* feat(decisions): add hosted_vllm provider
Adds HostedVLLMDecisionsConfig so hosted_vllm/<served model> works on
POST /v1/systemone and POST /v1/decisions against a vLLM server that
serves the System One body (vllm-project/vllm#59299, after v0.31.0).
vLLM only supports choice questions and rejects others, so the base
decisions config gains a per-provider health_check_questions attribute
and the evaluation-mode health probe sends a choice question for
hosted_vllm while every other provider keeps the noul probe.
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* refactor(decisions): move the hosted_vllm placeholder key to constants
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
---------
Co-authored-by: kerry <kerry@berri.ai>
Co-authored-by: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
(cherry picked from commit 61e5f2dd3c)
#45214 routed both decision formats through the shared decisions IR in
litellm/llms/base_llm/decisions/transformation.py and litellm/types/decisions.py.
That left litellm/llms/base_llm/decisions/systemone.py (to_system_one_request,
question_keys, to_decisions_response, the SystemOne* models and their adapter)
and litellm/types/openai_decisions.py with no importer outside their own unit
tests, so both modules and both test files go.
Co-authored-by: mateo-berri <277851410+mateo-berri@users.noreply.github.com>
(cherry picked from commit 0e48048bd5)
* feat(ui): add TypeSafe and Strands Decider to the Add Model provider list
* test(public_endpoints): type the provider fields helper against the route's model
---------
Co-authored-by: mateo-berri <277851410+mateo-berri@users.noreply.github.com>
(cherry picked from commit 3a3b8cad80)
* refactor(decisions): dispatch /v1/decisions through provider configs and the shared HTTP handler
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* fix(decisions): map OpenRouter connection failures to APIConnectionError and drop explanatory docstrings
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* fix(decisions): keep APIConnectionError for unreachable deployments on every provider
`_map_upstream_exception` now maps a status `_handle_error` synthesized for a
non-HTTP failure to `APIConnectionError` before `exception_type` runs, so a
Perplexity (openai-compatible) deployment that cannot be reached no longer
answers `InternalServerError` where main answers `APIConnectionError`.
Audit cells: the unreachable-deployment integration cell is parametrized over
the five providers, OpenRouter connection failures are pinned on chat, stream,
embeddings, decisions and the SDK, a deployment whose provider has no decisions
config gets the 400 naming every supported provider, and a unit cell pins the
mapper alone. The first docstring sentence of the five translation bases is
dropped.
---------
Co-authored-by: kerry <kerry@berri.ai>
Co-authored-by: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
Co-authored-by: mateo-berri <277851410+mateo-berri@users.noreply.github.com>
(cherry picked from commit 0e09e99146)
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* feat(decisions): add the OpenAI Decisions spec types and the System One translation
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* refactor(decisions): share one DecisionsModel config and require model and usage on responses
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* refactor(decisions): rename the shared pydantic parent to DecisionsObjectBase
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* fix(decisions): match the SDK on strict choice values and drop the invented min_length bounds
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* refactor(decisions): compose DecisionsRequest from model and body and read the translator top down
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* refactor(decisions): use the plural Decisions prefix only for the request and response envelopes
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* fix(decisions): import assert_never from typing_extensions for Python 3.10
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
---------
Co-authored-by: kerry <kerry@berri.ai>
Co-authored-by: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
(cherry picked from commit 79f62db620)
* feat(decisions): add unified /v1/decisions endpoint for Jev-compatible providers (#44236)
Backport of #44236 to rc/1.105.0.
Cherry-picked from 8b1990b4bc (main).
* feat(decisions): serve System One format at /v1/systemone and OpenAI format at /v1/decisions
Backport of #45184 to rc/1.105.0.
Cherry-picked from 3339469e73 (litellm_systemone_endpoint). The playground System One tab is not on this line, so its UI hunks are dropped.
* fix(decisions): import assert_never from typing_extensions for Python 3.10
Backport of the #45184 follow-up to rc/1.105.0.
Cherry-picked from 29e715473c (litellm_systemone_endpoint).
* fix(decisions): cap OpenAI-format questions at the System One limit
Backport of the #45184 follow-up to rc/1.105.0.
Cherry-picked from 7ef132947a (litellm_systemone_endpoint).
* fix(decisions): return 400 for bodies that are not JSON
Backport of the #45184 follow-up to rc/1.105.0.
Cherry-picked from b6bb885324 (litellm_systemone_endpoint).
* test(decisions): send System One bodies to /v1/systemone in integration tests
Backport of the #45184 follow-up to rc/1.105.0.
Cherry-picked from 77872f6cda (litellm_systemone_endpoint).
* feat(decisions): add OpenAI as a Decisions provider
System One requests to openai/ models are translated to OpenAI's Decisions shape on the way out and OpenAI's answers are translated back, so both /v1/systemone and /v1/decisions can route to gpt-6-luna.
(cherry picked from commit b94bbbc318)
* refactor(decisions): move the OpenAI wire translation into llms/openai
(cherry picked from commit d209a85076)
* feat(decisions): route both request formats through a shared decisions IR
System One and OpenAI-format bodies now convert to one internal representation, and each provider translates it to its own wire format. OpenAI deployments receive the caller's messages, images, names, typed choice values, level descriptions and safety_identifier unchanged. System One providers return a 400 for image input. Cached and cache-write tokens are priced on both response shapes, and provider response extras survive the round trip.
(cherry picked from commit 0dd207fde1)
* fix(decisions): keep provider fields inside System One answers
(cherry picked from commit 1348de864f)
* fix(decisions): load on Python 3.10 and 3.11 and bill cached tokens once under custom pricing
Decisions IR dataclasses used a mappingproxy default, which is only hashable from Python 3.12, so import litellm failed on 3.10 and 3.11. Decisions usage now reaches cost_per_token as a Usage with prompt_tokens_details, so custom pricing no longer adds cached and cache-write tokens on top of input_tokens that already include them
(cherry picked from commit 39012f1c22)
* fix(decisions): honor litellm OpenAI key and base settings for the openai provider
OpenAI decisions now resolve the key from litellm.api_key, litellm.openai_key, then OPENAI_API_KEY, and the base from litellm.api_base, OPENAI_BASE_URL, then OPENAI_API_BASE, matching other OpenAI calls. Providers own their configured key and base lookup through the endpoint config.
(cherry picked from commit e465919004)
* test(decisions): clear every OpenAI base setting in the proxy OpenAI deployment test
(cherry picked from commit a2951b318d)
* fix(decisions): send OpenAI instructions for System One questions without them and reject single-option questions
(cherry picked from commit e4081f77c6)
* fix(decisions): return guardrail_information on OpenAI-format decisions when requested
Backport of the #45184 and #45214 guardrail fix.
Cherry-picked from 855f0bb765 (litellm_systemone_endpoint) as merged into #45214 at c19ada66ed (litellm_openai_decisions_provider).
---------
Co-authored-by: devin-ai-integration[bot] <158243242+devin-ai-integration[bot]@users.noreply.github.com>
Backport of #44086 to rc/1.105.0.
Cherry-picked from f95446ea39 (main).
Co-authored-by: devin-ai-integration[bot] <158243242+devin-ai-integration[bot]@users.noreply.github.com>
Cherry-pick of the squash commit of #44307, which serves gpt-5.6+ chat
completions on the native Bedrock /openai/v1/chat/completions route by
default and adds the chat_completions/ opt-in for gpt-oss and grok.
Conflict resolution: tests/unit/llms/bedrock/test_bedrock_common_utils.py
kept the line's file plus the one test the pick adds, since rc/1.105.0
does not carry #44112's neighbouring tests. The pick's
bedrock_route_for_request builds a MappingProxyType, whose import #44112
brought to main, so the import is added to litellm/llms/bedrock/common_utils.py
here.
(cherry picked from commit 233db9f285)
* fix(proxy-extras): retry P3009 when a peer already recovered the named migration row
* fix(proxy-extras): pin P3009 recovery locals as Final and cover a recovered sibling row
* test(proxy-extras): stop the P3009 ledger fakes shadowing the partition detector cursor
* fix(proxy-extras): annotate the migration ledger reader locals as Final
---------
Co-authored-by: devin-ai-integration[bot] <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* test(integration): answer the model-info refresh GET in the mixed MCP responses wire peer
The proxy's periodic model-info refresh sends GET /v1/models to the deployment api_base, which tripped the peer's /responses-only assertion when a tick landed mid-test
* test(e2e): wait for the api-keys URL after clicking Virtual Keys in onboarding
/ui already renders the Virtual Keys heading, so the helper returned before navigation finished. The late route change moved focus and closed the account menu popover in hideLiteAdmin
* test(proxy): stop two unit modules leaking app.openapi_schema and a session-wide Router
test_custom_openapi cached a stripped schema on app.openapi_schema and never cleared it, breaking later openapi route tests. test_proxy_reject_logging built a module-level Router that stayed in the live router registry all session and re-added cost-map keys during a reload. Reset the schema via monkeypatch and make the Router a function-scoped fixture
(cherry picked from commit a292fd409f)
The failure spend-log row from #42695 is written for pass-through routes that
run as LLM API routes, which a config route only does with auth: true. The
test omitted auth and passed only while config wins (#41779) registered config
entries through the typed model, where auth defaults to true. #43962 restored
the pre-config-wins registration, so the route lost that status and the row
was never written. Set auth: true on the route so the test covers the logging
it was written for without depending on that side effect
(cherry picked from commit 1d9cd9b09c)
* test(proxy): stop the proxy_server app fixture leaking LITELLM_LOG
The session app fixture set LITELLM_LOG=ERROR with os.environ.setdefault and never removed it, so later tests on the same xdist worker inherited it. test_drop_params_env_var spawns a subprocess with os.environ and lost the warning it asserts on. Scope the variable to the import with a MonkeyPatch context
* test(secret-detection): give the hand-built redaction request an ASGI path
Since #43975 _read_request_body checks the route path via request.scope, and a scope without path raised KeyError that was swallowed into an empty body, so chat_completion failed with a missing messages parameter. Real ASGI scopes always carry path
* test(integration): isolate litellm callback lists per sdk test
usage-based-routing-v2 Routers register their selector in litellm.callbacks and nothing removes it, not even Router.reset(). The counter TTL and Redis service metrics tests left their selectors behind, and the next usage routing test ran their pre-call checks against its own rpm=1 deployments, raising "Deployment over defined rpm limit". An autouse fixture now gives each sdk test copies of the callback lists and restores the originals afterwards
* test(integration): keep the owner-lookup fault proxy off the shared read replica
The owned proxy points DATABASE_URL at a scratch database but inherited
DATABASE_URL_READ_REPLICA from the replica job, so auth read the shared
database and rejected the freshly created key with token_not_found_in_db.
Drop the replica variable like the other scratch-database owned proxies
* test(integration): request every seeded key in the team owner breakdown
The aggregated team activity endpoint now caps breakdown.api_keys at the top
100 keys by default (#43398), so the 300 seeded keys came back as 100 rows.
The test guarantees each key is reported with its own owner, so ask for an
api_key_limit that covers all seeded keys
* test(integration): give every owned Redis its own port in the redis-cache container
On CircleCI every owned Redis ran on the fixed port 16379 inside the shared
redis-cache container. When an earlier server still held that port, the new
one failed to bind, readiness pinged the old server, the pidfile read failed
and cleanup then reported "Owned Redis still serves after shutdown"
Reserve an ephemeral port for the docker-exec path the same way the local
binary path already does, and refuse to start when something already serves
the chosen port so the failure names the real cause
* test(e2e): skip the Vertex Mistral partner case the e2e project cannot reach
The e2e Vertex project gets a 404 publisher model not found for vertex_ai/mistral-small-2503, so the case can only fail
* test(e2e): skip the Vertex gpt-oss partner case the e2e project never serves
vertex_ai/openai/gpt-oss-120b-maas has hit a 60s read timeout with no response headers on every run in the e2e Vertex project since the case was ported, and no other Vertex partner chat model passes there to switch to
* test(e2e): check only stored message content for a leaked card number
The Presidio spend-log check ran the card-number pattern over the whole serialized response, so a Luhn-valid usage.cost float (0.0003466000000000001) failed the streaming /v1/messages case although the stored content was <CREDIT_CARD>. The check now reads the content and text strings of the stored response, which is where a raw card would land, and still requires the placeholder there
* test(e2e): assert the proxy decodes token-array embeddings for titan
The port in #44120 carried over a legacy SDK-direct test that expected Bedrock to reject token ids with a 400. Through the proxy, /embeddings decodes token arrays to text for providers that cannot embed tokens, so titan answers 200. The test now sends a token array and its decoded sentence and requires the two vectors to match, which fails if the proxy stops decoding or decodes with the wrong tokenizer
* test(e2e): run the Bedrock extended-thinking round trip on a model that honors enabled thinking
us.anthropic.claude-sonnet-5-5 is adaptive-only, so litellm sends thinking.type=enabled with a 1024 budget as adaptive with low effort, and Bedrock returned no reasoning blocks on 5 of 5 identical Converse calls (boto3 direct agreed). us.anthropic.claude-sonnet-4-6 accepts the legacy shape verbatim and returned reasoning on 5 of 5. The non-thinking Bedrock case stays on sonnet-5-5
* test(proxy): stop unit modules forcing DEBUG logging into the event-loop lag tests
Five tests/unit modules set verbose_proxy_logger to DEBUG at import, so every xdist worker that collected them logged the 2.4MB pass-through response from a worker thread, and secret redaction of that line held the GIL for ~0.8s+ inside the timed window. The lag tests now pin the LiteLLM loggers to WARNING and freeze gc while timing, and the module-level DEBUG overrides are removed
* test(e2e): cite the tokenizer and date behind the titan token-array fixture
* test(e2e): let migration seed replicas finish their request-log indexes before cloning
Since #43948 a serving proxy builds the two LiteLLM_SpendLogs indexes on a background thread after it reports ready. The seed fixtures stopped the replica at readiness, so every cloned legacy database lacked an index no real deployment would be missing, and the v2 baseline diff refused it. Seeds now wait until both indexes exist and are valid in the database's schema
* test(passthrough): give the pass-through MockRequest an httpx URL and ASGI scope
#43626 made get_request_route read request.scope during pass-through kwarg setup; the MockRequest in tests/unit/passthrough had neither a scope nor a URL object, so both stream-param tests raised before reaching the code they check. Mirrors the repair #43626 made to the tests/pass_through_unit_tests fake
* test(integration): ignore foreign allow_all_keys MCP servers in the access matrix tool list
test_toolset_gateway_url_serves_a_team_granted_toolset_to_a_key_without_its_own_grant (#43908) registers an allow_all_keys server on the shared gateway, and allow_all_keys servers are listed to every key by design, so a matrix case running on another xdist worker at the same time saw its tools. The matrix now drops tools of allow_all_keys servers it did not create, read from LiteLLM_MCPServerTable before and after listing, and still compares everything else exactly
(cherry picked from commit 9b5562f89b)
Backport of #44203 to rc/1.105.0.
Cherry-picked from 4758fce91a (main).
Co-authored-by: devin-ai-integration[bot] <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* feat(mcp)!: disable stdio MCP servers by default (#44066)
* feat(mcp)!: disable stdio MCP servers by default
stdio MCP servers now only run when the proxy is started with
LITELLM_ENABLE_MCP_STDIO=true. While it is off, existing stdio servers stay
registered but never start: tool listings skip them quietly, direct tool
calls and health checks return a 403 naming the env var, and creating or
updating a stdio server is rejected. The flag is read from the process
environment only, so DB-stored environment_variables cannot turn it on.
The UI reads mcp_stdio_enabled from /.well-known/litellm-ui-config to grey
out the stdio transport, show a banner on stdio forms, and badge stdio
server cards.
BREAKING CHANGE: stdio MCP servers are off by default. Set
LITELLM_ENABLE_MCP_STDIO=true in the proxy environment and restart to keep
using them.
* fix(mcp): ignore stdio flag from config file and read UI flag from the selected worker
LITELLM_ENABLE_MCP_STDIO set under environment_variables in config.yaml is now skipped like the DB-stored value, so only the process environment can enable stdio. The dashboard reads mcp_stdio_enabled from the proxy it is managing, so a control plane shows each worker's own setting.
* test(mcp): cover non-mapping payloads in the shared transport validator
* fix(mcp): skip blocked stdio servers quietly in every listing and keep the UI unchanged until the flag loads
Prompt, resource and resource-template listings now skip a blocked stdio server at debug level like tool listing does, instead of logging a warning per server on every call. The dashboard only treats stdio as disabled once the proxy explicitly reports mcp_stdio_enabled false, so a proxy with the flag on, or an older one without the field, renders exactly as before with no flicker while loading.
* fix(mcp): route blocked stdio tool calls to the flag error and warn once per server
A gateway tools/call naming a blocked stdio server's tool now returns the
LITELLM_ENABLE_MCP_STDIO message instead of "Tool not found".
The "will not start" warning moves out of build_mcp_server_from_table, which
DB reload re-runs on every cycle for rows with a NULL updated_at and which
drafts and test-connection also call. It now fires when a row first enters
the registry or changes transport.
* fix(ui): explain on the server detail page why a stdio server is inert
The Overview and MCP Tools tabs showed "No tools available" with no reason
while stdio is disabled. The detail page now shows the same warning banner
as the edit form, and hands off to the form's banner once editing starts.
* refactor(ui): name the stdio banner conditions on the server detail page
Keeps local/no-long-condition-chain within its budget
* fix(proxy): log the ignored DB-stored LITELLM_ENABLE_MCP_STDIO warning once
The DB config sync re-reads environment_variables on every cycle, so a stored
flag logged the warning on each sync per worker
(cherry picked from commit 2b85808011)
* chore: update Next.js build artifacts (2026-10-02 17:40 UTC, node v24.19.0)
* refactor(types): replace Any with proven types in 9 files
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* test(integration): pin xecguard-adjacent guardrail retry, mcp mixed tools, jwt routing and complexity router paths
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* refactor(types): keep only live-provable Any removals
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* refactor(types): type cache_hit as bool | None on the sync success path
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
---------
Co-authored-by: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* test(straiker): assert a saved api_version v1 with an sk_agt_ key routes to v3 (#44106)
Co-authored-by: yucheng <yucheng@berri.ai>
Co-authored-by: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* test(straiker): ignore zombie workers when counting uvicorn children after a kill
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* test(straiker): describe the saved v1 cell by the v3 route it asserts
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
---------
Co-authored-by: devin-ai-integration[bot] <158243242+devin-ai-integration[bot]@users.noreply.github.com>
Co-authored-by: yucheng <yucheng@berri.ai>
* test(e2e): move live-provider legacy tests into tests/e2e
Port legacy tests that exercise real providers into the tests/e2e suites that own them, using the harness (/model/new plus deferred cleanup) and asserting on what the caller receives. Delete legacy tests already covered at equal or stronger strength by e2e, integration or unit tests, and drop the now empty ocr_testing CircleCI job
* test(e2e): address review on the live-provider test move
Assert the SSE error frame a client actually receives when a post_call guardrail blocks a stream, and require a tool call for every requested city before checking the answer. Restore the OCR matrix and its CircleCI job, the Claude Agent SDK streaming test, and test_async_create_batch, since their SDK-level and callback assertions have no equivalent in tests/e2e
* test(e2e): accept both guardrail block shapes on a blocked stream
A post_call block before the first chunk reaches the client as HTTP 400 with either a JSON error body or a single SSE error frame, depending on whether the block surfaced as an exception or an error chunk. Assert the policy message is present and the blocked output is absent in both
* test(realtime): restore direct SDK realtime tests against OpenAI
The e2e realtime tests go through the proxy and the remaining SDK tests either mock the upstream or assert less, so keep the direct litellm._arealtime tests with and without intent, and TestOpenAIRealtime::test_realtime_connection, in place
* test: make realtime and Nova stream checks deterministic
The direct SDK realtime tests now fail on a refused connection instead of skipping. The with-intent test asserts OpenAI rejects the exact intent value sent, which only happens when the intent is forwarded. The Nova /v1/messages stream test asserts stream structure, stop reason and usage instead of model wording
* test(realtime): own intent forwarding with a unit test instead of a live rejection
Assert litellm._arealtime passes the intent query param into the OpenAI realtime websocket URL, which is the behavior LiteLLM owns, and drop the live test that depended on OpenAI's rejection wording
* fix(azure_storage): keep client call ids from sharing one Data Lake file
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* docs(azure_storage): tighten adls_safe_file_name docstring
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* fix(azure_storage): give ids with dot or empty path segments their own Data Lake file
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* test(azure_storage): cover failure, cache-hit and edge call ids in Data Lake file names
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* test(azure_storage): run the Data Lake file name cells on two proxy workers
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
---------
Co-authored-by: yucheng <yucheng@berri.ai>
Co-authored-by: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* test(integration): move legacy proxy, router and Redis tests into tests/integration
Port 39 legacy tests to the integration tier that owns them, running against
the scripted upstream, local Postgres and Redis, test-owned wire peers and
owned proxies. Delete 5 legacy tests whose contract is already owned by an
existing integration test, and remove the legacy functions, files and helpers
left unused.
* test(integration): cover recovery of a spent key after its budget is raised
* fix(guardrails): straiker v3 routes sk_agt_ keys to v3 and stops reading a missing verdict as allow
An sk_agt_ key always calls /api/v3/detect, even when the guardrail was saved with
api_version 'v1' by the old shared default: the v1 webhook rejects that key with 401.
A 200 with no decision, or permissionDecision 'ask', now takes the failure policy
instead of allowing the request. A detect-mode action is still not a block.
A response-phase block is no longer remembered under the request, so asking the same
question again is scored instead of refused from memory. The replay memory is scoped
by principal and session together, so two principals on one session id never share a
block. A text-only /guardrails/apply_guardrail call relays the text as a user turn.
The agent_ref description now matches the code: the configured value wins.
* fix(guardrails): straiker v3 relays text beside an empty messages list and keys the memory on the key
/guardrails/apply_guardrail sends `messages: []` beside `text`; an empty list is no
conversation, so the text is relayed as the user turn. A verdict whose blocked_by is not
a list states no decision and takes the failure policy, the same way a missing decision
does. A key that names no user is still the caller, so the replay memory is keyed on the
key when no user is known.
* test(guardrails): build the straiker replay-scope request data without mutating it
---------
Co-authored-by: PhimmStraiker <PhimmStraiker@users.noreply.github.com>
* feat(tracing): add SQL queries and schema-aware query help
* test(tracing): verify help requests and sync API types
* refactor(tracing): render query help with Askama
* refactor(tracing): use jinja extension for query guide
* fix(tracing): preserve query help when discovery fails
* feat(tracing): enforce team SQL scope with managed ClickHouse readers
* test(tracing): verify reads with one ClickHouse URL
* fix(tracing): revoke rotated trace reader credentials
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* refactor(tracing): streamline query help catalog assembly
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* fix(tracing): run query help discovery sequentially
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* test(tracing): update reader setup request expectations
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
---------
Co-authored-by: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
Remove the ENFORCE_PRISMA_MIGRATION_CHECK opt-in. When PrismaManager.setup_database returns False (database unreachable, connection retries exhausted, or prisma migrate deploy failing after retries) the proxy now always prints the red failure message and exits 1 instead of serving requests against a database whose schema may be behind the code. The --enforce_prisma_migration_check flag stays as a hidden no-op that prints a one-line deprecation warning so existing container args keep parsing; the env var is no longer read anywhere. The standalone migration entrypoint always runs run_server(("--skip_server_startup",)), and the integration launchers drop the flag.
Co-authored-by: yassin <yassin@berri.ai>
Co-authored-by: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
The daily spend tables store date as text, so a request date that strptime accepts but that is not spelled YYYY-MM-DD (2026-9-24, 2026-09-4, full-width digits) was compared as raw text against canonical rows and matched nothing, and the export route copied it into Content-Disposition, which fails latin-1 encoding and returned 500. A shared parse_canonical_date_range now rejects any spelling whose round trip differs from the input, so every bounded daily activity route (user, team, tag, organization, customer, agent: aggregated, aggregated/keys, search, model top keys, export, cache leakage) and the paginated get_daily_activity path answer 400 before touching the repository, and the export filename is built from the validated dates.
Co-authored-by: yassin <yassin@berri.ai>
Co-authored-by: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
The shared exclusion predicate negated a PostgreSQL ANY comparison without handling NULL, so NULL entity ids evaluated to UNKNOWN and dropped out of the Unassigned bucket whenever exclude_*_ids was set. The paginated daily rows Prisma filter had the same NOT IN shape and gets the same IS NULL OR NOT IN treatment
Co-authored-by: yassin <yassin@berri.ai>
Co-authored-by: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* feat(mcp): add Microsoft 365 (Graph) server to the MCP catalog
* fix(mcp): signpost the self-hosted Microsoft 365 URL and pin the catalog entry in tests
* test(mcp): pin both shipped copies of a catalog icon to the same bytes
---------
Co-authored-by: mateo-berri <277851410+mateo-berri@users.noreply.github.com>
* fix(proxy): persist SSO display name as user_alias on login
Generic/Microsoft SSO already parsed the IdP display_name, first_name and last_name into the SSO result, but the user upsert only wrote user_email and user_role, so the Users table never showed a name. Store the display name (first + last as fallback) as user_alias on first login and on later logins of users whose alias is still empty; never overwrite an alias already set. Whitespace-only names are treated as missing.
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* fix(proxy): keep stored user_email when SSO login carries no email claim
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* revert: keep stored user_email change, login writes the IdP email as before
Reverts 47c65cecff. A stored email staying eligible for email-based account linking after the IdP stops sending it is not wanted; the PR goes back to the user_alias fix only
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
---------
Co-authored-by: yassin <yassin@berri.ai>
Co-authored-by: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
The timeout reliability tests rely on a 1ms deadline the real backend always
misses. With E2E_PROVIDER_CACHE on, the deployment pointed at the cache edge,
and its healthy sibling in the same test had already recorded a response for
the same canonical request, so the edge answered from Redis inside the 1ms read
window. Build 342 of litellm-e2e saw test_timeout_trips_cooldown_then_recovers
get a 200 from the timing-out deployment itself, with a recording made about
12 hours earlier. Both timeout helpers now register on the live provider path,
which PROVIDER_CACHE.md reserves for tests that need real provider timing
* feat(proxy): add LITELLM_DISABLE_LAZY_ROUTES to register optional routers at startup
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* fix(proxy): register eager lazy routes at startup so late eager routes keep precedence
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* refactor(proxy): share one optional-feature install path between lazy and eager registration
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* docs(proxy): say eager lazy routes register at worker startup
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* refactor(proxy): name the import callable passed to _install
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* test(integration): prove a startup hook can drop eager lazy routes for good
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* test(integration): strip the lazy-routes flag from the lazy-mode control proxies
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* test(proxy): cover the lazy warmup route registering a feature
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* ci(unit): run tests/unit/proxy/test__lazy_features.py in the proxy-server-core shard
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* test(integration): audit cells for LITELLM_DISABLE_LAZY_ROUTES
Flag spellings, /openapi.json at boot, the warm-up route in both modes, /mcp/proxy ahead of the /mcp mount,
route table and OpenAPI parity with a fully warmed lazy proxy, a broken optional import in both modes, every
client SDK against the completion endpoints, a boot burst with a killed worker, and a restart
* fix(proxy): keep config pass-through routes ahead of eagerly registered features
With LITELLM_DISABLE_LAZY_ROUTES set, features registered before the proxy lifespan
added config pass-through endpoints, so a pass-through overlapping a feature path
(e.g. a self-hosted /langfuse) lost to the built-in route. Restore lazy mode's
registry order once startup finishes, without bringing back routes a startup hook removed
---------
Co-authored-by: ryan <ryan@berri.ai>
Co-authored-by: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* fix(bedrock): accept Converse messages with no content key
A user or tool message whose content key is missing (or null, which the
message cleanup strips) made every Bedrock Converse request fail with
APIConnectionError 'content' before reaching Bedrock. The Converse
transform now reads content with .get for those messages, as it already
did for assistant messages: a content-less user message adds no block
and a content-less tool message becomes a toolResult with empty content.
The str branch also sends the continue message text instead of the
original whitespace-only text.
* fix(bedrock): send the continue message for a content-less user turn
* test(bedrock): type the content-less Converse message test parameters
* fix(bedrock): accept a Converse system message with no content key
* refactor(bedrock): read the system message content with get
* test(integration): audit Bedrock Converse messages without content
Adds the /audit cells for a chat message whose content key is missing or
null on a Converse-routed Bedrock deployment: happy, sad, edge, and chaos
rows through the OpenAI SDK, the Anthropic SDK, and raw httpx against the
scripted upstream, asserting the caller's response, the body the peer
received, and the spend row. The owned-proxy readiness deadline in the
integration harness is now INTEGRATION_PROXY_READY_SECONDS (default 70).
* test(integration): bound stray spend rows in the mid-burst restart cell
---------
Co-authored-by: mateo-berri <277851410+mateo-berri@users.noreply.github.com>
* fix(mcp): expand team and dashboard grants when listing and serving toolsets
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* fix(mcp): honor team toolset grants on the responses gateway path and expose a public team permission lookup
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* chore(mcp): drop the toolset route docstring tweak so the OpenAPI snapshot stays unchanged
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* fix(mcp): scope inherited toolset grants by the key's own MCP ceiling
A key that declares any MCP grant of its own keeps only its own toolsets, one
that declares none inherits its team's, and require_key_mcp_access_defined
stops a virtual key inheriting while dashboard sessions and admitted users
still do. Adds the direct, no-grant, admin and key-ceiling integration cases
and makes the LLM gateway toolset case discriminate a scoped toolset from the
aggregate grant
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* fix(mcp): resolve toolset grants per admitted source and enforce the live team roster
Dashboard sessions and gateway-admitted users now expand into the admitted
subject's per-team sources when resolving toolset grants, so a team-granted
toolset is not capped by the user's own MCP row and is reachable on the
namespaced route. A cached team id no longer grants a toolset unless the live
roster still lists the user, a team lookup fault only drops that team's
inherited grant, and /team/member_add evicts the cached team object so the new
member is authoritative immediately
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* fix(mcp): read a key's named object permission before letting it inherit team toolsets
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* refactor(mcp): inject the toolset grant resolver into scope helpers so tests stop patching MCPRequestHandler
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* fix(mcp): drop the duplicate admitted_subject_sources wrapper after merging main and follow its renamed resolvers
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* fix(mcp): honour fresh policy and the session resource scope on pinned toolsets
A pinned toolset scope now reads the toolset through the writer when the admitted session
requires fresh policy, so a tool revoked from the toolset is gone on the next request. A
gateway bearer scoped to one server can only open a toolset that names that server
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* fix(mcp): keep operator-open servers out of toolset gateway urls
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* test(mcp): audit cells for team-granted toolsets across every surface
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* test(mcp): bound toolset edit convergence by both cache layers
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* test(mcp): drop toolset integration cells that test behavior this PR does not change
Repeat-read byte identity, 20 concurrent calls, a stopped peer and a killed worker are covered
generically by test_mcp_resilience.py and test_mcp_user_env_vars.py. The toolset edit cell asserts
pre-existing cache propagation and flaked locally with connection resets while polling
* chore(mcp): drop mutable-ok suppressions that main's LIT013 now flags as unused
* chore(mcp): keep the require_key_mcp_access_defined read from adding an unknown-argument type error
---------
Co-authored-by: ryan <ryan@berri.ai>
Co-authored-by: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
The two SpendLogs index migrations shipped in v1.103.0 each break one table shape: the plain CREATE INDEX holds a SHARE lock on a large unpartitioned table and the CONCURRENTLY one fails with 0A000 on a partitioned parent. Both files are now inert and the indexes are built by a table-driven, shape-aware step after migrate deploy: CONCURRENTLY on a plain table, ON ONLY the parent plus per-partition CONCURRENTLY and ATTACH PARTITION on a partitioned one. The migration job builds them synchronously and exits non-zero on failure; a serving proxy that ran migrate deploy itself builds them in the background off the readiness path. A valid index of the same definition under another name is renamed and reused, an invalid one is rebuilt, and extra copies are reported with their DROP INDEX statement instead of being dropped. The migration checker rejects any CREATE INDEX on LiteLLM_SpendLogs or LiteLLM_ErrorLogs in future migrations
Co-authored-by: yassin <yassin@berri.ai>
Co-authored-by: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* add test case for /rag/query and stronger auth check
* style(proxy): ruff format auth_checks.py
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* fix(proxy): build rag query vector store ids immutably and test the no-registry path
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* test(proxy): integration coverage for /v1/rag/query vector store allowlist
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* test(proxy): audit cells for /v1/rag/query vector store allowlist
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* test(proxy): consolidate vector store allowlist audit coverage
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* fix(proxy): type RAG vector store request body
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
---------
Co-authored-by: Mrinal Chanshetty <mrinal@berri.ai>
Co-authored-by: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
The usage card hid actual and baseline spend unless every older session
could be rebuilt from SpendLogs within two seconds, which on a real
gateway it never was. Each complexity router's actual spend is now its
rollup spend and its baseline is spend plus recorded savings, for old
and new requests alike, so the benchmarks and session endpoints never
scan SpendLogs. Adaptive and quality routers record no savings baseline
and stay out of the compared totals; savings_estimated_classifier_cost
is kept and covers the same compared requests
* feat(proxy): gzip buffered responses for clients that accept it
Large JSON reads like /user/daily/activity/aggregated shipped tens of MB
uncompressed. Compress single-message bodies of 500B or more when the
client's Accept-Encoding allows gzip (q-values and the wildcard honored).
Streamed and etagged responses pass through untouched, every negotiable
response carries Vary: Accept-Encoding, and bodies of 1MB or more are
compressed in a worker thread
* fix(proxy): skip partial and no-transform responses in gzip and always release the held start
The gzip gate now also skips 206 Partial Content and Cache-Control: no-transform,
since compressing either breaks byte ranges or ignores an explicit ban on transforms.
A response start without a headers key no longer raises, and a start the app never
follows with a body message is forwarded when the app returns instead of being dropped.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* feat(tool-policies): show the user who owns the key that discovered a tool
GET /v1/tool/list and GET /v1/tool/{tool_name} resolve the discovering key's owner from the verification token and user tables at response time and return it as a nullable user field. The Tool Policies page adds a User column that shows alias, then email, then ID, with the same cell the Virtual Keys page uses. Keys without an owner, deleted owners, and rows without a key hash show no user, and a database failure in the owner lookup keeps the tools listed with user null
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* fix(tool-policies): bound the owner lookup with chunked membership queries
The key-by-token and user-by-id lookups behind the tool rows' user field
put every distinct key hash into one IN list. BaseRepository gains
find_many_in, which runs the repository's chunked membership query and
converts the rows like find_many does, and the owner lookup uses it
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* test(tool-policies): cover the owner column across the tool routes and the dashboard
Integration cells for the direct, detail and filtered tool routes, owners without alias or email, deleted owners and keys, keyless and unknown-key historical rows, more keys than one membership chunk, repeated reads, two-worker reads during discovery and a failed owner lookup. A Playwright cell drives the bundled Tool Policies page against the live proxy and follows the owner link
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
---------
Co-authored-by: ryan <ryan@berri.ai>
Co-authored-by: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* ci(circleci): test Redis behavior against local Redis and print short tracebacks
The redis_caching_unit_tests job ran three legacy files against the shared remote Redis. The DualCache and
batch-read logic that never needed a server now lives in tests/unit/caching/test_dual_cache.py with a mocked
RedisCache, and the behavior that does need one (the increment-with-floor Lua script, read-through, deletes,
batch reads) moved to tests/integration, which starts a local Redis. test_returned_settings only read
REDIS_PORT and is replaced by a unit test of Router.get_settings
CircleCI pytest runs now use --tb=short so failure output stays readable in the test results tab
* ci(integration): print short tracebacks from run.py and allow Redis in the sdk shard
