litellm/schema.prisma
Mateo Wang 6d6eda8101
[internal copy of #28008] Support MCP OAuth passthrough and issuer-scoped JWT auth (#28356)
* fix(proxy): point /metrics 401 at the opt-out flag

Operators upgrading past 35bbca60b0 (which made /metrics auth
default-on) see "Malformed API Key passed in. Ensure Key has 'Bearer '
prefix." with no hint that
litellm_settings.require_auth_for_metrics_endpoint: false restores the
previous unauthenticated behavior. Append that discovery hint to the
existing 401 body so a Prometheus scraper that breaks after upgrade
has a clear migration path. No behavior change.

* fix(proxy): bound budget reservation per request instead of pinning to remaining headroom

reserve_budget_for_request fell back to reserving the entire remaining
team/key/user headroom whenever a request omitted max_tokens, which
pinned the spend counter at max_budget for the duration of the
in-flight request and false-positive-blocked every concurrent or
back-to-back request until the success callback reconciled. Surfaced
as an integration-test team being budget-blocked at its $2000 cap
while DB spend was $0.144.

Switch the missing-max_tokens path to a fixed default of 16384 output
tokens (mirrors parallel_request_limiter_v3's DEFAULT_MAX_TOKENS_ESTIMATE
precedent), and clamp explicit max_tokens at the model's
max_output_tokens for reservation accounting only. The outbound request
body is unchanged, so providers see whatever the caller actually sent;
only the local integer used to compute reservation cost is bounded.
This also prevents a hostile max_tokens=999999999 from inflating one
request's reservation up to the entire team headroom.

For Opus 4.7 (output $25/M, max_output 128K) on a $2000 budget the
worst-case per-request reservation drops from "everything left" to
$3.20, raising admittable concurrency from 1 to ~625.

* fix(proxy): reserve per-image cost for image-generation requests

Image-generation routes (dall-e-3, flux, etc.) have no per-token output
cost so they fell through to the no-reservation read-time-only path.
Concurrent image requests against a depleted budget could all pass
common_checks (counter exactly at max_budget passes the strict-`>`
gate) and reach the provider before reconciliation caught up.

Add per-image reservation in _estimate_request_max_cost_for_model:
when the model has a per-image cost field, reserve `n × cost_per_image`
upfront. The atomic counter increment serializes concurrent admissions,
so the second request sees the post-first-reservation counter and
raises BudgetExceededError instead of silently leaking through.

Both `output_cost_per_image` and `input_cost_per_image` are honored —
naming is inconsistent across providers (OpenAI dall-e-3 uses
input_cost_per_image, aiml/dall-e-3 uses output_cost_per_image for
the same per-generated-image price).

Per-pixel pricing (DALL-E 2 size variants) and TTS/STT routes still
fall through to read-time enforcement; those are follow-ups.

* fix(proxy): gate image-gen reservation strictly on model mode

The previous detection treated any model with input_cost_per_image
or output_cost_per_image as image generation. Several chat and
embedding models carry those fields to price multimodal vision input,
not generated images:

- gemini-3.1-pro-preview (mode=chat) has output_cost_per_image=0.00012
  alongside input/output token pricing.
- azure/gpt-realtime-* (mode=chat) has input_cost_per_image=5e-6.
- amazon.titan-embed-image-v1 (mode=embedding) has
  input_cost_per_image=6e-5.

For these models the image-gen branch fired first and reserved a
fraction of a cent per request, short-circuiting the token-priced
path entirely. Long Gemini chats reserved 1 × $0.00012 instead of
the true token cost.

Gate strictly on mode in {"image_generation", "image_edit"}. All 197
real image_generation entries and all 31 image_edit entries
(Flux Kontext, Stability inpaint/outpaint, etc.) carry the right mode,
so the field-presence fallback was unnecessary.

Adds regression tests for the chat-model-with-image-cost-field case
and for image_edit reservation.

* build(packaging): relax core runtime pins to ranges

Backport of #27241 onto litellm_1.84.0rc2.

The 12 entries in `[project.dependencies]` were exact `==` pins, a side
effect of the Poetry -> uv migration. This forces every downstream
package that lists litellm as a dependency to downgrade common runtime
libraries (openai, pydantic, aiohttp, click, jsonschema, ...) to the
exact versions we ship.

Switch to lower-bounded ranges with upper bounds where the upstream
package is pre-1.0 or has a known breaking-major-version policy.
Reproducibility for our Docker proxy and CI continues to come from
`uv.lock`, which is regenerated here as a metadata-only diff.

Conflict resolution vs upstream merge:
- The upstream merge commit also surfaced unrelated context entries
  (nvidia-riva-client, soundfile/stt-nvidia-riva extra) that exist in
  staging but not in rc2. Those are not part of #27241's intent and
  were dropped from the resolution; the rc2 uv.lock keeps its existing
  entry set, only the 12 specifier strings changed.
- `uv lock --check` passes (392 packages resolved, no drift).

* build(packaging): raise jinja2 floor to 3.1.6

Our `uv.lock` already resolves jinja2 to 3.1.6, so Docker / CI installs
get that version. The `pyproject.toml` floor was lagging at 3.1.0,
which means downstream consumers using `--resolution=lowest-direct` or
older constraint files can land on 3.1.0-3.1.5 instead of the version
we actually test against.

Aligns the declared floor with the resolved version so external
installers see the same baseline our test matrix exercises.

`uv lock` diff is metadata-only (no resolved-version drift).

* fix(mcp): forward extra_headers for OpenAPI MCP tools

OpenAPI-generated tools only applied static closure headers and BYOK
Authorization via ContextVar. Copy MCPServer.extra_headers from the
incoming MCP request into _request_extra_headers (set in server.py before
local tool dispatch), merge in openapi_to_mcp_generator via a small helper.

OAuth2 M2M: do not forward caller Authorization from raw_headers (same rule
as _prepare_mcp_server_headers for managed MCP).

Adds TestRequestExtraHeaders and clarifies mcp_server_manager registration
comment.

Fixes #26794

Co-authored-by: Cursor <cursoragent@cursor.com>

* refactor(mcp): access has_client_credentials on MCPServer directly

Greptile: getattr default was redundant; property exists on MCPServer and
mcp_server is non-None inside the extra_headers forwarding block.

Co-authored-by: Cursor <cursoragent@cursor.com>

* fix(mcp): static headers win over forwarded headers in OpenAPI MCP

Match the existing MCP invariant in merge_mcp_headers and the managed MCP
path: operator-configured static headers always override caller-forwarded
headers on name conflict, with case-insensitive comparison so different
casing cannot bypass the precedence. _request_auth_header (BYOK) still
overrides Authorization last.

Addresses Veria review on PR #27383.

Co-authored-by: Mateo Wang <mateo-berri@users.noreply.github.com>

* fix(proxy): always merge caller-supplied tags into request metadata

Caller-supplied tags (`x-litellm-tags` header, body `tags`, `metadata.tags`)
were silently dropped unless the key/team had
`metadata.allow_client_tags: true` set. Restore the documented behavior:
tags from the request always flow into `metadata.tags` and union with any
admin-configured static tags from key/team/project metadata.

Removes the `allow_client_tags` opt-in flag from the pre-call pipeline.
The flag was only ever read here; it has no schema or endpoint footprint,
so leftover values in existing key metadata are inert.

Test cleanup mirrors the simplification: drop the three tests that
verified the strip-when-not-opted-in path, drop the `allow_client_tags`
fixture lines from the merge/union tests.

* docs(proxy): refresh stale comments referencing removed tag strip

The tag-strip block was removed in the parent commit but two surrounding
comments still referenced "tags without opt-in" and "runs AFTER the
strip". Update them to describe the remaining user_api_key_* and
_pipeline_managed_guardrails strip that the snapshot/merge ordering
actually protects against.

* chore: reject bare str at file-input sinks to prevent local-file read (#27762)

Cherry-pick of #27762 onto litellm_1.84.0rc2.

* chore: reject bare str at file-input sinks to prevent local-file read (#27667)
* fix: use os.PathLike in ocr sink and check truthy reasoningSummary for bridge
  - ocr/main.py: widen Path check to os.PathLike for consistency with other sinks
  - main.py: bridge condition checks truthiness of reasoning_summary, not just None
* fix: remove unused pathlib.Path import in ocr/main.py

Co-authored-by: yuneng-jiang <yuneng@berri.ai>
Co-authored-by: ryan-crabbe-berri <ryan@berri.ai>
Co-authored-by: stuxf <70670632+stuxf@users.noreply.github.com>

* Strip SERVER_ROOT_PATH before lazy-feature prefix match

LazyFeatureMiddleware compared the raw scope path against registered
prefixes (e.g. /policies), so requests under a server root path like
/api/v1/policies/... never matched, the feature never loaded, and the
endpoint returned 404. Strip the configured root path before matching,
normalizing trailing slashes and enforcing a component boundary so
/api does not falsely match /apiv2.

* Cache normalized SERVER_ROOT_PATH at middleware init

SERVER_ROOT_PATH is a process-startup env var. Read it once in
__init__ instead of calling get_server_root_path() + rstrip on every
request that arrives before all lazy features have loaded.

* chore(proxy): backport /key/regenerate ownership-rebind + premium-gate guards (#27793)

Backport of #27793 onto litellm_1.84.0rc2.

A non-admin caller could rebind their own key's user_id via /key/regenerate.
_execute_virtual_key_regeneration had org/team guards but no user_id guard,
and prepare_key_update_data did not strip the field — it survived
model_dump(exclude_unset=True) into the Prisma update. On the next request,
_return_user_api_key_auth_obj resolved the rebound user_id against
litellm_usertable and returned PROXY_ADMIN whenever the target row's
user_role was admin.

/key/update had the equivalent guard inline at _validate_update_key_data;
extract it to a shared helper _validate_caller_can_change_key_ownership and
call from both /key/update and _execute_virtual_key_regeneration.

Also tighten the premium gate that allowed the master-key rotation branch to
skip the enterprise check. The previous predicate was a field-presence test,
not an identity check. Verify the caller actually holds the master key via
_is_master_key before allowing the non-premium path.

Block explicit-null user_id and empty-string user_id as removal attempts;
both 403-reject for non-admin callers.

* fix(proxy): expose db status on public /health/readiness

Backport of #27866 onto litellm_1.84.0rc2.

External readiness probes consumed the legacy detailed payload's `db`
field to drive alerting and pod-rotation decisions. Stripping the body
to {"status": "healthy"} broke those probes silently — the HTTP code
still flipped to 503, but probes checking body.db == "connected"
treated the response as healthy.

Add `db` back to the unauthenticated payload. The rest of the diagnostic
fields (litellm_version, callbacks, cache, log_level) stay behind
/health/readiness/details so the recon-leak gate from #26912 holds.
Values match the legacy contract: "connected", "disconnected",
"Not connected". The 503-on-DB-disconnect behavior from LIT-2607 is
preserved.

* fix(ui): fetch version + debug flag from /health/readiness/details

The proxy moved `litellm_version`, `is_detailed_debug`, and other
diagnostic fields off the public `/health/readiness` payload behind
an auth-gated `/health/readiness/details` endpoint. The navbar
version tag and the detailed-debug-mode banner stopped working
because they were still reading those fields from the unauthed
response, which no longer contains them.

Replace `useHealthReadiness` with a `useHealthReadinessDetails`
hook that takes an `accessToken` argument and sends a Bearer header
to the auth-gated endpoint. The hook stays disabled while
`accessToken` is falsy, so the navbar can keep rendering on the
public model hub (where the token is null) without triggering an
auth redirect or a 401-loop.

* fix(ui): disable retries on readiness/details + cover token forwarding

Two small follow-ups on the readiness/details migration:

- Set `retry: false` on the query. The payload feeds a passive
  navbar tag and a debug banner; a 401 from an expired token
  shouldn't fan out into three retries against the proxy.
- Add navbar specs that assert the `accessToken` prop is forwarded
  into the hook (matches the DebugWarningBanner spec). Without
  this, the navbar could silently regress to passing `undefined`
  and the existing tests wouldn't catch it.

* chore: update Next.js build artifacts (2026-05-14 03:52 UTC, node v20.20.2)

* Merge pull request #27898 from stuxf/chore/banned-params-extra-body-cover

chore(proxy): cover extra_body + azure_ad_token in banned-params check

(cherry picked from commit a6a9d8edf0)

* Merge pull request #27801 from stuxf/chore/get-instance-fn-runtime-s3-gate

chore(proxy): refuse remote-URL instance-fn loads outside config-file path

(cherry picked from commit e3e5209f51)

* fix: block client-side pricing injection via request body

Authenticated clients could supply CustomPricingLiteLLMParams fields
(input_cost_per_token, output_cost_per_token, etc.) in the request body.
These were forwarded to register_model() in main.py, permanently mutating
the shared global litellm.model_cost dict for all users on the instance.

Adds all CustomPricingLiteLLMParams fields to _BANNED_REQUEST_BODY_PARAMS
so is_request_body_safe() rejects them before they reach completion().
New pricing fields added to CustomPricingLiteLLMParams are auto-covered.

Admin opt-in via allow_client_side_credentials or
configurable_clientside_auth_params still works as before.

Co-Authored-By: Claude Sonnet 4.6 (1M context) <noreply@anthropic.com>

* fix: block SSRF fields in RAG ingest vector_store config

aws_sts_endpoint, aws_web_identity_token, and aws_bedrock_runtime_endpoint
in ingest_options.vector_store were passed directly to the Bedrock ingestion
class, which reads them into boto3 STS client construction. Any authenticated
caller could redirect AssumeRole calls to an attacker-controlled server,
leaking the proxy's instance profile credentials.

Calls is_request_body_safe() on ingest_options["vector_store"] before
forwarding to litellm.aingest(). Same banned-params list and admin opt-in
escape hatch (allow_client_side_credentials) as the /chat/completions path.
ValueError from the safety check is caught and re-raised as HTTP 400.

Co-Authored-By: Claude Sonnet 4.6 (1M context) <noreply@anthropic.com>

* fix: harden /key/update authorization checks (#27878)

* fix: patch Host-header auth bypass in get_request_route

Starlette reconstructs request.url from the Host header. A malformed
Host like `localhost/?x=1` causes Starlette to build the full URL as
`http://localhost/?x=1/health`, which url-parses to path="/". Since "/"
is in LiteLLMRoutes.public_routes, all protected routes became reachable
without authentication.

Fix: read scope["path"] (set by uvicorn from the HTTP request line,
not derivable from headers) instead of request.url.path. Sub-path
deployments are handled via scope["app_root_path"] / scope["root_path"],
mirroring Starlette's own base_url construction logic.

Affected variants confirmed fixed:
  Host: localhost/?x=1
  Host: localhost:4000/?x=1
  Host: localhost/#test
  Host: localhost:4000/#test

Co-Authored-By: Claude Sonnet 4.6 (1M context) <noreply@anthropic.com>

* style: reduce comments in route fix

Co-Authored-By: Claude Sonnet 4.6 (1M context) <noreply@anthropic.com>

* fix: block credential fields in RAG ingest vector_store options

Credential fields (vertex_credentials, aws_access_key_id, api_key, etc.)
in ingest_options.vector_store are now rejected at the API boundary with
a 400 error. Credentials must be configured server-side.

Previously any authenticated user could supply a vertex_credentials dict
with type=external_account pointing credential_source.file at an
arbitrary path (e.g. /proc/1/environ) and token_url at an
attacker-controlled server. google-auth's identity_pool.Credentials
refresh() would read the file and POST its contents to the attacker.

Co-Authored-By: Claude Sonnet 4.6 (1M context) <noreply@anthropic.com>

* fix: block /key/update self-escalation by assigned users

Non-admin users who were assigned a key (created_by != caller) could
update any non-budget field — models, rpm_limit, guardrails, etc. —
without admin authorization, allowing privilege self-escalation.

Gate: only the key creator (created_by == caller) may edit their own
key without admin check; budget changes always require admin regardless
of creator status. All other callers must pass _check_key_admin_access.

Co-Authored-By: Claude Sonnet 4.6 (1M context) <noreply@anthropic.com>

* fix: block user-controlled api_base in RAG ingest vector_store options

A user-supplied api_base in ingest_options.vector_store caused the server
to forward its configured provider credentials (Gemini, OpenAI) to an
attacker-controlled endpoint via SSRF.

Add api_base to the blocked credential params set alongside api_key and
the existing credential fields.

Co-Authored-By: Claude Sonnet 4.6 (1M context) <noreply@anthropic.com>

* fix: restrict /utils/transform_request to PROXY_ADMIN and apply body safety check

Any authenticated internal_user could POST arbitrary provider config
(aws_sts_endpoint, api_base, etc.) to /utils/transform_request and have
the server forward its credentials to an attacker-controlled endpoint.

- Gate the endpoint on PROXY_ADMIN role (403 for all other roles)
- Call is_request_body_safe() to reject banned params even for admins
- Convert ValueError from safety check to HTTP 400

Co-Authored-By: Claude Sonnet 4.6 (1M context) <noreply@anthropic.com>

* fix: apply banned-param check to /utils/transform_request

Without is_request_body_safe(), any authenticated user could pass
aws_sts_endpoint, api_base, or aws_web_identity_token to
/utils/transform_request and have the server forward its configured
provider credentials to an attacker-controlled endpoint during SDK
credential resolution.

Applies the same banned-param blocklist already used by LLM endpoints.
Endpoint remains accessible to all authenticated users.

Co-Authored-By: Claude Sonnet 4.6 (1M context) <noreply@anthropic.com>

* fix: block SSRF via api_base in /prompts/test dotprompt YAML frontmatter

Any frontmatter key not in ["model","input","output"] flowed into
optional_params and was merged into the LLM call data dict, bypassing
is_request_body_safe. An attacker with any bearer key could set
api_base in YAML to redirect the outbound LLM request — including the
provider API key — to an attacker-controlled host.

Fix: call is_request_body_safe on the constructed data dict after
optional_params are merged, before invoking ProxyBaseLLMRequestProcessing.
ValueError from the banned-param check is surfaced as HTTP 400.

Co-Authored-By: Claude Sonnet 4.6 (1M context) <noreply@anthropic.com>

* Update litellm/proxy/rag_endpoints/endpoints.py

Co-authored-by: veria-ai[bot] <224490171+veria-ai[bot]@users.noreply.github.com>

* fix: coerce nested config strings before banned-param check

_NESTED_CONFIG_KEYS descent used isinstance(nested, dict) which silently
skipped litellm_embedding_config when delivered as a JSON string via
multipart/form-data. Banned params (api_base, aws_sts_endpoint, etc.)
nested inside the stringified value were invisible to is_request_body_safe.

_NESTED_METADATA_KEYS already used _coerce_metadata_to_dict which parses
JSON strings before checking. Apply the same coercion to _NESTED_CONFIG_KEYS.

Co-Authored-By: Claude Sonnet 4.6 (1M context) <noreply@anthropic.com>

* fix: replace substring match with prefix match in is_llm_api_route

mapped_pass_through_routes used `_llm_passthrough_route in route` (substring)
so any admin-only path whose URL contained a provider name (openai, anthropic,
azure, bedrock, etc.) was misclassified as an LLM API route and bypassed the
admin gate in non_proxy_admin_allowed_routes_check.

Confirmed live: non-admin key could GET /credentials/by_name/openai (read
masked provider API key) and DELETE /credentials/openai (delete credential).

Fix: use exact match or startswith(prefix + "/") — the same pattern used
everywhere else in RouteChecks — so only routes that actually start with a
passthrough prefix are allowed through.

Co-Authored-By: Claude Sonnet 4.6 (1M context) <noreply@anthropic.com>

* fix: stabilize PR #27878 test failures

- key_management_endpoints: extend can_skip_admin_check to team keys so
  team members with /key/update permission can update non-budget fields.
  can_team_member_execute_key_management_endpoint already validates team
  membership + permission and raises if unauthorized; reaching the admin
  check on a team key means the caller was authorized.

- test: set created_by on mock key in
  test_update_key_non_budget_fields_allowed_for_internal_user so
  caller_is_creator resolves correctly (MagicMock default ≠ user_id).

- auth_utils.get_request_route: guard against non-dict request.scope
  (e.g. MagicMock in unit tests) to prevent a MagicMock leaking into
  UserAPIKeyAuth.request_route and failing Pydantic validation.

- ci: assign test_multipart_bypass_repro.py to the proxy-runtime shard
  in test-unit-proxy-db.yml to satisfy the shard-coverage check.

Co-Authored-By: Claude Sonnet 4.6 (1M context) <noreply@anthropic.com>

* fix(lint): add explicit str() cast in get_request_route for MyPy

scope.get() returns Any|None which MyPy cannot coerce to str implicitly.
Wrap both scope.get() calls in str() to satisfy the type checker.

Co-Authored-By: Claude Sonnet 4.6 (1M context) <noreply@anthropic.com>

* fix: guard bare-/ root_path strip + make total_spend migration idempotent

auth_utils.get_request_route: when Starlette sets scope["app_root_path"]
to "/" (e.g. behind some middleware), the old stripping logic would
remove the leading slash from every path ("/team/new" → "team/new"),
breaking route matching and causing auth to misclassify protected routes.
Skip stripping when root_path is bare "/".

migration: add IF NOT EXISTS to total_spend ALTER TABLE so the migration
is safe to replay when a prior partial run already created the column.
Without this guard, prisma migrate deploy fails on CI DBs that were
partially migrated, causing all subsequent DB operations (including
/team/new) to 500.

Co-Authored-By: Claude Sonnet 4.6 (1M context) <noreply@anthropic.com>

* fix: require creator still owns key for personal-key bypass in /key/update

caller_is_creator now requires both created_by == caller AND user_id ==
caller. Previously checking only created_by let a demoted admin who
originally created a key for another user continue editing non-budget
fields on it after reassignment, bypassing _check_key_admin_access.

Adds regression test: creator whose key was reassigned is blocked (403).

Co-Authored-By: Claude Sonnet 4.6 (1M context) <noreply@anthropic.com>

* fix: extract auth checks to fix PLR0915 + broaden max_budget assertion

internal_user_endpoints._update_single_user_helper exceeded 50 statements
(PLR0915). Extract authorization checks into _check_user_update_authz helper
to bring statement count under the limit.

test_validate_max_budget: assert "negative" (substring of both the local
"cannot be negative" and the CI "non-negative finite number" messages) so
the test is stable regardless of which exact wording the function uses.

Co-Authored-By: Claude Sonnet 4.6 (1M context) <noreply@anthropic.com>

---------

Co-authored-by: Claude Sonnet 4.6 (1M context) <noreply@anthropic.com>
Co-authored-by: veria-ai[bot] <224490171+veria-ai[bot]@users.noreply.github.com>

* bump: version 0.4.71 → 0.4.72

* uv lock

* feat(mcp): support OAuth passthrough discovery

* fix(mcp): support OAuth browser auth

* fix(mcp): refine upstream OAuth metadata fallback

* feat(proxy): support issuer-scoped JWT auth

* fix(mcp): validate oauth callback redirect sink

* feat(proxy): support issuer-scoped JWT auth

* test(mcp): align trusted proxy fixtures

* style(mcp): satisfy black formatting

* chore(ui): bump next to 16.2.6

* fix(mcp): address oauth passthrough review findings

* test(mcp): split oauth passthrough regressions

* fix(interactions): align openapi response fields

* security: prevent forwarding litellm api keys to upstream mcp servers

- Strip Authorization header from extra_headers for pass-through servers
- Pass-through servers (auth_type=None with extra_headers: [Authorization])
  must not receive the user's LiteLLM API key
- Only OAuth2 M2M and pass-through servers skip Authorization header
- Other headers (x-request-id, x-trace-id) are still forwarded normally
- Fixes credential leakage / authentication bypass in MCP pass-through mode

* fix(interactions): remove steps field not in google openapi spec

The steps field was added but is not present in the current Google
Interactions OpenAPI specification. Revert to using only the fields
that are actually defined in the spec.

* fix(mcp): forward Authorization in pass-through when x-litellm-api-key is admission

Commit 3753970cc9 widened the Authorization strip to cover all
is_oauth_passthrough servers — protecting against the LiteLLM admission
key leaking upstream when the caller used Authorization for admission,
but also silently stripping legitimate upstream OAuth bearers when the
caller used x-litellm-api-key for admission.

That broke transparent OAuth pass-through (EAI-506 V5/V6): standards-
compliant MCP clients (OpenCode, Claude Code, mcp-inspector) complete
PKCE against the upstream IdP and send the resulting token as plain
Authorization: Bearer per the MCP spec — with the wider strip in place,
that token never reaches the upstream and tools/list returns empty.

Narrow the strip: skip Authorization for pass-through servers only when
the caller did NOT supply x-litellm-api-key. When x-litellm-api-key is
present, admission is unambiguous and Authorization is free to carry
the upstream OAuth bearer.

The original security guarantee is preserved — a client that sends only
Authorization (no x-litellm-api-key) still has it stripped, so the
LiteLLM key cannot leak upstream via that path.

Tests:
- new: forwards Authorization when x-litellm-api-key is present
- new: still strips Authorization when only Authorization is present
- existing pass-through + M2M tests unchanged

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

* fix(interactions): align status enum with openapi spec

* fix(mcp,jwt): address greptile review concerns

- Cache _get_agent_object_permission via user_api_key_cache (sentinel for
  no-permission rows) so MCP requests from agent keys don't hit the DB on
  every tool-list / tool-call.
- Re-raise HTTPException in handle_sse_mcp so 401 + WWW-Authenticate
  challenges (and other HTTP errors) propagate to SSE clients instead of
  being swallowed as 500.
- Normalise booleans in _validate_token_response so admin rules written as
  JSON-style "true" / "false" match upstream responses that return
  Python True / False.
- Treat configured JWT issuer claim mappings as advisory: when a mapped
  field is absent or empty, leave the normalised claim unset instead of
  raising, matching the global litellm_jwtauth path.

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

* test: replace dall-e-3 with gpt-image-1 in health check and router tests (#27813)

OpenAI returns 'The model dall-e-3 does not exist' for the test account,
breaking test_openai_img_gen_health_check and test_image_generation.
Switch to gpt-image-1, matching the existing TestOpenAIGPTImage1 pattern.

(cherry picked from commit aee58db880)

* fix(tests): drop dall-e-only test classes; route live image tests via gpt-image-1

Second wave of failures from the 2026-05-12 DALL-E shutdown:
- tests/image_gen_tests/test_image_edits.py::TestOpenAIImageEditDallE2
  and tests/image_gen_tests/test_image_generation.py::TestOpenAIDalle3
  are explicitly named for the deprecated models and can't pass; remove.
  gpt-image-1 coverage already exists in sibling classes.
- tests/local_testing/test_router.py image gen tests use dall-e-3 only
  as a routing example; swap to gpt-image-1.
- tests/local_testing/test_custom_callback_input.py image_generation
  success/failure paths swapped to gpt-image-1.

(cherry picked from commit 945b10ded4)

* test(fireworks): replace deprecated llama-v3p3-70b-instruct model

Fireworks removed llama-v3p3-70b-instruct from serverless, so every
live test using it now fails with NotFoundError ("Model not found,
inaccessible, and/or not deployed").

Swap the 6 references (3 files) to the currently-served
accounts/fireworks/models/deepseek-v3p1 — the canonical model in
Fireworks' current docs examples and present in LiteLLM's cost map.
test_get_model_params_fireworks_ai is a pure pricing-heuristic test
(no network) asserting the >16b branch, so it uses llama-v3p1-70b-
instruct instead to keep the "fireworks-ai-above-16b" assertion and
branch coverage intact.

(cherry picked from commit 39a1d438f2)

* test(fireworks): mock remaining live smoke tests

test_completion_fireworks_ai and test_completion_cost_fireworks_ai
made real Fireworks calls and broke whenever Fireworks rotated its
serverless catalog (no externally-verifiable model list exists).
They also asserted nothing — just printed.

Mock the HTTP post and assert real behavior instead: the request is
built with the right model/messages and the OpenAI-compatible
response parses back; the cost path yields a non-zero cost against
the local cost map. No network, no model dependency, stronger than
the old smoke checks.

(cherry picked from commit b5db7ed37d)

* fix(tests): replace shut-down gpt-4o-audio-preview with gpt-audio-1.5 (#28281)

* fix(tests): replace shut-down gpt-4o-audio-preview with gpt-audio-1.5

OpenAI shut down gpt-4o-audio-preview on 2026-05-07, so the live audio
calls in test_stream_chunk_builder_openai_audio_output_usage and
test_standard_logging_payload_audio now hard-fail with a model-not-found
error on every PR. The error was not "openai-internal", so the except
block swallowed it and execution fell through to an unbound
completion/response (UnboundLocalError).

Switch both tests to gpt-audio-1.5, OpenAI's recommended successor
(GA, not deprecated, already present in the litellm cost map so the
response_cost assertion still resolves). Also broaden the except to
skip with the real error in the reason instead of crashing, so a
transient upstream blip can't reintroduce the UnboundLocalError.

* fix(tests): narrow audio-test skip to model-not-found, re-raise the rest

Address review feedback: an unconditional skip on any exception would
silently mask a litellm-internal regression in the audio path (broken
param transformation, serialization, bad header) instead of failing CI.

Skip only on the upstream-unavailable class (model_not_found / "does not
exist" / openai-internal) and re-raise everything else, so genuine
regressions still fail loudly. The UnboundLocalError is still fixed
because the handler either skips or raises - it never falls through.

* fix(tests): add budget_exceeded to expected Interaction status enum

Staging added budget_exceeded to the Interaction OpenAPI status enum; the staging merge into this branch picked up the spec change but not the matching test update, so test_status_enum_values failed in CI. Align the test's expected list (exact-match by design) with the live spec.

* fix(tests): mock HTTP fetch in test_img_url_token_counter

The test parameterized a live third-party image URL (blog.purpureus.net) which now 404s, causing get_image_dimensions to fall through to its base64 decode path and crash with 'not enough values to unpack' on every PR run. Mock safe_get with a tiny 1x1 PNG so the URL branch is still exercised without any network dependency.

* fix(tests): swap gpt-4o-audio-preview to gpt-audio-1.5 in test_gpt4o_audio

OpenAI shut down gpt-4o-audio-preview on 2026-05-07, so both live tests in test_gpt4o_audio.py (test_audio_output_from_model and test_audio_input_to_model) hard-fail model_not_found on every PR. Swap the hardcoded model to OpenAI's successor gpt-audio-1.5 (same chat-completions audio surface; already in the litellm cost map). Mirror the narrowed-skip pattern from the prior audio fixes: skip on model_not_found / does-not-exist / openai-internal, re-raise everything else so genuine litellm regressions still fail CI loudly.

(cherry picked from commit 92de7423ef)

* fix(tests): migrate realtime + rerank tests off shut-down upstream models (#28191)

* fix(tests): use gpt-realtime in realtime guardrails test

OpenAI shut down gpt-4o-realtime-preview-2024-12-17 on 2026-05-07, so
the live OpenAI realtime guardrails integration test now fails with
model_not_found (session.created never arrives, _wait_for_event times
out). Point OPENAI_REALTIME_URL at the current GA model, gpt-realtime.

Scope limited to this test: the pricing-catalog JSON keeps the retired
entries intentionally (historical cost calc + separate Azure timeline),
and the Azure realtime cost-calc test is unaffected.

* fix(tests): mock nvidia_nim rerank instead of hitting EOL'd endpoint

NVIDIA reached end-of-life for the hosted nvidia/llama-3.2-nv-rerankqa-1b-v2
rerank API on 2026-05-18 with no published replacement, so the live
BaseLLMRerankTest.test_basic_rerank for nvidia_nim now returns HTTP 410
("Gone"). NVIDIA's hosted catalog rotates on a schedule, so swapping in
another live model would only defer the failure.

Override test_basic_rerank in TestNvidiaNim to mock the sync/async HTTP
transport (same pattern as test_nvidia_nim_rerank_ranking_endpoint in this
file) and inject a fake NVIDIA_NIM_API_KEY via monkeypatch. The
request/response transformation and cost calculation stay covered offline.
Scope limited to nvidia_nim; other BaseLLMRerankTest providers untouched.

* fix(tests): migrate remaining realtime tests off shut-down gpt-4o-realtime-preview

OpenAI's 2026-05-07 shutdown removed the entire gpt-4o-realtime-preview
family, including the undated 'gpt-4o-realtime-preview' alias (not just the
dated snapshot fixed earlier). Three live tests still connected with the
dead alias and failed with messages_received=1 (an error event instead of
session.created):

- test_openai_realtime_simple.py: get_model() -> gpt-realtime (drives
  TestOpenAIRealtime.test_realtime_connection / test_realtime_with_query_params)
- test_openai_realtime.py: test_openai_realtime_direct_call_no_intent and
  test_openai_realtime_direct_call_with_intent -> openai/gpt-realtime
  (the with_intent test shares the same dead alias even though it was not
  in the failing set this run)

Mocked unit tests (test_realtime_query_params_construction,
test_realtime_query_params_use_normalized_model_name) are left as-is: they
never hit the network and assert string plumbing only.

Also fixes test_text_message_blocked_by_guardrail_no_ai_response, which now
connects (the earlier URL swap worked) but tripped a model-wording-brittle
assertion. The guardrail flow asks the model to voice the block message
verbatim; gpt-4o-realtime-preview complied (output contained 'blocked'),
gpt-realtime refuses verbatim-repeat instructions ('I'm sorry, but I can't
repeat that message.'). Since the original user message is blocked before
it reaches OpenAI, the refusal is still a safe outcome. Assertion #3 now
accepts both voicing and refusal, and adds a hard check that the blocked
phrase never leaks into AI output.

(cherry picked from commit ce87c411bf)

* fix(model_prices): register mistral/ministral-8b-2512

Mistral's API now returns model='ministral-8b-2512' when 'mistral-tiny'
is requested, so test_completion_mistral_api fails with 'This model
isn't mapped yet'. Adding the entry so completion_cost can resolve the
cost for that response.

Author: Claude <noreply@anthropic.com>

* fix(mcp,auth): address greptile review concerns

- handle_sse_mcp now calls _raise_preemptive_401_for_unauthenticated_servers
  so SSE clients to pass-through OAuth MCP servers receive the RFC 9728
  401 + WWW-Authenticate challenge that the streamable-HTTP path already emits.
- get_request_route strips a trailing slash from root_path before length-based
  prefix removal so non-canonical ASGI root_path values like "/litellm/"
  don't strip the leading slash from the returned route.
- _mcp_oauth_user_api_key_auth's cookie JWT decode now passes
  options={"verify_aud": False} so a future revision of the UI session
  JWT containing an aud claim cannot silently downgrade the request to
  unauthenticated.

Co-authored-by: Claude <claude@anthropic.com>

* fix(tests): backfill local model_cost into remote-fetched map

litellm.model_cost is loaded at import time from LITELLM_MODEL_COST_MAP_URL
(pinned to main), so pricing entries that exist only in this branch (e.g.
mistral/ministral-8b-2512, freshly added because Mistral's API now returns
this id from mistral-tiny) are absent at test time and completion_cost
lookups raise 'This model isn't mapped yet'. Backfill the in-tree backup
into litellm.model_cost in the local_testing conftest so cassette-driven
cost calculations resolve against the entries that ship with the branch
under test.

Fixes local_testing_part1 failures on test_completion_mistral_api and
test_completion_mistral_api_modified_input.

* fix(mcp,jwt): address greptile concurrency and code-quality concerns

- _apply_issuer_claim_mappings now builds a new dict and reads from the
  original token, rather than mutating its input. The change is
  behaviour-preserving (caller passes a fresh jwt.decode result), but
  avoids the surprise-mutation pattern flagged by greptile.
- is_network_error uses isinstance(exc, httpx.TransportError) instead of
  matching type(exc).__name__ against a hand-maintained string set, so
  ReadError / WriteError / ProxyError / etc. are also treated as
  transport-level failures and surfaced as HTTP 502.
- fetch_upstream_oauth_protected_resource now coalesces concurrent
  discovery requests per (server_id, resource_url) through an
  asyncio.Lock so concurrent .well-known calls share a single upstream
  fetch + cache write.
- Drop the redundant 'if trusted_ranges:' branch in get_mcp_client_ip;
  it is always true on the path that reaches it (the prior 'if not
  trusted_ranges:' early-returns).

Co-authored-by: Claude <claude@anthropic.com>

* fix(jwt,mcp): fall back to global JWKS on unknown issuer; prune fetch locks

- handle_jwt._get_configured_issuer now returns None for tokens whose 'iss'
  is not in the configured issuers list, letting auth_jwt fall through to
  the legacy JWT_PUBLIC_KEY_URL path instead of hard-raising. This keeps
  existing tokens from non-configured IdPs working when an operator adds
  the new 'issuers' list to a live deployment.

- discoverable_endpoints._prune_oauth_metadata_cache now also prunes
  entries in _OAUTH_METADATA_FETCH_LOCKS whose cache entry has been
  evicted and whose lock isn't currently held, bounding the locks dict
  to match the cache it guards.

Co-authored-by: Claude <claude@anthropic.com>

* fix(mcp,auth): restore client_ip in oauth2 target check, drop from delegate check

The merge of staging into the PR branch (d42a66adb6) misplaced the
client_ip=client_ip kwarg: it landed inside _target_servers_delegate_auth_to_upstream
(which never accepted client_ip and isn't called with it), while the
sibling _target_servers_use_oauth2 has client_ip in its signature but
stopped passing it through to get_mcp_server_by_name. That left ruff
flagging F821 on the undefined name and lint failing.

Move client_ip back into _target_servers_use_oauth2's lookup (matching
the call site that already forwards IPAddressUtils.get_mcp_client_ip)
and drop it from _target_servers_delegate_auth_to_upstream so its body
matches its signature again.

* fix(mcp): respect client ip for delegated auth

* fix(auth): address remaining greptile style findings

- get_request_route: require root_path to match whole path segments before
  stripping, so '/apifoo' isn't truncated to 'foo' when root_path='/api'.
- get_mcp_client_ip: collapse the two trusted-proxy validation branches into
  a single is_request_from_trusted_proxy call so the return value drives
  control flow instead of being discarded for the side-effect warning.

Co-authored-by: Claude <claude@anthropic.com>

* fix(jwt): strip internal _litellm_* claims in global JWKS auth path

Prevents identity spoofing where a token signed by the global JWKS
could inject _litellm_jwt_issuer and other _litellm_* claims that
downstream getters trust. The issuer-scoped path already strips these
via _apply_issuer_claim_mappings; mirror that behavior for the global
fallback path.

Co-authored-by: Yassin Kortam <yassin@berri.ai>

* fix(mcp): surface MCPUpstreamAuthError as 401 in SSE/HTTP transport handlers

Both handle_sse_mcp and handle_streamable_http_mcp only caught
HTTPException to preserve 401 + WWW-Authenticate challenges, but
MCPUpstreamAuthError (raised when a pass-through server's upstream
rejects a bearer token mid-session) inherits from Exception. It was
falling through to the generic handler and surfacing as an opaque 500.

Mirror the REST endpoint behavior: translate MCPUpstreamAuthError into
an HTTPException(status_code=e.status_code) with the upstream
www-authenticate header so standards-compliant MCP clients trigger the
upstream OAuth flow.

Co-authored-by: Yassin Kortam <yassin@berri.ai>

* fix(mcp): add upstream auth pre-flight in SSE handler

Mirror handle_streamable_http_mcp by calling _check_passthrough_upstream_auth
after the cold-start 401 emitter so expired/invalid upstream tokens surface a
proper 401 + WWW-Authenticate challenge before the SSE session commits 200
headers, instead of letting list_tools silently return [] when the upstream
rejects the token.

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

* fix(mcp): tighten cold-start bypass against CSV paths + dedupe upstream auth probe

- Return None from _parse_mcp_server_names_from_path for CSV multi-server
  paths (/mcp/a,b). The regex previously truncated at the first comma and
  silently passed a single server name to the cold-start gate.
- Switch _is_mcp_passthrough_cold_start to all-targets semantics, matching
  _target_servers_use_oauth2: one non-passthrough target in a co-targeted
  set must not flip the anonymous-admission bypass open for the others.
- Drop the redundant HTTPStatusError block in _extract_upstream_auth_failure
  - any HTTPStatusError carries a .response, so the preceding generic block
  already handles 401/403 detection.

Co-authored-by: Yassin Kortam <yassin@berri.ai>

* fix(mcp,tests): sync stubs and cold-start assertions with delegate-check

The merge of base-branch _target_servers_delegate_auth_to_upstream
into process_mcp_request inserts an additional
get_mcp_server_by_name(name) lookup ahead of the cold-start path,
which breaks two test patterns:

1. lookup_by_name(name) side-effect stubs in
   TestMCPDelegateAuthToUpstream are called positionally by the
   delegate check, then again by the cold-start path with
   client_ip=... — raising TypeError: unexpected keyword argument
   'client_ip'. Accept **_kwargs to match the real signature.

2. TestMCPPassthroughColdStartAdmission assertions count the lookup
   exactly once with client_ip=..., but the delegate check now adds
   a positional-only call ahead of it. Switch assert_called_once_with
   to assert_any_call for the cold-start invocation, and assert
   client_ip was *not* passed for the aggregate /mcp test where
   cold-start must not fire.

Both updates align with CLAUDE.md guidance to keep monkeypatch stubs in
sync with the real signature when an optional parameter is added.

Co-authored-by: Claude <claude@anthropic.com>

* fix(mcp): correct passthrough probe 401 + slashed-name cold start parser

- _check_passthrough_upstream_auth now emits
  'Bearer resource_metadata="..."' pointing at the gateway's
  oauth-protected-resource well-known URL, mirroring the
  pre-emptive 401 path. Pass-through servers don't use the gateway
  as an authorization server, so the previous 'authorization_uri='
  challenge sent clients to the wrong metadata endpoint.

- _parse_mcp_server_names_from_path now accepts server names that
  contain a single slash (e.g. custom_solutions/user_123), mirroring
  MCPRequestHandler._extract_target_server_names_from_path. Without
  this, the cold-start bypass missed slashed-name servers and the
  generic admission error propagated instead of the spec-compliant
  401 challenge.

- _is_mcp_passthrough_cold_start drops the unused scope parameter
  from its signature.

Co-authored-by: Yassin Kortam <yassin@berri.ai>

* style(mcp): format discoverable endpoints

* refactor(mcp): dedupe MCPUpstreamAuthError->HTTPException + thread client_ip into delegate-auth gate

Co-authored-by: Yassin Kortam <yassin@berri.ai>

* fix(mcp): handle passthrough OAuth metadata and startup auth errors

- discoverable_endpoints: For pass-through MCP servers, when upstream
  oauth-protected-resource returns a non-200/non-dict response, raise
  HTTP 502 instead of falling through to default gateway metadata.
  Falling through would direct MCP clients at the gateway, which is
  not the authorization server for pass-through configs.

- mcp_server_manager: Wrap _get_tools_from_server in startup tool name
  mapping with try/except. Since _get_tools_from_server now re-raises
  MCPUpstreamAuthError, an upstream 401 from a pass-through server at
  startup (when no user token is present) would otherwise abort the
  loop and leave subsequent servers unmapped.

Co-authored-by: Yassin Kortam <yassin@berri.ai>

* fix(mcp): restrict passthrough probe challenge to OAuth passthrough servers

The probe filter previously matched any server with Authorization in
extra_headers, including gateway-managed OAuth2 servers. Those would
then receive the resource_metadata= WWW-Authenticate challenge meant
for pass-through servers, instead of the authorization_uri= challenge
pointing at the gateway AS metadata. Use srv.is_oauth_passthrough so
only genuine pass-through servers get the resource-metadata challenge.

Co-authored-by: Yassin Kortam <yassin@berri.ai>

* test(proxy): cover issuer-scoped JWT auth

* fix(mcp): use resource metadata for passthrough reauth

* fix(mcp,tests): assert cold-start helper directly for aggregate /mcp

Threading client_ip into _target_servers_delegate_auth_to_upstream
made get_mcp_server_by_name(name, client_ip=...) also fire from the
delegate-auth check, so the call_args_list assertion on
client_ip-in-kwargs no longer uniquely signals a cold-start lookup.
Patch _is_mcp_passthrough_cold_start and assert it is not invoked,
which is the actual contract the test is pinning.

* fix(mcp,jwt): drop unneeded async helper + suppress misleading unscoped JWT warning

- _build_oauth_authorization_server_response: revert to sync (no awaits in body).
  The function only does dict construction and synchronous registry lookups;
  async added coroutine creation overhead per discovery call without need.
- _build_decode_kwargs: accept has_issuer_config so the global path's
  'JWT auth is unscoped' warning is suppressed when LiteLLM_JWTAuth.issuers
  provides per-issuer scoping. Previously the warning fired spuriously for
  admins who intentionally use only the new issuers config.

* fix(jwt,mcp): clarify issuers fallthrough + add TTL on mcp permission cache

- LiteLLM_JWTAuth.issuers docs now state explicitly that unlisted
  issuers fall back to the global JWT_AUDIENCE/JWT_ISSUER path; the
  field is additive routing, not an allow-list. Matches actual
  control flow in handle_jwt.auth_jwt and the regression tests
  asserting backwards compatibility with the global JWKS path.
- MCPRequestHandler._get_{org,agent}_object_permission now pass
  ttl=DEFAULT_MANAGEMENT_OBJECT_IN_MEMORY_CACHE_TTL on async_set_cache,
  mirroring the auth_checks.py pattern so the cache TTL is explicit
  on both DualCache layers.

* fix(tests): align merged JWT and MCP cold-start assertions

Update the tests carried over from PR #28008 to match the assertions on
the staging branch:

- tests/test_litellm/proxy/auth/test_handle_jwt.py: unknown issuers now
  fall back to the legacy JWT_PUBLIC_KEY_URL path (per
  litellm_feat/v1.84.0-mcp-gateway-jwt-auth's
  '\''fall back to global JWKS on unknown issuer'\''), and mapped issuer
  claims that are absent no longer fail closed — they simply leave the
  normalised LiteLLM internal claim absent.

- tests/test_litellm/proxy/_experimental/mcp_server/auth/test_user_api_key_auth_mcp.py:
  the aggregate '\''/mcp'\'' route still triggers the delegate-auth-to-upstream
  lookup once for the header-supplied server name; cold-start admission
  must NOT fire on top of that. Tighten the assertion to
  assert_called_once_with so a future regression that re-enters cold-start
  is caught.

Co-authored-by: Mateo Wang <mateo-berri@users.noreply.github.com>

* fix(jwt): guard litellm_jwtauth access in auth_jwt global path

JWTHandler() can be constructed without update_environment() being
called (tests do this directly), in which case self.litellm_jwtauth
does not exist. Accessing it raises AttributeError before getattr can
fall back. Use the same safe pattern other call sites use.

* Gate MCP OAuth pass-through on delegate_auth_to_upstream flag

Sameer's review on #28356/#28008 flagged that the new pass-through
behaviors (preemptive 401 challenges, /.well-known/oauth-protected-
resource proxying, upstream 401/403 propagation as MCPUpstreamAuthError,
and Authorization-stripping when no x-litellm-api-key is supplied)
were implicitly enabled for every server with auth_type=none plus
Authorization in extra_headers. Existing users doing static bearer
pass-through for non-OAuth reasons would have silently regressed.

Make the detection rule explicit: extend the existing
delegate_auth_to_upstream flag (previously oauth2-only) to also gate
is_oauth_passthrough. Now requires flag + auth_type=None + Authorization
in extra_headers, per Sameer's suggested detection rule. The UI toggle
now appears for both modes (oauth2 PKCE passthrough and auth_type=none
OAuth pass-through) with mode-appropriate copy.

Update test fixtures to set the flag where the test intent is to
exercise OAuth pass-through behavior, and add negative tests covering
the new default-false case.

* fix(mcp): route org object_permission lookup through shared auth helpers

Replace the bespoke litellm_organizationtable.find_unique + dedicated
cache key in _get_org_object_permission with get_org_object +
get_object_permission so MCP requests share the same user_api_key_cache
entries as the rest of the proxy and no longer fragment org-row caching.

* fix(mcp): wrap get_object_permission call in shared try/except

Ensure exceptions from get_object_permission in _get_org_object_permission are caught and return None, preserving the original fail-safe semantics.

Co-authored-by: Yassin Kortam <yassin@berri.ai>

* fix(jwt): validate issuer audience at config load + dedicated key-miss exception

- Move JWTIssuerConfig audience-required guard into a Pydantic model_validator
  so misconfiguration fails at startup instead of on the first request.
- Replace the string-match `No matching public key found` filter in
  get_public_key's multi-URL fallback with a dedicated
  NoMatchingJWTPublicKeyError; only that specific exception triggers
  continuation, every other error still surfaces.

* fix(mcp): admit and forward Authorization for passthrough OAuth return

For pass-through MCP servers (auth_type=none with delegate_auth_to_upstream)
the RFC 9728 cold-start flow sends the client back with only
"Authorization: Bearer <upstream-token>" after upstream OAuth discovery.
Previously this path 1) was rejected in process_mcp_request because the
oauth2_headers fallback only covered auth_type=oauth2 targets, and 2) had
the Authorization header stripped by _prepare_mcp_server_headers when no
x-litellm-api-key was present, treating the upstream token as a potential
LiteLLM key leak.

- Extend the elif oauth2_headers fallback to also admit anonymously when
  every target is a pass-through server.
- Pass user_api_key_auth into _prepare_mcp_server_headers so it can
  forward Authorization for pass-through servers when admission did not
  consume the bearer as a LiteLLM key (api_key is unset).

Co-authored-by: Yassin Kortam <yassin@berri.ai>

* fix(mcp): consistent www-authenticate casing + SSE toolset scoping

- Normalize the WWW-Authenticate header key emitted by
  _check_passthrough_upstream_auth to lowercase to match the other 401
  emitters in the OAuth pass-through flow.
- Mirror the streamable HTTP handler's toolset scoping in handle_sse_mcp:
  strip client-supplied x-mcp-toolset-id and apply _apply_toolset_scope
  before _check_passthrough_upstream_auth so the upstream probe list is
  derived from the fully-authorized server set.
- Tighten _has_client_supplied_mcp_auth signature so
  mcp_server_auth_headers is Optional, matching its caller in
  process_mcp_request.

Co-authored-by: Yassin Kortam <yassin@berri.ai>

* security(mcp): strip Authorization in call_tool when LiteLLM admission used legacy header

Mirror the OAuth pass-through admission check from _prepare_mcp_server_headers
(list-tools path) in _call_regular_mcp_tool (tool-call path): when the server
is OAuth pass-through and the caller did not supply x-litellm-api-key,
Authorization on the inbound request may itself be the LiteLLM API key — so
strip it before forwarding instead of leaking the gateway credential upstream.

When x-litellm-api-key is present, admission is unambiguous and Authorization
continues to carry the upstream OAuth bearer (transparent pass-through).

* refactor(mcp): centralize caller Authorization strip decision

Extracted the security-sensitive logic that decides whether the caller's
Authorization header is forwarded to (or stripped from) an outgoing MCP
request into a single helper, _should_strip_caller_authorization, in
mcp_server_manager.py.

Previously the same condition was duplicated across
_call_regular_mcp_tool (mcp_server_manager.py) and
_prepare_mcp_server_headers (server.py). Keeping two copies of this
check risked future divergence and credential-leak / broken-passthrough
bugs. Both call sites now share the helper, preserving exact behavior.

Co-authored-by: Yassin Kortam <yassin@berri.ai>

* log MCP OAuth discovery diagnostics for unmatched paths and non-transport upstream errors

* fix(jwt): include issuer-normalized team id in get_all_jwt_team_ids

The aggregator for team IDs only consulted the issuer-normalized claim
for the plural (team_ids) path and fell back to the global config for
the singular path. When an operator configures team_id_jwt_field only
at the issuer level, get_team_id correctly returned the mapped value
but get_all_jwt_team_ids silently dropped it, causing membership
reconciliation to disagree with request routing.

Co-authored-by: Yassin Kortam <yassin@berri.ai>

* fix(mcp/jwt): dedupe cold-start path parser; reject conflicting audience flags

- _parse_mcp_server_names_from_path now delegates to
  MCPRequestHandler._extract_target_server_names_from_path so the
  names used by the cold-start passthrough bypass cannot drift from the
  names used by downstream routing.
- JWTIssuerConfig now rejects the combination of audience and
  disable_audience_validation=True at validation time instead of
  silently ignoring the flag.

* fix(mcp): restrict passthrough cold-start bypass to 401 only

The new elif passthrough cold-start branch reused is_auth_error which
matches both 401 and 403. A 403 from user_api_key_auth indicates the
LiteLLM key WAS recognized but is forbidden (e.g. over budget / rate
limited); falling through to anonymous UserAPIKeyAuth() in that case
bypasses spend and rate-limit controls on passthrough servers.

Only trigger the cold-start anonymous admission on 401, which is the
signal that the bearer is an upstream OAuth token rather than a
recognized LiteLLM key.

Co-authored-by: Yassin Kortam <yassin@berri.ai>

* fix(jwt/mcp): warn on unscoped JWT fallback; route agent permission lookup through shared helper

- _build_decode_kwargs no longer suppresses the unscoped-fallback warning
  when LiteLLM_JWTAuth.issuers is set: tokens whose iss does not match
  any configured issuer still fall through to the global path, and that
  fallback is itself unscoped when JWT_AUDIENCE/JWT_ISSUER are absent.

- _get_agent_object_permission now caches the agent_id ->
  object_permission_id mapping and delegates the permission lookup to
  the shared get_object_permission helper, so the agent path reuses the
  same cache entries as the org / team / key paths.

* fix(mcp): fabricate resource_metadata challenge when upstream 401 omits WWW-Authenticate

When an upstream pass-through MCP server returns 401 without a
WWW-Authenticate header (non-compliant per RFC 7235 §3.1),
to_http_exception() now produces a synthetic Bearer challenge pointing
at the gateway's standard-pattern oauth-protected-resource well-known
endpoint for that server. This keeps MCP clients on the RFC 9728
discovery flow instead of receiving a bare 401 with no recovery hint.

* fix(jwt): make _get_decode_options explicitly control verify_iss

Previously, _get_decode_options only set verify_aud based on whether
audience was provided. The issuer JWT path relied on always passing
issuer=issuer_config.issuer to trigger PyJWT's default verify_iss=True,
making the helper's behavior implicitly dependent on caller behavior.

Now _get_decode_options accepts issuer as well, mirroring the verify_aud
handling and matching the dimensions handled by _build_decode_kwargs.

Co-authored-by: Yassin Kortam <yassin@berri.ai>

* fix(mcp): emit absolute resource_metadata URI in fabricated 401 challenge

Per RFC 9728 §3.2 the resource_metadata Bearer challenge must be an
absolute URI; strict MCP clients reject relative URIs and fail to
initiate discovery. MCPUpstreamAuthError.to_http_exception now accepts
the gateway base URL and prepends it when the upstream omitted
WWW-Authenticate, and all four call sites (streamable HTTP, SSE, and
the two REST tool-list paths) supply it.

* fix(mcp): correct 403 detail text and remove dead _list_tools_for_single_server duplicate

- MCPUpstreamAuthError.to_http_exception() now returns detail='Forbidden' for
  403 upstream responses (and 'Unauthorized' for 401), matching the
  _check_passthrough_upstream_auth pre-flight probe.
- Remove the shadowed first definition of _list_tools_for_single_server in
  rest_endpoints.py; the second definition was the live one and the dead copy
  was a maintenance trap.

Co-authored-by: Yassin Kortam <yassin@berri.ai>

* fix: address potential bugs in auth_utils, mcp discoverable endpoints, and mcp auth

- auth_utils.get_request_route: return '/' instead of empty string when
  raw_path exactly equals root_path so downstream route allowlist checks
  still see a leading slash
- discoverable_endpoints.fetch_upstream_oauth_protected_resource: also
  cache negative results (no upstream metadata) for a shorter TTL so we
  don't re-fetch on every discovery request and so the per-key fetch
  lock can be pruned
- user_api_key_auth_mcp: guard the oauth2_headers 401 cold-start
  passthrough bypass with _has_client_supplied_mcp_auth, matching the
  parallel bypass in the no-Authorization branch so MCP-auth-bearing
  requests don't silently downgrade to anonymous admission

Co-authored-by: Yassin Kortam <yassin@berri.ai>

* test(vertex): tolerate transient InternalServerError in google maps tool test

test_gemini_google_maps_tool_simple makes live calls to Vertex AI's Google
Maps grounding backend, which intermittently returns 500 INTERNAL ("Please
retry") — a transient upstream failure, not a LiteLLM bug. The test already
passes on RateLimitError; treat InternalServerError the same way so transient
Vertex-side failures don't fail CI.

* refactor(mcp): drop redundant has_client_credentials filter on passthrough probe

is_oauth_passthrough already requires auth_type in (None, MCPAuth.none),
which is mutually exclusive with has_client_credentials (auth_type ==
MCPAuth.oauth2), so the extra guard was always True and only added
confusion about whether a server could be both passthrough and M2M.

Co-authored-by: Yassin Kortam <yassin@berri.ai>

* fix: restore unreachable InternalServerError skip handler in vertex test

Co-authored-by: Yassin Kortam <yassin@berri.ai>

* feat(mcp): add dedicated oauth_passthrough flag for non-oauth2 pass-through

Previously is_oauth_passthrough reused delegate_auth_to_upstream — a flag
scoped to oauth2 servers (PKCE bypass) — to gate OAuth pass-through for
auth_type=none servers. Overloading it risked regressing existing
deployments that set delegate_auth_to_upstream, since the same flag would
silently start driving pass-through (discovery proxying, 401 challenges,
upstream 401/403 propagation) on non-oauth2 servers.

Introduce a separate oauth_passthrough opt-in so the two behaviors never
imply each other:
- MCPServer.is_oauth_passthrough now requires oauth_passthrough (not
  delegate_auth_to_upstream).
- Persist oauth_passthrough on LiteLLM_MCPServerTable (new column +
  migration) and wire it through config/DB load and API responses.
- UI splits the single toggle into two: "Delegate auth to upstream (PKCE
  passthrough)" for oauth2 and "OAuth pass-through" for auth_type=none
  servers forwarding Authorization.

Adds backend tests (property, round-trip, and a regression guard that
delegate_auth_to_upstream alone never enables pass-through) and UI tests
for the toggle split.

* fix(mcp): reconcile cold-start bypass with x-mcp-servers header and skip non-absolute WWW-Authenticate fabrication

- _parse_mcp_server_names_from_path now fails closed when the
  x-mcp-servers header introduces any target not present in the
  path-derived target set, closing a header/path mismatch where the
  cold-start passthrough bypass could otherwise admit anonymously
  while the header advertises a non-passthrough server.
- MCPUpstreamAuthError.to_http_exception no longer emits a relative
  resource_metadata URI when base_url is missing; per RFC 9728 3.2
  the URI must be absolute, so we skip fabrication entirely rather
  than send a challenge strict MCP clients will reject.

Co-authored-by: Yassin Kortam <yassin@berri.ai>

* fix(mcp): fabricate path-aware resource_metadata URI for upstream 401

When MCPUpstreamAuthError.to_http_exception fabricates a
`WWW-Authenticate: Bearer resource_metadata=...` challenge (because
the upstream 401 omitted one), the URL now matches the inbound MCP
transport pattern the client originally used:

  - /mcp/{server_name}      -> /.well-known/oauth-protected-resource/mcp/{server_name}
  - /{server_name}/mcp      -> /.well-known/oauth-protected-resource/{server_name}/mcp

This mirrors the path-aware behaviour of
_get_passthrough_resource_metadata_url in server.py so strict
RFC 9728 \xA73.2 clients on legacy routes get a resource_metadata URI
aligned with the resource pattern they originally targeted.

Co-authored-by: Yassin Kortam <yassin@berri.ai>

* fix(jwt+mcp): tighten issuer-scoped claim type handling, RFC-quote authorization_uri, surface MCP upstream auth errors, defense-in-depth on decode options

- handle_jwt: when an issuer-scoped _litellm_team_ids claim exists but
  has an unexpected type, return [] instead of falling through to the
  global team_ids_jwt_field path (different claim semantically).
- handle_jwt: _get_decode_options/_decode_jwt_with_public_key now take
  an explicit disable_audience_validation flag; passing audience=None
  without it raises, so audience checks can't silently disappear if the
  model validator is ever bypassed. _auth_jwt_with_issuer forwards the
  flag from JWTIssuerConfig.
- mcp_server: quote the authorization_uri WWW-Authenticate parameter
  value (RFC 6750 / 9728 auth-param must be quoted-string), matching
  the pass-through path.
- mcp_server: in _fetch_and_filter_server_tools, re-raise
  MCPUpstreamAuthError so the outer streamable-HTTP handler can surface
  a proper 401 + WWW-Authenticate challenge instead of returning an
  empty tool list.

Co-authored-by: Yassin Kortam <yassin@berri.ai>

* chore(docker): align Dockerfile.non_root/Dockerfile.database to current wolfi-base SHA

The older sha256:3258be... pin has been intermittently returning 500/not-found
from cgr.dev, breaking the test-server-root-path GitHub Action and the
build_docker_database_image CircleCI job. Move both Dockerfiles onto the
same sha256:31da65... digest already in use by Dockerfile, gateway/Dockerfile,
backend/Dockerfile, and migrations/Dockerfile so the base image is consistent
across the repo.

* ci(docker): bump wolfi-base pin to current working digest

The previously aligned sha256:31da6565f35a... and the older sha256:3258be...
both return HTTP 500 from cgr.dev's manifest endpoint, breaking the
build_docker_database_image CircleCI job and test-server-root-path GitHub
Action. The current 'latest' tag resolves to sha256:5743937d521c... which
serves manifests normally, so move docker/Dockerfile.database and
docker/Dockerfile.non_root onto that digest.

* ci(docker): retry apk add in Dockerfile.database for apk.cgr.dev flakes

Mirror the retry-loop pattern from #28888 (which fixed backend/Dockerfile,
gateway/Dockerfile, and migrations/Dockerfile) into docker/Dockerfile.database.
The build_docker_database_image CI job has been intermittently failing with
"remote server returned error (try 'apk update')" when apk.cgr.dev flakes
mid-fetch; bumping the wolfi-base SHA doesn't address the mirror, only a
retry does.

Same explicit-failure form as #28888: exit non-zero on the 3rd miss instead
of silently succeeding because `sleep 5` was the last command in the
`&& break || sleep 5` chain.

* fix(mcp): scope preemptive 401 to toolset-narrowed server set

Move _raise_preemptive_401_for_unauthenticated_servers after toolset
scoping in both the StreamableHTTP and SSE handlers, and add an
optional allowed_server_ids parameter so passthrough/oauth2 servers
that the active toolset excludes no longer trigger a spurious 401
challenge. Without this, a client targeting a toolset whose scope
excludes a passthrough server could be pushed into an OAuth flow for
a server it would be 403'd on immediately after authentication.

Co-authored-by: Yassin Kortam <yassin@berri.ai>

* revert(docker): drop unrelated Wolfi bump and apk retry loop from MCP/JWT PR

These Docker changes are out of scope for the MCP OAuth passthrough + JWT
auth work and duplicate the build-reliability fix already merged to
litellm_internal_staging in #28888, which adds the same apk retry loop on
the componentized backend/gateway/migrations Dockerfiles and also fixes the
underlying nodeenv/libatomic root cause. Restoring docker/Dockerfile.database
and docker/Dockerfile.non_root to the base so this PR is purely the MCP/JWT
change.

* fix(mcp): surface upstream 403 challenges from REST tools/list

The single-server pass-through path converted an upstream MCPUpstreamAuthError
into an HTTPException, but list_tool_rest_api only re-raised 401s; an upstream
403 (valid token, insufficient scope) collapsed into a 200 response with
error=unexpected_error, so clients never saw the status or WWW-Authenticate
challenge needed to refresh scopes. Let MCPUpstreamAuthError propagate and
convert it once in list_tool_rest_api so both 401 and 403 reach the client,
while internal access/IP 403s keep the legacy error-dict shape.

* fix(mcp): fail closed for IP access control when XFF trusted ranges unset

When use_x_forwarded_for is enabled but mcp_trusted_proxy_ranges is not
configured, get_mcp_client_ip previously fell back to the direct peer IP.
Behind an internal reverse proxy that peer is the proxy's private address,
so every external caller was classified as internal and could reach MCP
servers with available_on_public_internet=false. Return an empty string in
that case so is_internal_ip treats the caller as external.

---------

Co-authored-by: Yuneng Jiang <yuneng@berri.ai>
Co-authored-by: Milan <milan@berri.ai>
Co-authored-by: Cursor <cursoragent@cursor.com>
Co-authored-by: Mateo Wang <mateo-berri@users.noreply.github.com>
Co-authored-by: Krrish Dholakia <krrish+github@berri.ai>
Co-authored-by: ryan-crabbe-berri <ryan@berri.ai>
Co-authored-by: stuxf <70670632+stuxf@users.noreply.github.com>
Co-authored-by: Claude Sonnet 4.6 (1M context) <noreply@anthropic.com>
Co-authored-by: veria-ai[bot] <224490171+veria-ai[bot]@users.noreply.github.com>
Co-authored-by: gym-cmd <186399764+gym-cmd@users.noreply.github.com>
Co-authored-by: Artem Dudarev <artem.dudarev@justeattakeaway.com>
Co-authored-by: Claude <claude@anthropic.com>
Co-authored-by: Yassin Kortam <yassin@berri.ai>
2026-06-02 12:22:04 -07:00

1380 lines
53 KiB
Plaintext
Raw Blame History

This file contains ambiguous Unicode characters

This file contains Unicode characters that might be confused with other characters. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.

datasource client {
provider = "postgresql"
url = env("DATABASE_URL")
}
generator client {
provider = "prisma-client-py"
binaryTargets = ["native", "debian-openssl-1.1.x", "debian-openssl-3.0.x", "linux-musl", "linux-musl-openssl-3.0.x"]
}
// Budget / Rate Limits for an org
model LiteLLM_BudgetTable {
budget_id String @id @default(uuid())
max_budget Float?
soft_budget Float?
max_parallel_requests Int?
tpm_limit BigInt?
rpm_limit BigInt?
model_max_budget Json?
budget_duration String?
budget_reset_at DateTime?
allowed_models String[] @default([]) // per-member model scope; empty = inherit team models
created_at DateTime @default(now()) @map("created_at")
created_by String
updated_at DateTime @default(now()) @updatedAt @map("updated_at")
updated_by String
organization LiteLLM_OrganizationTable[] // multiple orgs can have the same budget
projects LiteLLM_ProjectTable[] // multiple projects can have the same budget
keys LiteLLM_VerificationToken[] // multiple keys can have the same budget
end_users LiteLLM_EndUserTable[] // multiple end-users can have the same budget
tags LiteLLM_TagTable[] // multiple tags can have the same budget
team_membership LiteLLM_TeamMembership[] // budgets of Users within a Team
organization_membership LiteLLM_OrganizationMembership[] // budgets of Users within a Organization
}
// Models on proxy
model LiteLLM_CredentialsTable {
credential_id String @id @default(uuid())
credential_name String @unique
credential_values Json
credential_info Json?
created_at DateTime @default(now()) @map("created_at")
created_by String
updated_at DateTime @default(now()) @updatedAt @map("updated_at")
updated_by String
}
// Models on proxy
model LiteLLM_ProxyModelTable {
model_id String @id @default(uuid())
model_name String
litellm_params Json
model_info Json?
blocked Boolean @default(false)
created_at DateTime @default(now()) @map("created_at")
created_by String
updated_at DateTime @default(now()) @updatedAt @map("updated_at")
updated_by String
}
// Agents on proxy
model LiteLLM_AgentsTable {
agent_id String @id @default(uuid())
agent_name String @unique
litellm_params Json?
agent_card_params Json
static_headers Json? @default("{}")
extra_headers String[] @default([])
agent_access_groups String[] @default([])
object_permission_id String?
object_permission LiteLLM_ObjectPermissionTable? @relation(fields: [object_permission_id], references: [object_permission_id])
spend Float @default(0.0)
tpm_limit Int?
rpm_limit Int?
session_tpm_limit Int?
session_rpm_limit Int?
created_at DateTime @default(now()) @map("created_at")
created_by String
updated_at DateTime @default(now()) @updatedAt @map("updated_at")
updated_by String
}
model LiteLLM_OrganizationTable {
organization_id String @id @default(uuid())
organization_alias String
budget_id String
metadata Json @default("{}")
models String[]
spend Float @default(0.0)
model_spend Json @default("{}")
object_permission_id String?
created_at DateTime @default(now()) @map("created_at")
created_by String
updated_at DateTime @default(now()) @updatedAt @map("updated_at")
updated_by String
litellm_budget_table LiteLLM_BudgetTable? @relation(fields: [budget_id], references: [budget_id])
teams LiteLLM_TeamTable[]
users LiteLLM_UserTable[]
keys LiteLLM_VerificationToken[]
members LiteLLM_OrganizationMembership[] @relation("OrganizationToMembership")
object_permission LiteLLM_ObjectPermissionTable? @relation(fields: [object_permission_id], references: [object_permission_id])
}
// Model info for teams, just has model aliases for now.
model LiteLLM_ModelTable {
id Int @id @default(autoincrement())
model_aliases Json? @map("aliases")
created_at DateTime @default(now()) @map("created_at")
created_by String
updated_at DateTime @default(now()) @updatedAt @map("updated_at")
updated_by String
team LiteLLM_TeamTable?
}
// Assign prod keys to groups, not individuals
model LiteLLM_TeamTable {
team_id String @id @default(uuid())
team_alias String?
organization_id String?
object_permission_id String?
admins String[]
members String[]
members_with_roles Json @default("{}")
metadata Json @default("{}")
max_budget Float?
soft_budget Float?
spend Float @default(0.0)
models String[]
max_parallel_requests Int?
tpm_limit BigInt?
rpm_limit BigInt?
budget_duration String?
budget_reset_at DateTime?
blocked Boolean @default(false)
created_at DateTime @default(now()) @map("created_at")
updated_at DateTime @default(now()) @updatedAt @map("updated_at")
model_spend Json @default("{}")
model_max_budget Json @default("{}")
router_settings Json? @default("{}")
team_member_permissions String[] @default([])
access_group_ids String[] @default([])
policies String[] @default([])
default_team_member_models String[] @default([]) // default allowed_models for newly added team members; empty = no per-member restriction
budget_limits Json? // per-model budget limits for the team
model_id Int? @unique // id for LiteLLM_ModelTable -> stores team-level model aliases
allow_team_guardrail_config Boolean @default(false) // if true, team admin can configure guardrails for this team
litellm_organization_table LiteLLM_OrganizationTable? @relation(fields: [organization_id], references: [organization_id])
litellm_model_table LiteLLM_ModelTable? @relation(fields: [model_id], references: [id])
object_permission LiteLLM_ObjectPermissionTable? @relation(fields: [object_permission_id], references: [object_permission_id])
projects LiteLLM_ProjectTable[]
@@index([organization_id])
@@index([team_alias])
@@index([created_at])
}
// Projects sit between teams and keys for use-case management
model LiteLLM_ProjectTable {
project_id String @id @default(uuid())
project_alias String?
description String?
team_id String?
budget_id String?
metadata Json @default("{}")
models String[]
spend Float @default(0.0)
model_spend Json @default("{}")
model_rpm_limit Json @default("{}")
model_tpm_limit Json @default("{}")
blocked Boolean @default(false)
object_permission_id String?
created_at DateTime @default(now()) @map("created_at")
created_by String
updated_at DateTime @default(now()) @updatedAt @map("updated_at")
updated_by String
// Relations
litellm_team_table LiteLLM_TeamTable? @relation(fields: [team_id], references: [team_id])
litellm_budget_table LiteLLM_BudgetTable? @relation(fields: [budget_id], references: [budget_id])
keys LiteLLM_VerificationToken[]
object_permission LiteLLM_ObjectPermissionTable? @relation(fields: [object_permission_id], references: [object_permission_id])
}
// Audit table for deleted teams - preserves spend and team information for historical tracking
model LiteLLM_DeletedTeamTable {
id String @id @default(uuid())
team_id String // Original team_id
team_alias String?
organization_id String?
object_permission_id String?
admins String[]
members String[]
members_with_roles Json @default("{}")
metadata Json @default("{}")
max_budget Float?
soft_budget Float?
spend Float @default(0.0)
models String[]
max_parallel_requests Int?
tpm_limit BigInt?
rpm_limit BigInt?
budget_duration String?
budget_reset_at DateTime?
blocked Boolean @default(false)
model_spend Json @default("{}")
model_max_budget Json @default("{}")
router_settings Json? @default("{}")
team_member_permissions String[] @default([])
access_group_ids String[] @default([])
policies String[] @default([])
model_id Int? // id for LiteLLM_ModelTable -> stores team-level model aliases
allow_team_guardrail_config Boolean @default(false)
// Original timestamps from team creation/updates
created_at DateTime? @map("created_at")
updated_at DateTime? @map("updated_at")
// Deletion metadata
deleted_at DateTime @default(now()) @map("deleted_at")
deleted_by String? @map("deleted_by") // User who deleted the team
deleted_by_api_key String? @map("deleted_by_api_key") // API key hash that performed the deletion
litellm_changed_by String? @map("litellm_changed_by") // From litellm-changed-by header if provided
@@index([team_id])
@@index([deleted_at])
@@index([organization_id])
@@index([team_alias])
@@index([created_at])
}
// Track spend, rate limit, budget Users
model LiteLLM_UserTable {
user_id String @id
user_alias String?
team_id String?
sso_user_id String? @unique
organization_id String?
object_permission_id String?
password String?
teams String[] @default([])
user_role String?
max_budget Float?
spend Float @default(0.0)
user_email String?
models String[]
metadata Json @default("{}")
max_parallel_requests Int?
tpm_limit BigInt?
rpm_limit BigInt?
budget_duration String?
budget_reset_at DateTime?
allowed_cache_controls String[] @default([])
policies String[] @default([])
model_spend Json @default("{}")
model_max_budget Json @default("{}")
created_at DateTime? @default(now()) @map("created_at")
updated_at DateTime? @default(now()) @updatedAt @map("updated_at")
// relations
litellm_organization_table LiteLLM_OrganizationTable? @relation(fields: [organization_id], references: [organization_id])
organization_memberships LiteLLM_OrganizationMembership[]
invitations_created LiteLLM_InvitationLink[] @relation("CreatedBy")
invitations_updated LiteLLM_InvitationLink[] @relation("UpdatedBy")
invitations_user LiteLLM_InvitationLink[] @relation("UserId")
object_permission LiteLLM_ObjectPermissionTable? @relation(fields: [object_permission_id], references: [object_permission_id])
}
model LiteLLM_ObjectPermissionTable {
object_permission_id String @id @default(uuid())
mcp_servers String[] @default([])
mcp_access_groups String[] @default([])
mcp_tool_permissions Json? // Tool-level permissions for MCP servers. Format: {"server_id": ["tool_name_1", "tool_name_2"]}
vector_stores String[] @default([])
agents String[] @default([])
agent_access_groups String[] @default([])
models String[] @default([])
blocked_tools String[] @default([]) // Tool names blocked for any key/team/user with this permission
mcp_toolsets String[] @default([]) // Toolset IDs granted to this key/team/user
search_tools String[] @default([]) // search_tool_name values this key/team/user may call
teams LiteLLM_TeamTable[]
projects LiteLLM_ProjectTable[]
verification_tokens LiteLLM_VerificationToken[]
organizations LiteLLM_OrganizationTable[]
users LiteLLM_UserTable[]
end_users LiteLLM_EndUserTable[]
agents_table LiteLLM_AgentsTable[]
}
// Holds the MCP server configuration
model LiteLLM_MCPServerTable {
server_id String @id @default(uuid())
server_name String?
alias String?
description String?
instructions String?
url String?
spec_path String?
transport String @default("sse")
auth_type String?
credentials Json? @default("{}")
created_at DateTime? @default(now()) @map("created_at")
created_by String?
updated_at DateTime? @default(now()) @updatedAt @map("updated_at")
updated_by String?
mcp_info Json? @default("{}")
mcp_access_groups String[]
allowed_tools String[] @default([])
tool_name_to_display_name Json? @default("{}")
tool_name_to_description Json? @default("{}")
extra_headers String[] @default([])
static_headers Json? @default("{}")
// Health check status
status String? @default("unknown")
last_health_check DateTime?
health_check_error String?
// Stdio-specific fields
command String?
args String[] @default([])
env Json? @default("{}")
authorization_url String?
token_url String?
registration_url String?
allow_all_keys Boolean @default(false)
available_on_public_internet Boolean @default(true)
delegate_auth_to_upstream Boolean @default(false)
oauth_passthrough Boolean @default(false)
is_byok Boolean @default(false)
byok_description String[] @default([])
byok_api_key_help_url String?
source_url String?
// BYOM submission lifecycle
approval_status String? @default("active")
submitted_by String?
submitted_at DateTime?
reviewed_at DateTime?
review_notes String?
@@index([approval_status])
}
// Named collection of {server_id, tool_name} pairs that can be granted to keys/teams
model LiteLLM_MCPToolsetTable {
toolset_id String @id @default(uuid())
toolset_name String @unique
description String?
tools Json @default("[]") // [{server_id: string, tool_name: string}]
created_at DateTime @default(now())
created_by String?
updated_at DateTime @default(now()) @updatedAt
updated_by String?
}
// Per-user BYOK credentials for MCP servers
model LiteLLM_MCPUserCredentials {
id String @id @default(uuid())
user_id String
server_id String
credential_b64 String
created_at DateTime @default(now()) @map("created_at")
updated_at DateTime @default(now()) @updatedAt @map("updated_at")
@@unique([user_id, server_id])
}
// Generate Tokens for Proxy
model LiteLLM_VerificationToken {
token String @id
key_name String?
key_alias String?
soft_budget_cooldown Boolean @default(false) // key-level state on if budget alerts need to be cooled down
spend Float @default(0.0)
expires DateTime?
models String[]
aliases Json @default("{}")
config Json @default("{}")
router_settings Json? @default("{}")
user_id String?
team_id String?
agent_id String?
project_id String?
permissions Json @default("{}")
max_parallel_requests Int?
metadata Json @default("{}")
blocked Boolean?
tpm_limit BigInt?
rpm_limit BigInt?
max_budget Float?
budget_duration String?
budget_reset_at DateTime?
allowed_cache_controls String[] @default([])
allowed_routes String[] @default([])
policies String[] @default([])
access_group_ids String[] @default([])
model_spend Json @default("{}")
model_max_budget Json @default("{}")
budget_id String?
organization_id String?
object_permission_id String?
created_at DateTime? @default(now()) @map("created_at")
created_by String?
updated_at DateTime? @default(now()) @updatedAt @map("updated_at")
updated_by String?
last_active DateTime? // When this key was last used
rotation_count Int? @default(0) // Number of times key has been rotated
auto_rotate Boolean? @default(false) // Whether this key should be auto-rotated
rotation_interval String? // How often to rotate (e.g., "30d", "90d")
last_rotation_at DateTime? // When this key was last rotated
key_rotation_at DateTime? // When this key should next be rotated
budget_limits Json? // per-model budget limits for the key
litellm_budget_table LiteLLM_BudgetTable? @relation(fields: [budget_id], references: [budget_id])
litellm_organization_table LiteLLM_OrganizationTable? @relation(fields: [organization_id], references: [organization_id])
litellm_project_table LiteLLM_ProjectTable? @relation(fields: [project_id], references: [project_id])
object_permission LiteLLM_ObjectPermissionTable? @relation(fields: [object_permission_id], references: [object_permission_id])
jwt_key_mappings LiteLLM_JWTKeyMapping[]
// SELECT COUNT(*) FROM (SELECT "public"."LiteLLM_VerificationToken"."token" FROM "public"."LiteLLM_VerificationToken" WHERE ("public"."LiteLLM_VerificationToken"."user_id" = $1 AND ("public"."LiteLLM_VerificationToken"."team_id" IS NULL OR "public"."LiteLLM_VerificationToken"."team_id" <> $2)) OFFSET $3 ) AS "sub"
// SELECT ... FROM "public"."LiteLLM_VerificationToken" WHERE "public"."LiteLLM_VerificationToken"."user_id" = $1 OFFSET $2
@@index([user_id, team_id])
// SELECT ... FROM "public"."LiteLLM_VerificationToken" WHERE "public"."LiteLLM_VerificationToken"."team_id" = $1 OFFSET $2
@@index([team_id])
// SELECT ... FROM "public"."LiteLLM_VerificationToken" WHERE (("public"."LiteLLM_VerificationToken"."expires" IS NULL OR "public"."LiteLLM_VerificationToken"."expires" > $1) AND "public"."LiteLLM_VerificationToken"."budget_reset_at" < $2) OFFSET $3
@@index([budget_reset_at, expires])
}
model LiteLLM_JWTKeyMapping {
id String @id @default(uuid())
jwt_claim_name String // e.g. "sub", "email"
jwt_claim_value String // The claim value to match
token String // Hashed virtual key (FK)
description String?
is_active Boolean @default(true)
created_at DateTime @default(now())
created_by String?
updated_at DateTime @default(now()) @updatedAt
updated_by String?
litellm_verification_token LiteLLM_VerificationToken @relation(fields: [token], references: [token])
@@unique([jwt_claim_name, jwt_claim_value])
@@index([jwt_claim_name, jwt_claim_value, is_active])
}
// Deprecated keys during grace period - allows old key to work until revoke_at
model LiteLLM_DeprecatedVerificationToken {
id String @id @default(uuid())
token String // Hashed old key
active_token_id String // Current token hash in LiteLLM_VerificationToken
revoke_at DateTime // When the old key stops working
created_at DateTime @default(now()) @map("created_at")
@@unique([token])
@@index([token, revoke_at])
@@index([revoke_at])
}
// Audit table for deleted keys - preserves spend and key information for historical tracking
model LiteLLM_DeletedVerificationToken {
id String @id @default(uuid())
token String // Original token (hashed)
key_name String?
key_alias String?
soft_budget_cooldown Boolean @default(false)
spend Float @default(0.0)
expires DateTime?
models String[]
aliases Json @default("{}")
config Json @default("{}")
user_id String?
team_id String?
agent_id String?
project_id String?
permissions Json @default("{}")
max_parallel_requests Int?
metadata Json @default("{}")
blocked Boolean?
tpm_limit BigInt?
rpm_limit BigInt?
max_budget Float?
budget_duration String?
budget_reset_at DateTime?
allowed_cache_controls String[] @default([])
allowed_routes String[] @default([])
policies String[] @default([])
access_group_ids String[] @default([])
model_spend Json @default("{}")
model_max_budget Json @default("{}")
router_settings Json? @default("{}")
budget_id String?
organization_id String?
object_permission_id String?
created_at DateTime? // Original creation timestamp
created_by String? // Original creator
updated_at DateTime? // Last update timestamp before deletion
updated_by String? // Last user who updated before deletion
last_active DateTime? // When this key was last used before deletion
rotation_count Int? @default(0)
auto_rotate Boolean? @default(false)
rotation_interval String?
last_rotation_at DateTime?
key_rotation_at DateTime?
// Deletion metadata
deleted_at DateTime @default(now()) @map("deleted_at")
deleted_by String? @map("deleted_by") // User who deleted the key
deleted_by_api_key String? @map("deleted_by_api_key") // API key hash that performed the deletion
litellm_changed_by String? @map("litellm_changed_by") // From litellm-changed-by header if provided
@@index([token])
@@index([deleted_at])
@@index([user_id])
@@index([team_id])
@@index([organization_id])
@@index([key_alias])
@@index([created_at])
}
model LiteLLM_EndUserTable {
user_id String @id
alias String? // admin-facing alias
spend Float @default(0.0)
allowed_model_region String? // require all user requests to use models in this specific region
default_model String? // use along with 'allowed_model_region'. if no available model in region, default to this model.
budget_id String?
object_permission_id String?
litellm_budget_table LiteLLM_BudgetTable? @relation(fields: [budget_id], references: [budget_id])
object_permission LiteLLM_ObjectPermissionTable? @relation(fields: [object_permission_id], references: [object_permission_id])
blocked Boolean @default(false)
}
// Track tags with budgets and spend
model LiteLLM_TagTable {
tag_name String @id
description String?
models String[]
model_info Json? // maps model_id to model_name
spend Float @default(0.0)
budget_id String?
litellm_budget_table LiteLLM_BudgetTable? @relation(fields: [budget_id], references: [budget_id])
created_at DateTime @default(now()) @map("created_at")
created_by String?
updated_at DateTime @default(now()) @updatedAt @map("updated_at")
}
// store proxy config.yaml
model LiteLLM_Config {
param_name String @id
param_value Json?
}
// View spend, model, api_key per request
model LiteLLM_SpendLogs {
request_id String @id
call_type String
api_key String @default ("") // Hashed API Token. Not the actual Virtual Key. Equivalent to 'token' column in LiteLLM_VerificationToken
spend Float @default(0.0)
total_tokens Int @default(0)
prompt_tokens Int @default(0)
completion_tokens Int @default(0)
startTime DateTime // Assuming start_time is a DateTime field
endTime DateTime // Assuming end_time is a DateTime field
request_duration_ms Int?
completionStartTime DateTime? // Assuming completionStartTime is a DateTime field
model String @default("")
model_id String? @default("") // the model id stored in proxy model db
model_group String? @default("") // public model_name / model_group
custom_llm_provider String? @default("") // litellm used custom_llm_provider
api_base String? @default("")
user String? @default("")
metadata Json? @default("{}") // project_id stored here
cache_hit String? @default("")
cache_key String? @default("")
request_tags Json? @default("[]")
team_id String?
organization_id String?
end_user String?
requester_ip_address String?
messages Json? @default("{}")
response Json? @default("{}")
session_id String?
status String?
mcp_namespaced_tool_name String?
agent_id String?
proxy_server_request Json? @default("{}")
@@index([startTime])
@@index([startTime, request_id])
@@index([end_user])
@@index([session_id])
}
// View spend, model, api_key per request
model LiteLLM_ErrorLogs {
request_id String @id @default(uuid())
startTime DateTime // Assuming start_time is a DateTime field
endTime DateTime // Assuming end_time is a DateTime field
api_base String @default("")
model_group String @default("") // public model_name / model_group
litellm_model_name String @default("") // model passed to litellm
model_id String @default("") // ID of model in ProxyModelTable
request_kwargs Json @default("{}")
exception_type String @default("")
exception_string String @default("")
status_code String @default("")
}
// Beta - allow team members to request access to a model
model LiteLLM_UserNotifications {
request_id String @id
user_id String
models String[]
justification String
status String // approved, disapproved, pending
}
model LiteLLM_TeamMembership {
// Use this table to track the Internal User's Spend within a Team + Set Budgets, rpm limits for the user within the team
user_id String
team_id String
spend Float @default(0.0)
total_spend Float @default(0.0)
budget_id String?
litellm_budget_table LiteLLM_BudgetTable? @relation(fields: [budget_id], references: [budget_id])
@@id([user_id, team_id])
}
model LiteLLM_OrganizationMembership {
// Use this table to track Internal User and Organization membership. Helps tracking a users role within an Organization
user_id String
organization_id String
user_role String?
spend Float? @default(0.0)
budget_id String?
created_at DateTime? @default(now()) @map("created_at")
updated_at DateTime? @default(now()) @updatedAt @map("updated_at")
// relations
user LiteLLM_UserTable @relation(fields: [user_id], references: [user_id])
organization LiteLLM_OrganizationTable @relation("OrganizationToMembership", fields: [organization_id], references: [organization_id])
litellm_budget_table LiteLLM_BudgetTable? @relation(fields: [budget_id], references: [budget_id])
@@id([user_id, organization_id])
@@unique([user_id, organization_id])
}
model LiteLLM_InvitationLink {
// use this table to track invite links sent by admin for people to join the proxy
id String @id @default(uuid())
user_id String
is_accepted Boolean @default(false)
accepted_at DateTime? // when link is claimed (user successfully onboards via link)
expires_at DateTime // till when is link valid
created_at DateTime // when did admin create the link
created_by String // who created the link
updated_at DateTime // when was invite status updated
updated_by String // who updated the status (admin/user who accepted invite)
// Relations
liteLLM_user_table_user LiteLLM_UserTable @relation("UserId", fields: [user_id], references: [user_id])
liteLLM_user_table_created LiteLLM_UserTable @relation("CreatedBy", fields: [created_by], references: [user_id])
liteLLM_user_table_updated LiteLLM_UserTable @relation("UpdatedBy", fields: [updated_by], references: [user_id])
}
model LiteLLM_AuditLog {
id String @id @default(uuid())
updated_at DateTime @default(now())
changed_by String @default("") // user or system that performed the action
changed_by_api_key String @default("") // api key hash that performed the action
action String // create, update, delete
table_name String // on of LitellmTableNames.TEAM_TABLE_NAME, LitellmTableNames.USER_TABLE_NAME, LitellmTableNames.PROXY_MODEL_TABLE_NAME,
object_id String // id of the object being audited. This can be the key id, team id, user id, model id
before_value Json? // value of the row
updated_values Json? // value of the row after change
}
// Track daily user spend metrics per model and key
model LiteLLM_DailyUserSpend {
id String @id @default(uuid())
user_id String?
date String
api_key String
model String?
model_group String?
custom_llm_provider String?
mcp_namespaced_tool_name String?
endpoint String?
prompt_tokens BigInt @default(0)
completion_tokens BigInt @default(0)
cache_read_input_tokens BigInt @default(0)
cache_creation_input_tokens BigInt @default(0)
spend Float @default(0.0)
api_requests BigInt @default(0)
successful_requests BigInt @default(0)
failed_requests BigInt @default(0)
created_at DateTime @default(now())
updated_at DateTime @updatedAt
@@unique([user_id, date, api_key, model, custom_llm_provider, mcp_namespaced_tool_name, endpoint])
@@index([date])
@@index([user_id, date])
@@index([api_key])
@@index([model])
@@index([mcp_namespaced_tool_name])
@@index([endpoint])
}
// Track daily organization spend metrics per model and key
model LiteLLM_DailyOrganizationSpend {
id String @id @default(uuid())
organization_id String?
date String
api_key String
model String?
model_group String?
custom_llm_provider String?
mcp_namespaced_tool_name String?
endpoint String?
prompt_tokens BigInt @default(0)
completion_tokens BigInt @default(0)
cache_read_input_tokens BigInt @default(0)
cache_creation_input_tokens BigInt @default(0)
spend Float @default(0.0)
api_requests BigInt @default(0)
successful_requests BigInt @default(0)
failed_requests BigInt @default(0)
created_at DateTime @default(now())
updated_at DateTime @updatedAt
@@unique([organization_id, date, api_key, model, custom_llm_provider, mcp_namespaced_tool_name, endpoint])
@@index([date])
@@index([organization_id, date])
@@index([api_key])
@@index([model])
@@index([mcp_namespaced_tool_name])
@@index([endpoint])
}
// Track daily end user (customer) spend metrics per model and key
model LiteLLM_DailyEndUserSpend {
id String @id @default(uuid())
end_user_id String?
date String
api_key String
model String?
model_group String?
custom_llm_provider String?
mcp_namespaced_tool_name String?
endpoint String?
prompt_tokens BigInt @default(0)
completion_tokens BigInt @default(0)
cache_read_input_tokens BigInt @default(0)
cache_creation_input_tokens BigInt @default(0)
spend Float @default(0.0)
api_requests BigInt @default(0)
successful_requests BigInt @default(0)
failed_requests BigInt @default(0)
created_at DateTime @default(now())
updated_at DateTime @updatedAt
@@unique([end_user_id, date, api_key, model, custom_llm_provider, mcp_namespaced_tool_name, endpoint])
@@index([date])
@@index([end_user_id, date])
@@index([api_key])
@@index([model])
@@index([mcp_namespaced_tool_name])
@@index([endpoint])
}
// Track daily agent spend metrics per model and key
model LiteLLM_DailyAgentSpend {
id String @id @default(uuid())
agent_id String?
date String
api_key String
model String?
model_group String?
custom_llm_provider String?
mcp_namespaced_tool_name String?
endpoint String?
prompt_tokens BigInt @default(0)
completion_tokens BigInt @default(0)
cache_read_input_tokens BigInt @default(0)
cache_creation_input_tokens BigInt @default(0)
spend Float @default(0.0)
api_requests BigInt @default(0)
successful_requests BigInt @default(0)
failed_requests BigInt @default(0)
created_at DateTime @default(now())
updated_at DateTime @updatedAt
@@unique([agent_id, date, api_key, model, custom_llm_provider, mcp_namespaced_tool_name, endpoint])
@@index([date])
@@index([agent_id, date])
@@index([api_key])
@@index([model])
@@index([mcp_namespaced_tool_name])
@@index([endpoint])
}
// Track daily team spend metrics per model and key
model LiteLLM_DailyTeamSpend {
id String @id @default(uuid())
team_id String?
date String
api_key String
model String?
model_group String?
custom_llm_provider String?
mcp_namespaced_tool_name String?
endpoint String?
prompt_tokens BigInt @default(0)
completion_tokens BigInt @default(0)
cache_read_input_tokens BigInt @default(0)
cache_creation_input_tokens BigInt @default(0)
spend Float @default(0.0)
api_requests BigInt @default(0)
successful_requests BigInt @default(0)
failed_requests BigInt @default(0)
created_at DateTime @default(now())
updated_at DateTime @updatedAt
@@unique([team_id, date, api_key, model, custom_llm_provider, mcp_namespaced_tool_name, endpoint])
@@index([date])
@@index([team_id, date])
@@index([api_key])
@@index([model])
@@index([mcp_namespaced_tool_name])
@@index([endpoint])
}
// Track daily team spend metrics per model and key
model LiteLLM_DailyTagSpend {
id String @id @default(uuid())
request_id String?
tag String?
date String
api_key String
model String?
model_group String?
custom_llm_provider String?
mcp_namespaced_tool_name String?
endpoint String?
prompt_tokens BigInt @default(0)
completion_tokens BigInt @default(0)
cache_read_input_tokens BigInt @default(0)
cache_creation_input_tokens BigInt @default(0)
spend Float @default(0.0)
api_requests BigInt @default(0)
successful_requests BigInt @default(0)
failed_requests BigInt @default(0)
created_at DateTime @default(now())
updated_at DateTime @updatedAt
@@unique([tag, date, api_key, model, custom_llm_provider, mcp_namespaced_tool_name, endpoint])
@@index([date])
@@index([tag, date])
@@index([api_key])
@@index([model])
@@index([mcp_namespaced_tool_name])
@@index([endpoint])
}
// Track the status of cron jobs running. Only allow one pod to run the job at a time
model LiteLLM_CronJob {
cronjob_id String @id @default(cuid()) // Unique ID for the record
pod_id String // Unique identifier for the pod acting as the leader
status JobStatus @default(INACTIVE) // Status of the cron job (active or inactive)
last_updated DateTime @default(now()) // Timestamp for the last update of the cron job record
ttl DateTime // Time when the leader's lease expires
}
enum JobStatus {
ACTIVE
INACTIVE
}
model LiteLLM_ManagedFileTable {
id String @id @default(uuid())
unified_file_id String @unique // The base64 encoded unified file ID
file_object Json? // Stores the OpenAIFileObject
model_mappings Json
flat_model_file_ids String[] @default([]) // Flat list of model file id's - for faster querying of model id -> unified file id
storage_backend String? // Storage backend name (e.g., "azure_storage", "gcs", "default")
storage_url String? // The actual storage URL where the file is stored
created_at DateTime @default(now())
created_by String?
team_id String? // Team that owns the resource; populated for service-account keys without a user_id so listings can isolate by team.
updated_at DateTime @updatedAt
updated_by String?
@@index([unified_file_id])
@@index([team_id, created_at(sort: Desc)])
}
model LiteLLM_ManagedObjectTable { // for batches or finetuning jobs which use the
id String @id @default(uuid())
unified_object_id String @unique // The base64 encoded unified file ID
model_object_id String @unique // the id returned by the backend API provider
file_object Json // Stores the OpenAIFileObject
file_purpose String // either 'batch' or 'fine-tune'
status String? // check if batch cost has been tracked
batch_processed Boolean @default(false) // set to true by CheckBatchCost after cost is computed
created_at DateTime @default(now())
created_by String?
team_id String?
updated_at DateTime @updatedAt
updated_by String?
@@index([unified_object_id])
@@index([model_object_id])
@@index([team_id, created_at(sort: Desc)])
}
model LiteLLM_ManagedVectorStoreTable {
id String @id @default(uuid())
unified_resource_id String @unique // The base64 encoded unified vector store ID
resource_object Json? // Stores the VectorStoreCreateResponse
model_mappings Json // Maps model_id -> provider_vector_store_id
flat_model_resource_ids String[] @default([]) // Flat list of provider vector store IDs for faster querying
storage_backend String? // Storage backend name (if applicable)
storage_url String? // Storage URL (if applicable)
created_at DateTime @default(now())
created_by String?
team_id String?
updated_at DateTime @updatedAt
updated_by String?
@@index([unified_resource_id])
@@index([team_id, created_at(sort: Desc)])
}
model LiteLLM_ManagedVectorStoresTable {
vector_store_id String @id
custom_llm_provider String
vector_store_name String?
vector_store_description String?
vector_store_metadata Json?
created_at DateTime @default(now())
updated_at DateTime @updatedAt
litellm_credential_name String?
litellm_params Json?
team_id String?
user_id String?
@@index([team_id])
@@index([user_id])
}
// Guardrails table for storing guardrail configurations
model LiteLLM_GuardrailsTable {
guardrail_id String @id @default(uuid())
guardrail_name String @unique
litellm_params Json
guardrail_info Json?
team_id String?
created_at DateTime @default(now())
updated_at DateTime @updatedAt
// Submission lifecycle. Possible values: pending_review (team-registered, awaiting approval), active (approved), rejected
status String @default("active")
submitted_at DateTime?
reviewed_at DateTime?
// submitted_by_user_id and submitted_by_email live in guardrail_info JSON
@@index([status])
}
// Daily guardrail metrics for usage dashboard (one row per guardrail per day)
model LiteLLM_DailyGuardrailMetrics {
guardrail_id String // logical id; may not FK if guardrail from config
date String // YYYY-MM-DD
requests_evaluated BigInt @default(0)
passed_count BigInt @default(0)
blocked_count BigInt @default(0)
flagged_count BigInt @default(0)
avg_score Float?
avg_latency_ms Float?
created_at DateTime @default(now())
updated_at DateTime @updatedAt
@@id([guardrail_id, date])
@@index([date])
@@index([guardrail_id])
}
// Daily policy metrics for usage dashboard (one row per policy per day)
model LiteLLM_DailyPolicyMetrics {
policy_id String
date String // YYYY-MM-DD
requests_evaluated BigInt @default(0)
passed_count BigInt @default(0)
blocked_count BigInt @default(0)
flagged_count BigInt @default(0)
avg_score Float?
avg_latency_ms Float?
created_at DateTime @default(now())
updated_at DateTime @updatedAt
@@id([policy_id, date])
@@index([date])
@@index([policy_id])
}
// Index for fast "last N logs for guardrail/policy" from SpendLogs
model LiteLLM_SpendLogGuardrailIndex {
request_id String
guardrail_id String
policy_id String? // set when run as part of a policy pipeline
start_time DateTime
@@id([request_id, guardrail_id])
@@index([guardrail_id, start_time])
@@index([policy_id, start_time])
}
// Index for fast "last N logs for tool" from SpendLogs see how a tool is called in production
model LiteLLM_SpendLogToolIndex {
request_id String
tool_name String // matches LiteLLM_ToolTable.tool_name; join for input_policy/output_policy etc.
start_time DateTime
@@id([request_id, tool_name])
@@index([tool_name, start_time])
}
// Prompt table for storing prompt configurations
model LiteLLM_PromptTable {
id String @id @default(uuid())
prompt_id String
version Int @default(1)
environment String @default("development")
created_by String?
litellm_params Json
prompt_info Json?
created_at DateTime @default(now())
updated_at DateTime @updatedAt
@@unique([prompt_id, version, environment])
@@index([prompt_id, environment])
@@index([prompt_id])
}
model LiteLLM_HealthCheckTable {
health_check_id String @id @default(uuid())
model_name String
model_id String?
status String
healthy_count Int @default(0)
unhealthy_count Int @default(0)
error_message String?
response_time_ms Float?
details Json?
checked_by String?
checked_at DateTime @default(now())
created_at DateTime @default(now())
updated_at DateTime @updatedAt
@@index([model_name])
@@index([checked_at])
@@index([status])
@@index([model_id, model_name, checked_at(sort: Desc)], map: "LiteLLM_HealthCheckTable_model_id_model_name_checked_at_idx")
}
// Search Tools table for storing search tool configurations
model LiteLLM_SearchToolsTable {
search_tool_id String @id @default(uuid())
search_tool_name String @unique
litellm_params Json
search_tool_info Json?
created_at DateTime @default(now())
updated_at DateTime @updatedAt
}
// SSO configuration table
model LiteLLM_SSOConfig {
id String @id @default("sso_config")
sso_settings Json
created_at DateTime @default(now())
updated_at DateTime @updatedAt
}
model LiteLLM_ManagedVectorStoreIndexTable {
id String @id @default(uuid())
index_name String @unique
litellm_params Json
index_info Json?
created_at DateTime @default(now())
created_by String?
updated_at DateTime @updatedAt
updated_by String?
}
// Cache configuration table
model LiteLLM_CacheConfig {
id String @id @default("cache_config")
cache_settings Json
created_at DateTime @default(now())
updated_at DateTime @updatedAt
}
// UI Settings configuration table
model LiteLLM_UISettings {
id String @id @default("ui_settings")
ui_settings Json
created_at DateTime @default(now())
updated_at DateTime @updatedAt
}
// Generic config overrides table - one row per config_type
model LiteLLM_ConfigOverrides {
config_type String @id
config_value Json
created_at DateTime @default(now())
updated_at DateTime @updatedAt
}
// Skills table for storing LiteLLM-managed skills
model LiteLLM_SkillsTable {
skill_id String @id @default(uuid())
display_title String?
description String?
instructions String? // The skill instructions/prompt (from SKILL.md)
source String @default("custom") // "custom" or "anthropic"
latest_version String?
file_content Bytes? // Binary content of the skill files (zip)
file_name String? // Original filename
file_type String? // MIME type (e.g., "application/zip")
metadata Json? @default("{}")
created_at DateTime @default(now())
created_by String?
updated_at DateTime @default(now()) @updatedAt
updated_by String?
}
// Policy table for storing guardrail policies (versioned)
model LiteLLM_PolicyTable {
policy_id String @id @default(uuid())
policy_name String // No longer @unique; use @@unique([policy_name, version_number])
version_number Int @default(1)
version_status String @default("production") // "draft" | "published" | "production"
parent_version_id String?
is_latest Boolean @default(true)
published_at DateTime?
production_at DateTime?
inherit String? // Name of parent policy to inherit from
description String?
guardrails_add String[] @default([])
guardrails_remove String[] @default([])
condition Json? @default("{}") // Policy conditions (e.g., model matching)
pipeline Json? // Optional guardrail pipeline (mode + steps[])
created_at DateTime @default(now())
created_by String?
updated_at DateTime @default(now()) @updatedAt
updated_by String?
@@unique([policy_name, version_number])
@@index([policy_name, version_status])
}
// Policy attachment table for defining where policies apply
model LiteLLM_PolicyAttachmentTable {
attachment_id String @id @default(uuid())
policy_name String // Name of the policy to attach
scope String? // Use '*' for global scope
teams String[] @default([]) // Team aliases or patterns
keys String[] @default([]) // Key aliases or patterns
models String[] @default([]) // Model names or patterns
tags String[] @default([]) // Tag patterns (e.g., ["healthcare", "prod-*"])
created_at DateTime @default(now())
created_by String?
updated_at DateTime @default(now()) @updatedAt
updated_by String?
}
// Global tool registry - auto-discovered from LLM responses; admins set input/output policies here
model LiteLLM_ToolTable {
tool_id String @id @default(uuid())
tool_name String @unique // e.g. "huggingface_remote-mcp__dynamic_space"
origin String? // MCP server name or "user_defined"
input_policy String @default("untrusted") // "trusted" | "untrusted" | "blocked"
output_policy String @default("untrusted") // "trusted" | "untrusted"
call_count Int @default(0) // cumulative number of times this tool was seen
assignments Json? @default("{}")
key_hash String? // hash of the virtual key that first called this tool
team_id String? // team that first called this tool
key_alias String? // human-readable alias of the virtual key
user_agent String? // user-agent of the first request that discovered this tool
last_used_at DateTime? // timestamp of the most recent call
created_at DateTime @default(now())
created_by String?
updated_at DateTime @default(now()) @updatedAt
updated_by String?
@@index([input_policy])
@@index([output_policy])
@@index([team_id])
}
// Per-(tool, team/key) policy overrides. When present, override replaces global tool policy for that scope.
//Unified Access Groups table for storing unified access groups
model LiteLLM_AccessGroupTable {
access_group_id String @id @default(uuid())
access_group_name String @unique
description String?
// Resource memberships - explicit arrays per type
access_model_names String[] @default([])
access_mcp_server_ids String[] @default([])
access_agent_ids String[] @default([])
assigned_team_ids String[] @default([])
assigned_key_ids String[] @default([])
created_at DateTime @default(now())
created_by String?
updated_at DateTime @default(now()) @updatedAt
updated_by String?
}
// Claude Code Plugin Marketplace table
model LiteLLM_ClaudeCodePluginTable {
id String @id @default(uuid())
name String @unique
version String?
description String?
manifest_json String?
files_json String? @default("{}")
enabled Boolean @default(true)
created_at DateTime? @default(now())
updated_at DateTime? @default(now()) @updatedAt
created_by String?
@@map("LiteLLM_ClaudeCodePluginTable")
}
// User/team-scoped memory store with a GLOBAL unique key.
// `value` is a string (typically markdown/text meant for LLM context);
// `metadata` is an optional JSON envelope for structured tags without schema changes.
// Note: `key` is globally unique across all users/teams — callers should
// namespace their keys (e.g. `user:123:notes`) if they need per-user isolation.
// `user_id` / `team_id` stamp ownership for visibility filtering, but do NOT
// participate in the unique constraint.
model LiteLLM_MemoryTable {
memory_id String @id @default(uuid())
key String @unique
value String
metadata Json?
user_id String?
team_id String?
created_at DateTime @default(now())
created_by String?
updated_at DateTime @default(now()) @updatedAt
updated_by String?
@@index([user_id])
@@index([team_id])
}
// Per-(router, request_type, model) Beta posterior for the adaptive router.
model LiteLLM_AdaptiveRouterState {
router_name String
request_type String
model_name String
alpha Float
beta Float
total_samples Int @default(0)
last_updated_at DateTime @default(now()) @updatedAt
@@id([router_name, request_type, model_name])
}
// Per-(session, router, model) signal counters for the adaptive router.
model LiteLLM_AdaptiveRouterSession {
session_id String
router_name String
model_name String
classified_type String
misalignment_count Int @default(0)
stagnation_count Int @default(0)
disengagement_count Int @default(0)
satisfaction_count Int @default(0)
failure_count Int @default(0)
loop_count Int @default(0)
exhaustion_count Int @default(0)
last_user_content String?
last_assistant_content String?
tool_call_history Json @default("[]")
pending_tool_calls Json @default("{}")
turn_count Int @default(0)
last_processed_turn Int @default(-1)
clean_credit_awarded Boolean @default(false)
terminal_status Int?
last_activity_at DateTime @default(now()) @updatedAt
@@id([session_id, router_name, model_name])
@@index([last_activity_at], map: "idx_adaptive_router_session_activity")
}
// ---------------------------------------------------------------------------
// Workflow Run Tracking
//
// Generic durable state tracking for any agent or automated workflow.
// Design: three tables — run (header + materialized status), event (append-only
// source of truth for state transitions), message (conversation inbox/outbox).
//
// Usage:
// - Set `workflow_type` to identify the owning system (e.g. "shin-builder").
// - Store domain-specific fields in `metadata` (worktree_path, pr_url, etc.).
// - `session_id` on WorkflowRun matches `x-litellm-session-id` header sent to
// the proxy — all spend logs for this run are automatically tagged.
// ---------------------------------------------------------------------------
// One instance of work being done. `status` is a materialized cache of the
// latest event; the event log is the authoritative source of truth.
model LiteLLM_WorkflowRun {
run_id String @id @default(uuid())
session_id String @unique @default(uuid())
workflow_type String
status String @default("pending")
created_by String? // user_id of the key that created this run; null = created by master key
created_at DateTime @default(now())
updated_at DateTime @updatedAt
input Json?
output Json?
metadata Json?
events LiteLLM_WorkflowEvent[]
messages LiteLLM_WorkflowMessage[]
@@index([workflow_type, status])
@@index([session_id])
@@index([created_at])
@@index([created_by])
}
// Append-only log of state transitions. Never mutate rows here.
// `step_name` and `event_type` are caller-defined strings — no hardcoded enums.
// Status auto-update rules (applied by the append endpoint):
// step.started → run.status = running
// step.failed → run.status = failed
// hook.waiting → run.status = paused
// hook.received → run.status = running
model LiteLLM_WorkflowEvent {
event_id String @id @default(uuid())
run_id String
event_type String
step_name String
sequence_number Int
data Json?
created_at DateTime @default(now())
run LiteLLM_WorkflowRun @relation(fields: [run_id], references: [run_id])
@@unique([run_id, sequence_number])
@@index([run_id])
}
// Conversation inbox/outbox — full message content, separate from the durable
// event log. Spend logs truncate messages; this table stores them in full.
// `session_id` here is the Claude --resume session ID (or similar).
model LiteLLM_WorkflowMessage {
message_id String @id @default(uuid())
run_id String
role String
content String
sequence_number Int
session_id String?
created_at DateTime @default(now())
run LiteLLM_WorkflowRun @relation(fields: [run_id], references: [run_id])
@@unique([run_id, sequence_number])
@@index([run_id])
}