d/cryptography-46.0.6-pp311-pypy311_pp73-manylinux_2_28_aarch64.whl", hash = "sha256:a3e84d5ec9ba01f8fd03802b2147ba77f0c8f2617b2aff254cedd551844209c8", size = 4218227, upload-time = "2026-03-25T23:34:40.871Z" },
-    { url = "https://files.pythonhosted.org/packages/c6/65/707be3ffbd5f786028665c3223e86e11c4cda86023adbc56bd72b1b6bab5/cryptography-46.0.6-pp311-pypy311_pp73-manylinux_2_28_x86_64.whl", hash = "sha256:12f0fa16cc247b13c43d56d7b35287ff1569b5b1f4c5e87e92cc4fcc00cd10c0", size = 4381399, upload-time = "2026-03-25T23:34:42.609Z" },
-    { url = "https://files.pythonhosted.org/packages/f3/6d/73557ed0ef7d73d04d9aba745d2c8e95218213687ee5e76b7d236a5030fc/cryptography-46.0.6-pp311-pypy311_pp73-manylinux_2_34_aarch64.whl", hash = "sha256:50575a76e2951fe7dbd1f56d181f8c5ceeeb075e9ff88e7ad997d2f42af06e7b", size = 4217595, upload-time = "2026-03-25T23:34:44.205Z" },
-    { url = "https://files.pythonhosted.org/packages/9e/c5/e1594c4eec66a567c3ac4400008108a415808be2ce13dcb9a9045c92f1a0/cryptography-46.0.6-pp311-pypy311_pp73-manylinux_2_34_x86_64.whl", hash = "sha256:90e5f0a7b3be5f40c3a0a0eafb32c681d8d2c181fc2a1bdabe9b3f611d9f6b1a", size = 4380912, upload-time = "2026-03-25T23:34:46.328Z" },
-    { url = "https://files.pythonhosted.org/packages/1a/89/843b53614b47f97fe1abc13f9a86efa5ec9e275292c457af1d4a60dc80e0/cryptography-46.0.6-pp311-pypy311_pp73-win_amd64.whl", hash = "sha256:6728c49e3b2c180ef26f8e9f0a883a2c585638db64cf265b49c9ba10652d430e", size = 3409955, upload-time = "2026-03-25T23:34:48.465Z" },
+    { url = "https://files.pythonhosted.org/packages/f7/81/b0bb27f2ba931a65409c6b8a8b358a7f03c0e46eceacddff55f7c84b1f3b/cryptography-46.0.5-cp311-abi3-macosx_10_9_universal2.whl", hash = "sha256:351695ada9ea9618b3500b490ad54c739860883df6c1f555e088eaf25b1bbaad", size = 7176289, upload-time = "2026-02-10T19:17:08.274Z" },
+    { url = "https://files.pythonhosted.org/packages/ff/9e/6b4397a3e3d15123de3b1806ef342522393d50736c13b20ec4c9ea6693a6/cryptography-46.0.5-cp311-abi3-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:c18ff11e86df2e28854939acde2d003f7984f721eba450b56a200ad90eeb0e6b", size = 4275637, upload-time = "2026-02-10T19:17:10.53Z" },
+    { url = "https://files.pythonhosted.org/packages/63/e7/471ab61099a3920b0c77852ea3f0ea611c9702f651600397ac567848b897/cryptography-46.0.5-cp311-abi3-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:4d7e3d356b8cd4ea5aff04f129d5f66ebdc7b6f8eae802b93739ed520c47c79b", size = 4424742, upload-time = "2026-02-10T19:17:12.388Z" },
+    { url = "https://files.pythonhosted.org/packages/37/53/a18500f270342d66bf7e4d9f091114e31e5ee9e7375a5aba2e85a91e0044/cryptography-46.0.5-cp311-abi3-manylinux_2_28_aarch64.whl", hash = "sha256:50bfb6925eff619c9c023b967d5b77a54e04256c4281b0e21336a130cd7fc263", size = 4277528, upload-time = "2026-02-10T19:17:13.853Z" },
+    { url = "https://files.pythonhosted.org/packages/22/29/c2e812ebc38c57b40e7c583895e73c8c5adb4d1e4a0cc4c5a4fdab2b1acc/cryptography-46.0.5-cp311-abi3-manylinux_2_28_ppc64le.whl", hash = "sha256:803812e111e75d1aa73690d2facc295eaefd4439be1023fefc4995eaea2af90d", size = 4947993, upload-time = "2026-02-10T19:17:15.618Z" },
+    { url = "https://files.pythonhosted.org/packages/6b/e7/237155ae19a9023de7e30ec64e5d99a9431a567407ac21170a046d22a5a3/cryptography-46.0.5-cp311-abi3-manylinux_2_28_x86_64.whl", hash = "sha256:3ee190460e2fbe447175cda91b88b84ae8322a104fc27766ad09428754a618ed", size = 4456855, upload-time = "2026-02-10T19:17:17.221Z" },
+    { url = "https://files.pythonhosted.org/packages/2d/87/fc628a7ad85b81206738abbd213b07702bcbdada1dd43f72236ef3cffbb5/cryptography-46.0.5-cp311-abi3-manylinux_2_31_armv7l.whl", hash = "sha256:f145bba11b878005c496e93e257c1e88f154d278d2638e6450d17e0f31e558d2", size = 3984635, upload-time = "2026-02-10T19:17:18.792Z" },
+    { url = "https://files.pythonhosted.org/packages/84/29/65b55622bde135aedf4565dc509d99b560ee4095e56989e815f8fd2aa910/cryptography-46.0.5-cp311-abi3-manylinux_2_34_aarch64.whl", hash = "sha256:e9251e3be159d1020c4030bd2e5f84d6a43fe54b6c19c12f51cde9542a2817b2", size = 4277038, upload-time = "2026-02-10T19:17:20.256Z" },
+    { url = "https://files.pythonhosted.org/packages/bc/36/45e76c68d7311432741faf1fbf7fac8a196a0a735ca21f504c75d37e2558/cryptography-46.0.5-cp311-abi3-manylinux_2_34_ppc64le.whl", hash = "sha256:47fb8a66058b80e509c47118ef8a75d14c455e81ac369050f20ba0d23e77fee0", size = 4912181, upload-time = "2026-02-10T19:17:21.825Z" },
+    { url = "https://files.pythonhosted.org/packages/6d/1a/c1ba8fead184d6e3d5afcf03d569acac5ad063f3ac9fb7258af158f7e378/cryptography-46.0.5-cp311-abi3-manylinux_2_34_x86_64.whl", hash = "sha256:4c3341037c136030cb46e4b1e17b7418ea4cbd9dd207e4a6f3b2b24e0d4ac731", size = 4456482, upload-time = "2026-02-10T19:17:25.133Z" },
+    { url = "https://files.pythonhosted.org/packages/f9/e5/3fb22e37f66827ced3b902cf895e6a6bc1d095b5b26be26bd13c441fdf19/cryptography-46.0.5-cp311-abi3-musllinux_1_2_aarch64.whl", hash = "sha256:890bcb4abd5a2d3f852196437129eb3667d62630333aacc13dfd470fad3aaa82", size = 4405497, upload-time = "2026-02-10T19:17:26.66Z" },
+    { url = "https://files.pythonhosted.org/packages/1a/df/9d58bb32b1121a8a2f27383fabae4d63080c7ca60b9b5c88be742be04ee7/cryptography-46.0.5-cp311-abi3-musllinux_1_2_x86_64.whl", hash = "sha256:80a8d7bfdf38f87ca30a5391c0c9ce4ed2926918e017c29ddf643d0ed2778ea1", size = 4667819, upload-time = "2026-02-10T19:17:28.569Z" },
+    { url = "https://files.pythonhosted.org/packages/ea/ed/325d2a490c5e94038cdb0117da9397ece1f11201f425c4e9c57fe5b9f08b/cryptography-46.0.5-cp311-abi3-win32.whl", hash = "sha256:60ee7e19e95104d4c03871d7d7dfb3d22ef8a9b9c6778c94e1c8fcc8365afd48", size = 3028230, upload-time = "2026-02-10T19:17:30.518Z" },
+    { url = "https://files.pythonhosted.org/packages/e9/5a/ac0f49e48063ab4255d9e3b79f5def51697fce1a95ea1370f03dc9db76f6/cryptography-46.0.5-cp311-abi3-win_amd64.whl", hash = "sha256:38946c54b16c885c72c4f59846be9743d699eee2b69b6988e0a00a01f46a61a4", size = 3480909, upload-time = "2026-02-10T19:17:32.083Z" },
+    { url = "https://files.pythonhosted.org/packages/00/13/3d278bfa7a15a96b9dc22db5a12ad1e48a9eb3d40e1827ef66a5df75d0d0/cryptography-46.0.5-cp314-cp314t-macosx_10_9_universal2.whl", hash = "sha256:94a76daa32eb78d61339aff7952ea819b1734b46f73646a07decb40e5b3448e2", size = 7119287, upload-time = "2026-02-10T19:17:33.801Z" },
+    { url = "https://files.pythonhosted.org/packages/67/c8/581a6702e14f0898a0848105cbefd20c058099e2c2d22ef4e476dfec75d7/cryptography-46.0.5-cp314-cp314t-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:5be7bf2fb40769e05739dd0046e7b26f9d4670badc7b032d6ce4db64dddc0678", size = 4265728, upload-time = "2026-02-10T19:17:35.569Z" },
+    { url = "https://files.pythonhosted.org/packages/dd/4a/ba1a65ce8fc65435e5a849558379896c957870dd64fecea97b1ad5f46a37/cryptography-46.0.5-cp314-cp314t-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:fe346b143ff9685e40192a4960938545c699054ba11d4f9029f94751e3f71d87", size = 4408287, upload-time = "2026-02-10T19:17:36.938Z" },
+    { url = "https://files.pythonhosted.org/packages/f8/67/8ffdbf7b65ed1ac224d1c2df3943553766914a8ca718747ee3871da6107e/cryptography-46.0.5-cp314-cp314t-manylinux_2_28_aarch64.whl", hash = "sha256:c69fd885df7d089548a42d5ec05be26050ebcd2283d89b3d30676eb32ff87dee", size = 4270291, upload-time = "2026-02-10T19:17:38.748Z" },
+    { url = "https://files.pythonhosted.org/packages/f8/e5/f52377ee93bc2f2bba55a41a886fd208c15276ffbd2569f2ddc89d50e2c5/cryptography-46.0.5-cp314-cp314t-manylinux_2_28_ppc64le.whl", hash = "sha256:8293f3dea7fc929ef7240796ba231413afa7b68ce38fd21da2995549f5961981", size = 4927539, upload-time = "2026-02-10T19:17:40.241Z" },
+    { url = "https://files.pythonhosted.org/packages/3b/02/cfe39181b02419bbbbcf3abdd16c1c5c8541f03ca8bda240debc467d5a12/cryptography-46.0.5-cp314-cp314t-manylinux_2_28_x86_64.whl", hash = "sha256:1abfdb89b41c3be0365328a410baa9df3ff8a9110fb75e7b52e66803ddabc9a9", size = 4442199, upload-time = "2026-02-10T19:17:41.789Z" },
+    { url = "https://files.pythonhosted.org/packages/c0/96/2fcaeb4873e536cf71421a388a6c11b5bc846e986b2b069c79363dc1648e/cryptography-46.0.5-cp314-cp314t-manylinux_2_31_armv7l.whl", hash = "sha256:d66e421495fdb797610a08f43b05269e0a5ea7f5e652a89bfd5a7d3c1dee3648", size = 3960131, upload-time = "2026-02-10T19:17:43.379Z" },
+    { url = "https://files.pythonhosted.org/packages/d8/d2/b27631f401ddd644e94c5cf33c9a4069f72011821cf3dc7309546b0642a0/cryptography-46.0.5-cp314-cp314t-manylinux_2_34_aarch64.whl", hash = "sha256:4e817a8920bfbcff8940ecfd60f23d01836408242b30f1a708d93198393a80b4", size = 4270072, upload-time = "2026-02-10T19:17:45.481Z" },
+    { url = "https://files.pythonhosted.org/packages/f4/a7/60d32b0370dae0b4ebe55ffa10e8599a2a59935b5ece1b9f06edb73abdeb/cryptography-46.0.5-cp314-cp314t-manylinux_2_34_ppc64le.whl", hash = "sha256:68f68d13f2e1cb95163fa3b4db4bf9a159a418f5f6e7242564fc75fcae667fd0", size = 4892170, upload-time = "2026-02-10T19:17:46.997Z" },
+    { url = "https://files.pythonhosted.org/packages/d2/b9/cf73ddf8ef1164330eb0b199a589103c363afa0cf794218c24d524a58eab/cryptography-46.0.5-cp314-cp314t-manylinux_2_34_x86_64.whl", hash = "sha256:a3d1fae9863299076f05cb8a778c467578262fae09f9dc0ee9b12eb4268ce663", size = 4441741, upload-time = "2026-02-10T19:17:48.661Z" },
+    { url = "https://files.pythonhosted.org/packages/5f/eb/eee00b28c84c726fe8fa0158c65afe312d9c3b78d9d01daf700f1f6e37ff/cryptography-46.0.5-cp314-cp314t-musllinux_1_2_aarch64.whl", hash = "sha256:c4143987a42a2397f2fc3b4d7e3a7d313fbe684f67ff443999e803dd75a76826", size = 4396728, upload-time = "2026-02-10T19:17:50.058Z" },
+    { url = "https://files.pythonhosted.org/packages/65/f4/6bc1a9ed5aef7145045114b75b77c2a8261b4d38717bd8dea111a63c3442/cryptography-46.0.5-cp314-cp314t-musllinux_1_2_x86_64.whl", hash = "sha256:7d731d4b107030987fd61a7f8ab512b25b53cef8f233a97379ede116f30eb67d", size = 4652001, upload-time = "2026-02-10T19:17:51.54Z" },
+    { url = "https://files.pythonhosted.org/packages/86/ef/5d00ef966ddd71ac2e6951d278884a84a40ffbd88948ef0e294b214ae9e4/cryptography-46.0.5-cp314-cp314t-win32.whl", hash = "sha256:c3bcce8521d785d510b2aad26ae2c966092b7daa8f45dd8f44734a104dc0bc1a", size = 3003637, upload-time = "2026-02-10T19:17:52.997Z" },
+    { url = "https://files.pythonhosted.org/packages/b7/57/f3f4160123da6d098db78350fdfd9705057aad21de7388eacb2401dceab9/cryptography-46.0.5-cp314-cp314t-win_amd64.whl", hash = "sha256:4d8ae8659ab18c65ced284993c2265910f6c9e650189d4e3f68445ef82a810e4", size = 3469487, upload-time = "2026-02-10T19:17:54.549Z" },
+    { url = "https://files.pythonhosted.org/packages/e2/fa/a66aa722105ad6a458bebd64086ca2b72cdd361fed31763d20390f6f1389/cryptography-46.0.5-cp38-abi3-macosx_10_9_universal2.whl", hash = "sha256:4108d4c09fbbf2789d0c926eb4152ae1760d5a2d97612b92d508d96c861e4d31", size = 7170514, upload-time = "2026-02-10T19:17:56.267Z" },
+    { url = "https://files.pythonhosted.org/packages/0f/04/c85bdeab78c8bc77b701bf0d9bdcf514c044e18a46dcff330df5448631b0/cryptography-46.0.5-cp38-abi3-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:7d1f30a86d2757199cb2d56e48cce14deddf1f9c95f1ef1b64ee91ea43fe2e18", size = 4275349, upload-time = "2026-02-10T19:17:58.419Z" },
+    { url = "https://files.pythonhosted.org/packages/5c/32/9b87132a2f91ee7f5223b091dc963055503e9b442c98fc0b8a5ca765fab0/cryptography-46.0.5-cp38-abi3-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:039917b0dc418bb9f6edce8a906572d69e74bd330b0b3fea4f79dab7f8ddd235", size = 4420667, upload-time = "2026-02-10T19:18:00.619Z" },
+    { url = "https://files.pythonhosted.org/packages/a1/a6/a7cb7010bec4b7c5692ca6f024150371b295ee1c108bdc1c400e4c44562b/cryptography-46.0.5-cp38-abi3-manylinux_2_28_aarch64.whl", hash = "sha256:ba2a27ff02f48193fc4daeadf8ad2590516fa3d0adeeb34336b96f7fa64c1e3a", size = 4276980, upload-time = "2026-02-10T19:18:02.379Z" },
+    { url = "https://files.pythonhosted.org/packages/8e/7c/c4f45e0eeff9b91e3f12dbd0e165fcf2a38847288fcfd889deea99fb7b6d/cryptography-46.0.5-cp38-abi3-manylinux_2_28_ppc64le.whl", hash = "sha256:61aa400dce22cb001a98014f647dc21cda08f7915ceb95df0c9eaf84b4b6af76", size = 4939143, upload-time = "2026-02-10T19:18:03.964Z" },
+    { url = "https://files.pythonhosted.org/packages/37/19/e1b8f964a834eddb44fa1b9a9976f4e414cbb7aa62809b6760c8803d22d1/cryptography-46.0.5-cp38-abi3-manylinux_2_28_x86_64.whl", hash = "sha256:3ce58ba46e1bc2aac4f7d9290223cead56743fa6ab94a5d53292ffaac6a91614", size = 4453674, upload-time = "2026-02-10T19:18:05.588Z" },
+    { url = "https://files.pythonhosted.org/packages/db/ed/db15d3956f65264ca204625597c410d420e26530c4e2943e05a0d2f24d51/cryptography-46.0.5-cp38-abi3-manylinux_2_31_armv7l.whl", hash = "sha256:420d0e909050490d04359e7fdb5ed7e667ca5c3c402b809ae2563d7e66a92229", size = 3978801, upload-time = "2026-02-10T19:18:07.167Z" },
+    { url = "https://files.pythonhosted.org/packages/41/e2/df40a31d82df0a70a0daf69791f91dbb70e47644c58581d654879b382d11/cryptography-46.0.5-cp38-abi3-manylinux_2_34_aarch64.whl", hash = "sha256:582f5fcd2afa31622f317f80426a027f30dc792e9c80ffee87b993200ea115f1", size = 4276755, upload-time = "2026-02-10T19:18:09.813Z" },
+    { url = "https://files.pythonhosted.org/packages/33/45/726809d1176959f4a896b86907b98ff4391a8aa29c0aaaf9450a8a10630e/cryptography-46.0.5-cp38-abi3-manylinux_2_34_ppc64le.whl", hash = "sha256:bfd56bb4b37ed4f330b82402f6f435845a5f5648edf1ad497da51a8452d5d62d", size = 4901539, upload-time = "2026-02-10T19:18:11.263Z" },
+    { url = "https://files.pythonhosted.org/packages/99/0f/a3076874e9c88ecb2ecc31382f6e7c21b428ede6f55aafa1aa272613e3cd/cryptography-46.0.5-cp38-abi3-manylinux_2_34_x86_64.whl", hash = "sha256:a3d507bb6a513ca96ba84443226af944b0f7f47dcc9a399d110cd6146481d24c", size = 4452794, upload-time = "2026-02-10T19:18:12.914Z" },
+    { url = "https://files.pythonhosted.org/packages/02/ef/ffeb542d3683d24194a38f66ca17c0a4b8bf10631feef44a7ef64e631b1a/cryptography-46.0.5-cp38-abi3-musllinux_1_2_aarch64.whl", hash = "sha256:9f16fbdf4da055efb21c22d81b89f155f02ba420558db21288b3d0035bafd5f4", size = 4404160, upload-time = "2026-02-10T19:18:14.375Z" },
+    { url = "https://files.pythonhosted.org/packages/96/93/682d2b43c1d5f1406ed048f377c0fc9fc8f7b0447a478d5c65ab3d3a66eb/cryptography-46.0.5-cp38-abi3-musllinux_1_2_x86_64.whl", hash = "sha256:ced80795227d70549a411a4ab66e8ce307899fad2220ce5ab2f296e687eacde9", size = 4667123, upload-time = "2026-02-10T19:18:15.886Z" },
+    { url = "https://files.pythonhosted.org/packages/45/2d/9c5f2926cb5300a8eefc3f4f0b3f3df39db7f7ce40c8365444c49363cbda/cryptography-46.0.5-cp38-abi3-win32.whl", hash = "sha256:02f547fce831f5096c9a567fd41bc12ca8f11df260959ecc7c3202555cc47a72", size = 3010220, upload-time = "2026-02-10T19:18:17.361Z" },
+    { url = "https://files.pythonhosted.org/packages/48/ef/0c2f4a8e31018a986949d34a01115dd057bf536905dca38897bacd21fac3/cryptography-46.0.5-cp38-abi3-win_amd64.whl", hash = "sha256:556e106ee01aa13484ce9b0239bca667be5004efb0aabbed28d353df86445595", size = 3467050, upload-time = "2026-02-10T19:18:18.899Z" },
+    { url = "https://files.pythonhosted.org/packages/eb/dd/2d9fdb07cebdf3d51179730afb7d5e576153c6744c3ff8fded23030c204e/cryptography-46.0.5-pp311-pypy311_pp73-macosx_11_0_arm64.whl", hash = "sha256:3b4995dc971c9fb83c25aa44cf45f02ba86f71ee600d81091c2f0cbae116b06c", size = 3476964, upload-time = "2026-02-10T19:18:20.687Z" },
+    { url = "https://files.pythonhosted.org/packages/e9/6f/6cc6cc9955caa6eaf83660b0da2b077c7fe8ff9950a3c5e45d605038d439/cryptography-46.0.5-pp311-pypy311_pp73-manylinux_2_28_aarch64.whl", hash = "sha256:bc84e875994c3b445871ea7181d424588171efec3e185dced958dad9e001950a", size = 4218321, upload-time = "2026-02-10T19:18:22.349Z" },
+    { url = "https://files.pythonhosted.org/packages/3e/5d/c4da701939eeee699566a6c1367427ab91a8b7088cc2328c09dbee940415/cryptography-46.0.5-pp311-pypy311_pp73-manylinux_2_28_x86_64.whl", hash = "sha256:2ae6971afd6246710480e3f15824ed3029a60fc16991db250034efd0b9fb4356", size = 4381786, upload-time = "2026-02-10T19:18:24.529Z" },
+    { url = "https://files.pythonhosted.org/packages/ac/97/a538654732974a94ff96c1db621fa464f455c02d4bb7d2652f4edc21d600/cryptography-46.0.5-pp311-pypy311_pp73-manylinux_2_34_aarch64.whl", hash = "sha256:d861ee9e76ace6cf36a6a89b959ec08e7bc2493ee39d07ffe5acb23ef46d27da", size = 4217990, upload-time = "2026-02-10T19:18:25.957Z" },
+    { url = "https://files.pythonhosted.org/packages/ae/11/7e500d2dd3ba891197b9efd2da5454b74336d64a7cc419aa7327ab74e5f6/cryptography-46.0.5-pp311-pypy311_pp73-manylinux_2_34_x86_64.whl", hash = "sha256:2b7a67c9cd56372f3249b39699f2ad479f6991e62ea15800973b956f4b73e257", size = 4381252, upload-time = "2026-02-10T19:18:27.496Z" },
+    { url = "https://files.pythonhosted.org/packages/bc/58/6b3d24e6b9bc474a2dcdee65dfd1f008867015408a271562e4b690561a4d/cryptography-46.0.5-pp311-pypy311_pp73-win_amd64.whl", hash = "sha256:8456928655f856c6e1533ff59d5be76578a7157224dbd9ce6872f25055ab9ab7", size = 3407605, upload-time = "2026-02-10T19:18:29.233Z" },
 ]
 
 [[package]]
@@ -3048,7 +3048,7 @@ wheels = [
 
 [[package]]
 name = "langchain-core"
-version = "1.2.22"
+version = "1.2.11"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "jsonpatch" },
@@ -3060,9 +3060,9 @@ dependencies = [
     { name = "typing-extensions" },
     { name = "uuid-utils" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/b1/a3/c4cd6827a1df46c821e7214b7f7b7a28b189e6c9b84ef15c6d629c5e3179/langchain_core-1.2.22.tar.gz", hash = "sha256:8d8f726d03d3652d403da915126626bb6250747e8ba406537d849e68b9f5d058", size = 842487, upload-time = "2026-03-24T18:48:44.9Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/12/17/1943cedfc118e04b8128e4c3e1dbf0fa0ea58eefddbb6198cfd699d19f01/langchain_core-1.2.11.tar.gz", hash = "sha256:f164bb36602dd74a3a50c1334fca75309ad5ed95767acdfdbb9fa95ce28a1e01", size = 831211, upload-time = "2026-02-10T20:35:28.35Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/c7/a6/2ffacf0f1a3788f250e75d0b52a24896c413be11be3a6d42bcdf46fbea48/langchain_core-1.2.22-py3-none-any.whl", hash = "sha256:7e30d586b75918e828833b9ec1efc25465723566845dd652c277baf751e9c04b", size = 506829, upload-time = "2026-03-24T18:48:43.286Z" },
+    { url = "https://files.pythonhosted.org/packages/10/30/1f80e3fc674353cad975ed5294353d42512535d2094ef032c06454c2c873/langchain_core-1.2.11-py3-none-any.whl", hash = "sha256:ae11ceb8dda60d0b9d09e763116e592f1683327c17be5b715f350fd29aee65d3", size = 500062, upload-time = "2026-02-10T20:35:26.698Z" },
 ]
 
 [[package]]
@@ -4439,7 +4439,7 @@ requires-dist = [
     { name = "jsonref", marker = "extra == 'backend'", specifier = "==1.1.0" },
     { name = "kubernetes", specifier = ">=31.0.0" },
     { name = "kubernetes", marker = "extra == 'backend'", specifier = "==31.0.0" },
-    { name = "langchain-core", marker = "extra == 'backend'", specifier = "==1.2.22" },
+    { name = "langchain-core", marker = "extra == 'backend'", specifier = "==1.2.11" },
     { name = "langfuse", marker = "extra == 'backend'", specifier = "==3.10.0" },
     { name = "lazy-imports", marker = "extra == 'backend'", specifier = "==1.0.1" },
     { name = "litellm", specifier = "==1.81.6" },
@@ -4458,7 +4458,7 @@ requires-dist = [
     { name = "numpy", marker = "extra == 'model-server'", specifier = "==2.4.1" },
     { name = "oauthlib", marker = "extra == 'backend'", specifier = "==3.2.2" },
     { name = "office365-rest-python-client", marker = "extra == 'backend'", specifier = "==2.6.2" },
-    { name = "onyx-devtools", marker = "extra == 'dev'", specifier = "==0.7.2" },
+    { name = "onyx-devtools", marker = "extra == 'dev'", specifier = "==0.7.1" },
     { name = "openai", specifier = "==2.14.0" },
     { name = "openapi-generator-cli", marker = "extra == 'dev'", specifier = "==7.17.0" },
     { name = "openinference-instrumentation", marker = "extra == 'backend'", specifier = "==0.1.42" },
@@ -4563,19 +4563,19 @@ requires-dist = [{ name = "onyx", extras = ["backend", "dev", "ee"], editable =
 
 [[package]]
 name = "onyx-devtools"
-version = "0.7.2"
+version = "0.7.1"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "fastapi" },
     { name = "openapi-generator-cli" },
 ]
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/22/b0/765ed49157470e8ccc8ab89e6a896ade50cde3aa2a494662ad4db92a48c4/onyx_devtools-0.7.2-py3-none-macosx_10_12_x86_64.whl", hash = "sha256:553a2b5e61b29b7913c991c8d5aed78f930f0f81a0f42229c6a8de2b1e8ff57e", size = 4203859, upload-time = "2026-03-27T15:09:49.63Z" },
-    { url = "https://files.pythonhosted.org/packages/f7/9d/bba0a44a16d2fc27e5441aaf10727e10514e7a49bce70eca02bced566eb9/onyx_devtools-0.7.2-py3-none-macosx_11_0_arm64.whl", hash = "sha256:5cf0782dca8b3d861de9e18e65e990cfce5161cd559df44d8fabd3fefd54fdcd", size = 3879750, upload-time = "2026-03-27T15:09:42.413Z" },
-    { url = "https://files.pythonhosted.org/packages/4d/d8/c5725e8af14c74fe0aeed29e4746400bb3c0a078fd1240df729dc6432b84/onyx_devtools-0.7.2-py3-none-manylinux_2_17_aarch64.whl", hash = "sha256:9a0d67373e16b4fbb38a5290c0d9dfd4cfa837e5da0c165b32841b9d37f7455b", size = 3743529, upload-time = "2026-03-27T15:09:44.546Z" },
-    { url = "https://files.pythonhosted.org/packages/1a/82/b7c398a21dbc3e14fd7a29e49caa86b1bc0f8d7c75c051514785441ab779/onyx_devtools-0.7.2-py3-none-manylinux_2_17_x86_64.whl", hash = "sha256:794af14b2de575d0ae41b94551399eca8f8ba9b950c5db7acb7612767fd228f9", size = 4166562, upload-time = "2026-03-27T15:09:49.471Z" },
-    { url = "https://files.pythonhosted.org/packages/26/76/be129e2baafc91fe792d919b1f4d73fc943ba9c2b728a60f1fb98e0c115a/onyx_devtools-0.7.2-py3-none-win_amd64.whl", hash = "sha256:83b3eb84df58d865e4f714222a5fab3ea464836e2c8690569454a940bbb651ff", size = 4282270, upload-time = "2026-03-27T15:09:44.676Z" },
-    { url = "https://files.pythonhosted.org/packages/3b/72/29b8c8dbcf069c56475f00511f04c4aaa5ba3faba1dfc8276107d4b3ef7f/onyx_devtools-0.7.2-py3-none-win_arm64.whl", hash = "sha256:62f0836624ee6a5b31e64fd93162e7fce142ac8a4f959607e411824bc2b88174", size = 3823053, upload-time = "2026-03-27T15:09:43.546Z" },
+    { url = "https://files.pythonhosted.org/packages/65/9d/74bcd02583706bdf90c8ac9084eb60bd71d0671392152410ab21b7b68ea1/onyx_devtools-0.7.1-py3-none-macosx_10_12_x86_64.whl", hash = "sha256:178385dce0b413fd2a1f761055a99f556ec536ef5c32963fc273e751813621eb", size = 4007974, upload-time = "2026-03-17T21:10:39.267Z" },
+    { url = "https://files.pythonhosted.org/packages/f0/f8/d8ddb32120428c083c60eb07244479da6e07eaebd31847658a049ab33815/onyx_devtools-0.7.1-py3-none-macosx_11_0_arm64.whl", hash = "sha256:7960ae6e440ebf1584e02d9e1d0c9ef543b1d54c2584cdcace15695aec3121b2", size = 3696924, upload-time = "2026-03-17T21:10:50.716Z" },
+    { url = "https://files.pythonhosted.org/packages/87/21/1e427280066db42ff9dd5f34c70b9dca5d9781f96d0d9a88aaa454fdb432/onyx_devtools-0.7.1-py3-none-manylinux_2_17_aarch64.whl", hash = "sha256:6785dda88ca0a3d8464a9bfab76a253ed90da89d53a9c4a67227980f37df1ccf", size = 3568300, upload-time = "2026-03-17T21:10:41.997Z" },
+    { url = "https://files.pythonhosted.org/packages/0e/0e/afbbe1164b3d016ddb5352353cb2541eef5a8b2c04e8f02d5d1319cb8b8c/onyx_devtools-0.7.1-py3-none-manylinux_2_17_x86_64.whl", hash = "sha256:9e77f2b725c0c00061a3dda5eba199404b51638cec0bf54fc7611fee1f26db34", size = 3974668, upload-time = "2026-03-17T21:10:43.879Z" },
+    { url = "https://files.pythonhosted.org/packages/8a/a5/22840643289ef4ca83931b7a79fba8f1db7e626b4b870d4b4f8206c4ff5f/onyx_devtools-0.7.1-py3-none-win_amd64.whl", hash = "sha256:de37daa0e4db9b5dccf94408a3422be4f821e380ab70081bd1032cec1e3c91e6", size = 4078640, upload-time = "2026-03-17T21:10:40.275Z" },
+    { url = "https://files.pythonhosted.org/packages/1e/c1/a0295506a521d9942b0f55523781a113e4555420d800a386d5a2eb46a7ad/onyx_devtools-0.7.1-py3-none-win_arm64.whl", hash = "sha256:ab88c53ebda6dff27350316b4ac9bd5f258cd586c2109971a9d976411e1e22ea", size = 3636787, upload-time = "2026-03-17T21:10:37.492Z" },
 ]
 
 [[package]]

web/lib/opal/src/components/buttons/button/components.tsx

@@ -104,7 +104,7 @@ function Button({
           isLarge ? "default" : size === "2xs" ? "mini" : "compact"
         }
       >
-        <div className="flex flex-row items-center gap-1">
+        <div className="flex flex-row items-center gap-1 interactive-foreground">
           {iconWrapper(Icon, size, !!children)}
 
           {labelEl}

web/lib/opal/src/components/buttons/filter-button/components.tsx

@@ -67,7 +67,7 @@ function FilterButton({
         state={active ? "selected" : "empty"}
       >
         <Interactive.Container type="button">
-          <div className="flex flex-row items-center gap-1">
+          <div className="interactive-foreground flex flex-row items-center gap-1">
             {iconWrapper(Icon, "lg", true)}
             <Text font="main-ui-action" color="inherit" nowrap>
               {children}

web/lib/opal/src/components/buttons/line-item-button/components.tsx

@@ -16,7 +16,7 @@ import * as TooltipPrimitive from "@radix-ui/react-tooltip";
 
 type ContentPassthroughProps = DistributiveOmit<
   ContentActionProps,
-  "paddingVariant" | "widthVariant" | "ref"
+  "paddingVariant" | "widthVariant" | "ref" | "withInteractive"
 >;
 
 type LineItemButtonOwnProps = Pick<
@@ -92,6 +92,7 @@ function LineItemButton({
       >
         <ContentAction
           {...(contentActionProps as ContentActionProps)}
+          withInteractive
           paddingVariant="fit"
         />
       </Interactive.Container>

web/lib/opal/src/components/buttons/open-button/components.tsx

@@ -132,7 +132,7 @@ function OpenButton({
       >
         <div
           className={cn(
-            "flex flex-row items-center",
+            "interactive-foreground flex flex-row items-center",
             justifyContent === "between" ? "w-full justify-between" : "gap-1",
             foldable &&
               justifyContent !== "between" &&

web/lib/opal/src/components/buttons/select-button/components.tsx

@@ -1,5 +1,3 @@
-"use client";
-
 import "@opal/components/buttons/select-button/styles.css";
 import {
   Interactive,
@@ -86,13 +84,11 @@ function SelectButton({
   const isLarge = size === "lg";
 
   const labelEl = children ? (
-    <Text
-      font={isLarge ? "main-ui-body" : "secondary-body"}
-      color="inherit"
-      nowrap
-    >
-      {children}
-    </Text>
+    <span className="opal-select-button-label">
+      <Text font={isLarge ? "main-ui-body" : "secondary-body"} color="inherit">
+        {children}
+      </Text>
+    </span>
   ) : null;
 
   const button = (
@@ -107,7 +103,7 @@ function SelectButton({
       >
         <div
           className={cn(
-            "opal-select-button",
+            "opal-select-button interactive-foreground",
             foldable && "interactive-foldable-host"
           )}
         >

web/lib/opal/src/components/buttons/select-button/styles.css

@@ -3,3 +3,7 @@
 .opal-select-button {
   @apply flex flex-row items-center gap-1;
 }
+
+.opal-select-button-label {
+  @apply whitespace-nowrap;
+}

web/lib/opal/src/components/pagination/components.tsx

@@ -390,12 +390,14 @@ function PaginationCount({
         <span className={textClasses(size, "muted")}>of</span>
         {totalItems}
         {units && (
-          <Text
-            color="inherit"
-            font={size === "sm" ? "secondary-body" : "main-ui-muted"}
-          >
-            {units}
-          </Text>
+          <span className="ml-1">
+            <Text
+              color="inherit"
+              font={size === "sm" ? "secondary-body" : "main-ui-muted"}
+            >
+              {units}
+            </Text>
+          </span>
         )}
       </span>
 

web/lib/opal/src/components/tag/components.tsx

@@ -1,4 +1,5 @@
 import "@opal/components/tag/styles.css";
+
 import type { IconFunctionComponent, RichStr } from "@opal/types";
 import { Text } from "@opal/components";
 import { cn } from "@opal/utils";
@@ -45,22 +46,20 @@ function Tag({ icon: Icon, title, color = "gray", size = "sm" }: TagProps) {
   const config = COLOR_CONFIG[color];
 
   return (
-    <div
-      className={cn("opal-auxiliary-tag", config.bg, config.text)}
-      data-size={size}
-    >
+    <div className={cn("opal-auxiliary-tag", config.bg)} data-size={size}>
       {Icon && (
         <div className="opal-auxiliary-tag-icon-container">
           <Icon className={cn("opal-auxiliary-tag-icon", config.text)} />
         </div>
       )}
-      <Text
-        font={size === "md" ? "secondary-body" : "figure-small-value"}
-        color="inherit"
-        nowrap
-      >
-        {title}
-      </Text>
+      <span className={cn("opal-auxiliary-tag-title px-[2px]", config.text)}>
+        <Text
+          font={size === "md" ? "secondary-body" : "figure-small-value"}
+          color="inherit"
+        >
+          {title}
+        </Text>
+      </span>
     </div>
   );
 }

web/lib/opal/src/components/text/InlineMarkdown.tsx

@@ -13,9 +13,7 @@ const SAFE_PROTOCOL = /^https?:|^mailto:|^tel:/i;
 const ALLOWED_ELEMENTS = ["p", "br", "a", "strong", "em", "code", "del"];
 
 const INLINE_COMPONENTS = {
-  p: ({ children }: { children?: ReactNode }) => (
-    <span className="block">{children}</span>
-  ),
+  p: ({ children }: { children?: ReactNode }) => <>{children}</>,
   a: ({ children, href }: { children?: ReactNode; href?: string }) => {
     if (!href || !SAFE_PROTOCOL.test(href)) {
       return <>{children}</>;

web/lib/opal/src/components/text/components.tsx

@@ -125,7 +125,6 @@ function Text({
   ...rest
 }: TextProps) {
   const resolvedClassName = cn(
-    "px-[2px]",
     FONT_CONFIG[font],
     COLOR_CONFIG[color],
     nowrap && "whitespace-nowrap",

web/lib/opal/src/core/index.ts

@@ -18,11 +18,9 @@ export {
 import { InteractiveStateless } from "@opal/core/interactive/stateless/components";
 import { InteractiveStateful } from "@opal/core/interactive/stateful/components";
 import { InteractiveContainer } from "@opal/core/interactive/container/components";
-import { InteractiveSimple } from "@opal/core/interactive/simple/components";
 import { Foldable } from "@opal/core/interactive/foldable/components";
 
 const Interactive = {
-  Simple: InteractiveSimple,
   Stateless: InteractiveStateless,
   Stateful: InteractiveStateful,
   Container: InteractiveContainer,
@@ -52,5 +50,3 @@ export type {
 } from "@opal/core/interactive/container/components";
 
 export type { FoldableProps } from "@opal/core/interactive/foldable/components";
-
-export type { InteractiveSimpleProps } from "@opal/core/interactive/simple/components";

web/lib/opal/src/core/interactive/Interactive.stories.tsx

@@ -9,6 +9,7 @@ const VARIANT_PROMINENCE_MAP: Record<string, string[]> = {
   default: ["primary", "secondary", "tertiary", "internal"],
   action: ["primary", "secondary", "tertiary", "internal"],
   danger: ["primary", "secondary", "tertiary", "internal"],
+  none: [],
 };
 
 const SIZE_VARIANTS = ["lg", "md", "sm", "xs", "2xs", "fit"] as const;
@@ -42,7 +43,7 @@ export const Default: StoryObj = {
         onClick={() => {}}
       >
         <Interactive.Container border>
-          <span>Secondary</span>
+          <span className="interactive-foreground">Secondary</span>
         </Interactive.Container>
       </Interactive.Stateless>
 
@@ -52,7 +53,7 @@ export const Default: StoryObj = {
         onClick={() => {}}
       >
         <Interactive.Container border>
-          <span>Primary</span>
+          <span className="interactive-foreground">Primary</span>
         </Interactive.Container>
       </Interactive.Stateless>
 
@@ -62,7 +63,7 @@ export const Default: StoryObj = {
         onClick={() => {}}
       >
         <Interactive.Container border>
-          <span>Tertiary</span>
+          <span className="interactive-foreground">Tertiary</span>
         </Interactive.Container>
       </Interactive.Stateless>
     </div>
@@ -114,7 +115,9 @@ export const VariantMatrix: StoryObj = {
                     onClick={() => {}}
                   >
                     <Interactive.Container border>
-                      <span>{prominence}</span>
+                      <span className="interactive-foreground">
+                        {prominence}
+                      </span>
                     </Interactive.Container>
                   </Interactive.Stateless>
                   <span
@@ -147,7 +150,7 @@ export const Sizes: StoryObj = {
           onClick={() => {}}
         >
           <Interactive.Container border heightVariant={size}>
-            <span>{size}</span>
+            <span className="interactive-foreground">{size}</span>
           </Interactive.Container>
         </Interactive.Stateless>
       ))}
@@ -165,7 +168,7 @@ export const WidthFull: StoryObj = {
         onClick={() => {}}
       >
         <Interactive.Container border widthVariant="full">
-          <span>Full width container</span>
+          <span className="interactive-foreground">Full width container</span>
         </Interactive.Container>
       </Interactive.Stateless>
     </div>
@@ -184,7 +187,7 @@ export const Rounding: StoryObj = {
           onClick={() => {}}
         >
           <Interactive.Container border roundingVariant={rounding}>
-            <span>{rounding}</span>
+            <span className="interactive-foreground">{rounding}</span>
           </Interactive.Container>
         </Interactive.Stateless>
       ))}
@@ -204,7 +207,7 @@ export const DisabledStory: StoryObj = {
           onClick={() => {}}
         >
           <Interactive.Container border>
-            <span>Disabled</span>
+            <span className="interactive-foreground">Disabled</span>
           </Interactive.Container>
         </Interactive.Stateless>
       </Disabled>
@@ -215,7 +218,7 @@ export const DisabledStory: StoryObj = {
         onClick={() => {}}
       >
         <Interactive.Container border>
-          <span>Enabled</span>
+          <span className="interactive-foreground">Enabled</span>
         </Interactive.Container>
       </Interactive.Stateless>
     </div>
@@ -233,7 +236,7 @@ export const Interaction: StoryObj = {
         onClick={() => {}}
       >
         <Interactive.Container border>
-          <span>Forced hover</span>
+          <span className="interactive-foreground">Forced hover</span>
         </Interactive.Container>
       </Interactive.Stateless>
 
@@ -244,7 +247,7 @@ export const Interaction: StoryObj = {
         onClick={() => {}}
       >
         <Interactive.Container border>
-          <span>Forced active</span>
+          <span className="interactive-foreground">Forced active</span>
         </Interactive.Container>
       </Interactive.Stateless>
 
@@ -254,7 +257,7 @@ export const Interaction: StoryObj = {
         onClick={() => {}}
       >
         <Interactive.Container border>
-          <span>Normal (rest)</span>
+          <span className="interactive-foreground">Normal (rest)</span>
         </Interactive.Container>
       </Interactive.Stateless>
     </div>
@@ -271,7 +274,7 @@ export const WithBorder: StoryObj = {
         onClick={() => {}}
       >
         <Interactive.Container border>
-          <span>With border</span>
+          <span className="interactive-foreground">With border</span>
         </Interactive.Container>
       </Interactive.Stateless>
 
@@ -281,7 +284,7 @@ export const WithBorder: StoryObj = {
         onClick={() => {}}
       >
         <Interactive.Container>
-          <span>Without border</span>
+          <span className="interactive-foreground">Without border</span>
         </Interactive.Container>
       </Interactive.Stateless>
     </div>
@@ -293,7 +296,7 @@ export const AsLink: StoryObj = {
   render: () => (
     <Interactive.Stateless variant="action" href="/settings">
       <Interactive.Container border>
-        <span>Go to Settings</span>
+        <span className="interactive-foreground">Go to Settings</span>
       </Interactive.Container>
     </Interactive.Stateless>
   ),
@@ -309,7 +312,7 @@ export const SelectVariant: StoryObj = {
         onClick={() => {}}
       >
         <Interactive.Container border>
-          <span>Selected (light)</span>
+          <span className="interactive-foreground">Selected (light)</span>
         </Interactive.Container>
       </Interactive.Stateful>
 
@@ -319,7 +322,7 @@ export const SelectVariant: StoryObj = {
         onClick={() => {}}
       >
         <Interactive.Container border>
-          <span>Unselected (light)</span>
+          <span className="interactive-foreground">Unselected (light)</span>
         </Interactive.Container>
       </Interactive.Stateful>
 
@@ -329,7 +332,7 @@ export const SelectVariant: StoryObj = {
         onClick={() => {}}
       >
         <Interactive.Container border>
-          <span>Selected (heavy)</span>
+          <span className="interactive-foreground">Selected (heavy)</span>
         </Interactive.Container>
       </Interactive.Stateful>
 
@@ -339,7 +342,7 @@ export const SelectVariant: StoryObj = {
         onClick={() => {}}
       >
         <Interactive.Container border>
-          <span>Unselected (heavy)</span>
+          <span className="interactive-foreground">Unselected (heavy)</span>
         </Interactive.Container>
       </Interactive.Stateful>
     </div>