* fix: rebuild models_by_provider in add_known_models so cost map reloads reach wildcard expansion
* fix: refresh models_by_provider in place so captured references survive reloads
A connection test that redirects the destination already leaves the configured
credentials behind. It kept litellm_credential_name, which names the same stored
secrets and is resolved further down the call, so the reference is now dropped
with them. A request that sets no connection fields of its own is unaffected,
which is how the Admin UI tests a configured model.
The proxy-wide opt-in that already governs callers supplying their own
connection parameters now also governs whether a connection test may pair a
request-supplied endpoint with the configured deployment's credentials. Off by
default, which keeps configured credentials scoped to the endpoint the
configuration names; on, the previous merge behaviour is available unchanged.
A request that supplies its own connection fields describes a connection of its
own, so the configured deployment's credentials are no longer merged underneath
it. Anything the request leaves unset still comes from the configuration, so
naming a configured model and testing it as configured is unchanged, and adding
a second deployment for an already-configured name works as before.
Replaces the earlier outright rejection, which also refused requests that
supplied a complete connection of their own.
When a connection test names a model that resolves to a configured deployment,
that deployment's routing and credential parameters are authoritative. A request
supplying a complete connection of its own is unaffected.
BREAKING CHANGE: /health/test_connection no longer lets a request replace the
routing or credential parameters of a configured model it names. Supply the full
connection parameters instead of naming a configured model.
Multipart callers express nested metadata as flat bracket-notation keys, which
reach the request-body check as literal keys rather than as a metadata dict.
The check now rebuilds them with the same helper the endpoints use, so both
encodings are handled identically and cannot drift apart.
BREAKING CHANGE: a multipart field such as `litellm_metadata[api_base]` is now
subject to the same request-body parameter rules as its JSON equivalent. Set
`general_settings.allow_client_side_credentials`, or the deployment's
`configurable_clientside_auth_params`, to keep passing these.
The URL-destination check previously ran over request-body fields only. The
per-field logic moves into reject_url_valued_destination(field, value) so a
deployment name resolved from the request path runs the same check against the
same admin allowlist.
BREAKING CHANGE: a deployment name supplied in the request path that parses as
an http/https destination is now refused. Add the host to
`provider_url_destination_allowed_hosts` in litellm_settings to keep it working.
* feat(pre-commit): save full lint output to a per-worktree log file
* docs(claude): point agents at the pre-commit log instead of rerunning
* fix(pre-commit): warn when the log cannot be created or fully written
An SSE stream that cannot be positively identified as Anthropic (no
parseable message_start event) now blocks instead of passing through
unscanned, closing the bypass where any raw-SSE backend skipped tool
permission checks entirely. Buffered chunks are joined back into one
stream before parsing, so events split across network chunk boundaries
assemble correctly instead of being silently dropped. Rewrite mode now
resets finish_reason to stop when no tool call survives, so the
re-encoded Anthropic stream reports stop_reason end_turn and clients do
not wait for a tool result that never comes
A cached HTTPHandler hands its raw httpx.Client out to consumers that keep it
for the process lifetime. When the shared client cache expires the entry on its
TTL or evicts it under the 200-entry cap, nothing references the handler, so it
is collected and its finalizer closed the client those consumers still hold.
Langfuse ingestion then failed silently on the SDK's background flush thread
until the process restarted.
A finalizer running proves only that nothing references the handler; it proves
nothing about the client. Both handlers now close the client during finalization
only when they built it and are still its sole referrer, so an unshared client is
still released promptly and a handed-out one is left alone. That keeps the
pooled-socket reclamation the finalizer was providing, which measures identical
to base over 2000 handler create-and-drop cycles.
Explicit close() stays, now gated on _owns_client so the wrapper never closes a
caller-injected client, and __aexit__ routes through it.
LangFuseLogger also keeps a reference to the handler whose client it hands the
SDK. Previously that handler was a local that went out of scope immediately,
leaving the client reachable only from the SDK. It still shares the cached
client, so no extra clients are created per logger.
test_reapply_runtime_registrations_replays_register_model_overrides asserts that
a fetched catalog value survives the replay for a key an operator override does
not mention. Any Router still alive in the process re-asserts its own deployments
first, so a router serving openai/gpt-4o writes its model_info over that catalog
value and the assertion reads the router's number instead. Routers built by
earlier tests stay in the weak set until they are collected, which made the test
depend on collection timing and fail intermittently in shards that run the router
tests alongside it.
The live-router rebuild is covered in test_router_model_cost_isolation.py, so
this test now runs with the replay callback unset and exercises the recorded
registrations it is about.
The redaction filter was attached to the handler shared by litellm's own
loggers, so it only covered records litellm emits. A litellm value can also
reach a log record through a dependency logging on its own logger, and those
records never pass through a litellm handler.
Attach the filter to each dependency logger that can carry one. The filter goes
on the emitting logger rather than on the root logger or a root handler, since
Logger.handle applies the emitting logger's filters before any handler runs, so
every downstream handler is covered regardless of who owns it.
The gate required a litellm. prefix on get_secret and get_secret_str, so any
module importing either function directly bypassed it: 335 environment variables
read under litellm/ were invisible to it. The three patterns collapse into one
with the prefix optional, a negative lookbehind so attribute calls on unrelated
objects cannot match, and litellm.utils. accepted since four call sites reach
get_secret that way.
Widening the patterns alone would demand about 320 new rows in the central
reference table, most of them provider credentials that are already documented
on their own provider pages. So the gate now looks across every page of the docs
site rather than only that one table, which leaves 143 keys genuinely
undocumented instead of 322.
* fix(auto-router): stop the embedding model's context window from failing long requests
The auto-router embeds the last user message to pick a model and sent it to the
embedding model unbounded. Embedding models carry 512 to 8k token windows while the
chat models they route to carry 200k+, so any prompt over the encoder's window failed
at the routing step with a 400 the destination model would never have raised.
Cut every doc to a character cap inside LiteLLMRouterEncoder, which is the one choke
point the auto-router, complexity-router, semantic guard and MCP tool filter all share.
Default 2000 chars, roughly 500 tokens, which fits even a 512-token self-hosted encoder,
overridable per deployment with auto_router_max_input_chars and globally with
DEFAULT_MAX_EMBEDDING_INPUT_CHARS.
Truncation alone cannot cover provider-side batch and byte limits, so any failure of
the route call now falls back to the auto-router's default model instead of propagating.
That path also fixes two latent bugs: a no-match left the auto-router alias in place as
the model name, which fails downstream with "Unmapped LLM provider" rather than reaching
default_model, and an empty route list raised IndexError.
Fixes#17869Fixes#20277
* fix(auto-router): make the embedding input cap opt-in so guards still see whole prompts
Defaulting the cap inside the shared encoder truncated every consumer, not just the
auto-router. The semantic guard builds the same encoder, so its pre-call check would
have classified only the first 2000 characters while the full message still reached the
model, which a benign opener in front of an injection payload walks straight past. The
MCP tool filter and complexity router were silently narrowed the same way.
The encoder now defaults to sending docs whole and cuts only when a caller passes
max_input_chars. The auto-router is the only caller that does, so guard, MCP filter and
complexity-router behaviour is unchanged from before this branch.
DEFAULT_MAX_EMBEDDING_INPUT_CHARS becomes DEFAULT_AUTO_ROUTER_MAX_INPUT_CHARS, since it
is now specific to the auto-router, and drops its env override: the per-deployment
auto_router_max_input_chars already covers it, and every env var in constants.py has to
be documented, which is what broke the documentation and code-quality checks.
Also drops the added comments and the redundant type: ignore that review flagged.
* test(auto-router): cover the max_input_chars wiring from litellm_params
Nothing asserted that auto_router_max_input_chars on the deployment reaches the
AutoRouter that embeds prompts. Dropping the wiring left every test green while the cap
silently reverted to the default, so an operator with a 512-token embedding model could
not lower it and every long prompt would fall back to the default model instead of
being routed.
* test(auto-router): cover the populated route-choice list branch
The route layer can hand back a list, and picking its first element is where the
IndexError lived: the empty case was covered but the populated one was not, so the
branch that reads route_choice[0].name could be deleted with every test still green.
`is_database_connection_error` answered True for any `PrismaError` it did not
recognize, on the reasoning that an unclassified failure might be an outage and
the safer default was to keep serving. That default is inverted for faults that
never resolve. A query engine that is missing or version-skewed, a malformed
generated query, or a misused transaction all satisfied the predicate, so with
`allow_requests_on_db_unavailable` enabled the proxy would absorb one, boot
clean, and keep issuing fallback identities for as long as the process ran.
The predicate is now an allowlist: the httpx transport errors, prisma's
`EngineConnectionError`, and a `no_db_connection` ProxyException. That is what a
real outage produces, since the query engine is a local HTTP server and an
unreachable database surfaces as a transport failure against it, so the
high-availability path is unchanged. Anything unrecognized is now treated as
permanent and surfaces instead of being absorbed.
Deciding whether to serve without a database and deciding what to tell the
caller are different questions, so they no longer share a predicate.
`is_database_infrastructure_error` keeps the previous broad behavior and now
backs the reporting and recovery paths: service-unavailable classification, the
access-group endpoint's status mapping, and the health watchdog's reconnect
trigger. Their behavior is unchanged. Without that split, a permanently faulted
engine would have started reporting as an authentication failure, sending an
operator after a credential problem that does not exist.
The Jina key fallback chain read JINA_AI_API_KEY three times in a row
before falling through to JINA_AI_TOKEN, so two of the four slots were
dead. Jina's own documentation publishes JINA_API_KEY, and litellm's
rerank validate_environment already tells users to set that name, but
nothing ever read it: a user who set only JINA_API_KEY got no key
resolved and Jina answered AUTH_MISSING_API_KEY.
Replace one of the repeats with JINA_API_KEY and drop the other.
JINA_AI_API_KEY stays first so no install that resolves a key today
changes which key it picks.
Guardrails silently skipped three surfaces on the Anthropic Messages
path, so an agent loop driven by /v1/messages ran unguarded:
- The Anthropic input translation never walked tool_result blocks, so
content returned by a local tool (a curl, a file read, an MCP call)
reached the model unscanned in both the string and list content
shapes, images inside a tool_result included.
- tool_permission only understood ModelResponse, so an Anthropic
non-streaming response or a raw SSE stream carrying tool_use blocks
passed through with no rule ever evaluated.
- ContentFilterGuardrail scanned inputs["texts"] but never
inputs["tool_calls"], so the arguments a model proposes for a tool
call went unchecked.
Tool call arguments are parsed as JSON before filtering so a MASK
action rewrites the value and leaves the payload valid JSON; non-JSON
arguments fall back to scanning the raw string. Denied tool_use blocks
are dropped from the Anthropic content array and replaced with a text
block, and stop_reason resets to end_turn when nothing tool-shaped
survives.
Align with the API gateway translation: instead of raising BadRequestError
on alias or competing-constraint conflicts, the explicit Fireworks-native
param wins and the NIM/vLLM extra is dropped with a debug log. Covers
truncate_prompt_tokens vs prompt_truncate_len, chat_template_kwargs
enable_thinking vs reasoning_effort/thinking, guided_* vs response_format
(including response_format nested in an explicit extra_body, which the
previous conflict check missed), and multiple guided_* params (priority
order json, grammar, choice). Malformed non-object chat_template_kwargs
is also dropped with a log instead of raising.
min_tokens is accepted natively by the Fireworks API (verified live), so
stop stripping it and let it pass through extra_body. Add the NIM-specific
include_reasoning and nvext keys to the strip set. enable_thinking=true
now omits reasoning_effort (model default) instead of forcing medium,
matching the gateway translation and preserving default-off models'
behavior; enable_thinking=false still maps to none.
get_api_key resolved the ai21 key from AI211_API_KEY, with a doubled 1. Every other
ai21 code path reads AI21_API_KEY, including the validate_environment branches that
report it as the missing one, so the name a user is told to set was ignored here.
No user path reaches this branch today, since every provider-resolution site rewrites
custom_llm_provider to ai21_chat and sets the key from a correctly spelled read first,
so this is a correctness fix rather than a bug fix. It is worth making because the
env-var documentation gate reads this call site: leaving the misspelling in place would
require a row for AI211_API_KEY in the environment variables reference table, which
would turn a typo into public API
S3 rebuilds the canonical request from the wire path with single
percent-encoding, which botocore models as S3SigV4Auth. Generic SigV4Auth
quotes the already encoded path a second time, so an object key holding any
character that percent-encodes was signed over %2520 while the request
carried %20, and S3 answered 403 SignatureDoesNotMatch.
The S3 logger was corrected in #35726; the Bedrock managed-files upload and
retrieval paths copied that same pre-fix pattern and were left behind, so a
configured bucket prefix with a space 403s every file upload and every file
content read.
Fireworks rejects reasoning_effort="auto" (accepted set: low, medium,
high, xhigh, max, none, adaptive), so OpenAI-compatible clients sending
it 400. Omitting the param means model default on Fireworks, which is
exactly what auto means on OpenAI's side, so skip it in
map_openai_params instead of forwarding.
Folds every successful auto-routed request into LiteLLM_AutoRouterSession with one
conditional upsert at spend-write time, classifying each turn (same model, first
visit, return to tier, out of order) against the row's own columns so nothing is
read before the write. The upsert's placeholders and argument tuple both derive
from the transaction dataclass's own field order, so the SQL and the call site
cannot drift apart. GET /auto_router/benchmarks aggregates the rollup, grouped
by the full (router, type) identity, and never scans LiteLLM_SpendLogs. A turn's
cache interaction is derived once from its usage record (savings.py owns the
extraction; compute_savings_spend derives cache reads from usage_object itself),
hits are counted order-independently so the overall hit rate matches its covered
denominator, caller-chosen session ids are bounded before entering the primary
key, and a poisoned statement drops only its own session's remaining turns.
Return misses inside the recorded TTL are named for what the telemetry shows
(within_ttl) rather than a presumed cause, since a provider can evict early.
Savings ride each router's derived baseline by default, so the response carries
no deployment-wide baseline label. Rollup retention has its own
maximum_autorouter_session_retention_period setting, pattern-identical to the
spend-logs knob and running in the same cleanup job on its own cutoff. Every
drain trigger sizes the queues through one owner and the enqueue honors
disable_spend_logs beside the tool-usage queue it mirrors.
* fix(autorouter): match CJK keyword_tier_rules that regex word boundaries miss
Single-word keywords were matched with a \b...\b regex. Every CJK character is a
regex word character and CJK is written without spaces, so \b never fires between
two of them and a rule like 发票 silently missed 我需要开发票, falling through to
complexity scoring instead of the configured tier.
Keywords containing CJK now match as plain substrings, the same way multi-word
phrases already did. The gate reads the keyword rather than the prompt, so a
keyword with no CJK in it keeps word boundary matching regardless of the script
the prompt is written in.
* fix(autorouter): cover Han extensions in planes 2 and 3, not just up to U+2FA1F
The supplementary range stopped at U+2FA1F, so Extension G and H ideographs kept
the word boundary path and stayed unmatchable. Both planes are dedicated to CJK
ideographs, so covering them whole also handles later extensions without chasing
each new block.
A price data reload replaced litellm.model_cost wholesale, discarding every
runtime registration: the deployment model_info the Router registers from
model_list, and pricing overrides passed to litellm.register_model. Custom
model groups lost max_input_tokens / max_output_tokens in /model_group/info,
and a deployment whose backend model is in the catalog silently reverted to
upstream values. Runtime registrations are now recorded and replayed on top of
the freshly fetched catalog.
Router._pre_call_checks resolved the per-deployment model name only after the
model-info lookup, so an unregistered model left it unset and the supported
params check ran against the bare model group name, raising "LLM Provider NOT
provided" out of deployment selection. The name is now resolved first, and an
unresolvable provider skips that check rather than failing the request.
Resolves LIT-4675
The enterprise gate on the OAuth2 auth path raised a bare `ValueError`,
which the terminal handler in auth_exception_handler.py converts to a 401.
Every sibling enterprise gate answers 403, including `_premium_user_check`
and the SSO gate. A 401 tells the client its credential was wrong and to
retry with a better one, and no credential can satisfy that while the
install is unlicensed, so it invites a retry loop that can never succeed.
It now raises a 403 `ProxyException` shaped like the SSO gate. Two response
fields move with it: the `Authentication Error, ` prefix goes away, since
the catch-all built that around `str(e)` and a `ProxyException` is re-raised
unmodified, and `param` becomes `premium_user`, naming the condition an
operator has to clear.
The gate's own text also gains the sentence break it was missing. The
message concatenated straight onto `CommonProxyErrors.not_premium_user`,
rendering as "premium usersYou must be a LiteLLM Enterprise user".
* feat(auto-router): let operators replace the LLM classifier's system prompt
The complexity router's LLM classifier has always sent one built-in rubric, so the
router could only ever grade difficulty. Operators can now supply their own system
prompt, which replaces the rubric outright and repurposes the same tier machinery for
whatever taxonomy the prompt defines, data sensitivity being the obvious case.
Replacement is total: neither the rubric nor its closing line is appended, since both
describe grading difficulty over a "current message" and a prompt grading something
else is entitled to contradict them. That closing paragraph is also the classifier's
prompt-injection defense, so the config field and the dashboard editor both warn that
a replacement omitting it lets a caller ask for a tier and get it.
The heuristic fallback still scores complexity, which is meaningless for a repurposed
taxonomy, so classifier_fallback now chooses between the heuristic scorer and routing
straight to default_model. The default_model path bypasses tier pools, the adaptive
bandit, and escalation, because no tier was decided and the point of that fallback is
a known destination. It reports itself as default_model_fallback in the spend logs.
The dashboard's prompt editor prefills from a new
/auto_router/classifier/default_prompt endpoint rather than a copy of the rubric in
the frontend, and stores no override when the draft matches the default, so later
rubric improvements still reach every router that never customized it.
Tier names stay SIMPLE/MEDIUM/COMPLEX/REASONING; a custom prompt redefines what they
mean, not what they are called.
* fix(complexity-router): don't let the default_model classifier fallback bypass routing plugins
* fix(complexity-router): don't pin a session to the default model after a classifier failure
* fix(complexity-router): omit the tier from a default-model-fallback routing decision
The classifier never answered, so no tier was decided. The record reported the
tier whose pool happens to hold default_model, which reads in the spend log and
the UI as if the request was classified. Matches how default_fallback already
records a route that no tier produced.
* fix(proxy): allowlist /auto_router/ on the UI backend component
The new GET /auto_router/classifier/default_prompt is a UI-consumed management
route, so it belongs on the control plane. Without the prefix it was exposed by
neither component and test_gateway_plus_backend_covers_full_app failed.
* docs(ui): reword the classifier prompt disclaimer
Frames the closing paragraph as a strong recommendation rather than a
description of what gets dropped, names prompt injection explicitly, and
notes the tier names stay fixed regardless of their display names.
* fix(complexity-router): stop logging a fabricated tier on the plugin fallback path
The classifier-failed fallback resolves a tier so the routing-plugin pipeline has a
pool to filter, but nothing about the request produced that tier. The non-plugin
short-circuit already dropped it from the logged decision; the plugin path still
reported it, so a spend log claimed a classification the request never received.
Record the pool as a plugin-filtered-pool signal instead.
Also name the real problem when the resolved tier has no models at all: that raised
"No candidate models left after routing-plugin filtering" and sent operators hunting
for a policy plugin that never narrowed anything.
SGR has had two independent definitions. The admin UI derived it from
SpendLogs, so it counted what litellm's logging callbacks observed and could
attribute and price. BillableRequestMetricsMiddleware counted what the proxy
actually answered at the ASGI edge, but only exported to OTLP for enterprise
metering. The two disagree by design in places, and the SpendLogs figure goes
quiet whenever spend logging is disabled or the callbacks are bypassed.
This adds LiteLLM_DailyGatewayRequests, written by the middleware, and points
the dashboard's Successful Requests tile at it.
Requests fold into an in-memory map at record time rather than going through a
queue like the spend path. A count is a pure aggregate, and every dimension of
the key is chosen by the proxy from a closed set: the date, the category, and a
route that the classifier maps to one of a fixed list of strings rather than
passing the raw path through. Nothing a caller sends can add a key, so the fold
and the table are bounded by (days x categories x routes) however much traffic
arrives; the spend queue blocks once full, which is not acceptable in the
response path. A scheduler job drains it on the existing batch interval, and a
failed flush merges its counts back so a database blip undercounts nothing.
The middleware previously returned early when no billing recorder was
injected, which is the unlicensed case. The new sink is not license-gated, so
that early return now requires both sinks to be absent. The billing recorder
keeps its 2xx-only gate; the sink takes every status so failed_requests is
real. The sink is not told which deployment served the request, unlike the
billing recorder. That id is a sha256 over litellm_params, credentials
included, so a caller who puts a credential in the request body mints a fresh
one per distinct value. No configuration is needed for that: api_base and
base_url are on _BANNED_REQUEST_BODY_PARAMS and need allow_client_side_
credentials, but api_key is not on that list, and both reach the same
_handle_clientside_credential branch. The read endpoint aggregates the
dimension away regardless, so the key is better off without it.
The new table carries no key, user or team dimension, so /gateway/daily/activity
is restricted to proxy admin roles and the per-key and per-model breakdowns
keep reading the daily spend tables. The old path is left running and marked
with TODOs.
A fetched result carries the range key it was fetched for, and the render
selects it only when that key matches the range on screen. Both the gateway
counts and the spend aggregate go through that rule: the request tiles read the
first and fall through to the second, so stamping only one of them would leave
the tile showing a superseded range by the other route.
The paginated pages behind that aggregate are reached through a failure flag,
so the flag is stamped too. A flag left over from the previous range would let
those pages through while a new range is in flight, which is the same defect
one fallback further down.
`_setup_prisma_client` ran `connect()`, then a `SELECT 1` health check, then
armed the DB health watchdog. Any failure fell into one handler that, with
`allow_requests_on_db_unavailable` set, swallowed the error and returned None,
which the caller assigns to the module-level `prisma_client`. A single
transient timeout on that health check therefore discarded a client that had
already connected, for the life of the process, and skipped the watchdog that
exists to reconnect it.
The watchdog now starts before the health check, and a swallowed post-connect
failure returns the connected client instead of None. A client whose
`connect()` failed is still discarded, and startup still hard-fails when
`allow_requests_on_db_unavailable` is not set.
The same check also misreported its own failure. `health_check()` labelled its
error `disconnect()`, a copy-paste from the real `disconnect()` below it, so
grepping the logs for the health check turned up nothing and read as "the check
never ran". Both it and the sibling `connect()` failure reported through
`print_verbose`, which reaches `verbose_proxy_logger.debug` and otherwise prints
only under the deprecated `litellm.set_verbose`, leaving a startup-blocking
database fault invisible at the verbosity operators actually run. Both now log
at warning under their own names. The proxy logger's handler carries the secret
redaction filter, so a connection string in the exception text is redacted
exactly as it was on the old print path.
ContentFilterGuardrail implements apply_guardrail, which is everything the
generic post_mcp_call_hook machinery needs to scan an MCP tool result before
it reaches the model, but post_mcp_call was missing from
get_supported_event_hooks. _validate_event_hook rejects any mode outside that
list, so a config with `mode: post_mcp_call` failed proxy startup with
"Event hook GuardrailEventHooks.post_mcp_call is not in the supported event
hooks" instead of scanning tool output.
Declaring the hook makes the indirect-prompt-injection case enforceable: an
MCP fetch tool returns a page whose body carries "IGNORE ALL PREVIOUS
INSTRUCTIONS ...", and the gateway blocks the result rather than handing it
to the model.
The gate only matched os.getenv(, litellm.get_secret( and
litellm.get_secret_str(, so a bare get_secret_bool("X") matched nothing and
the key bypassed the documentation requirement entirely. Add a fourth pattern
for get_secret_bool, with or without the litellm. prefix, and a negative
lookbehind so an unrelated receiver's .get_secret*( call is not mistaken for
an env var read.
Extraction and table parsing move into functions behind a __main__ guard so
the patterns can be unit tested; the script is still invoked exactly the same
way by CI.
This surfaces 13 keys the gate never checked, 8 of which have no reference
row yet.
* fix(proxy): give proxy_admin_viewer read parity with proxy_admin
Route-level checks already default-allow management GETs for the viewer
role, but ~15 handlers compared user_role to PROXY_ADMIN only, dropping
viewers into regular-user scoping (/key/list, /user/info, /model/info,
guardrails, prompts, agents, memory, workflows, MCP catalog, coordination
redis settings, credential migration check, enterprise projects). Swap
those read paths to user_api_key_has_admin_view; write gates unchanged.
The dashboard now presents the viewer session as Admin for all gating
(effectiveSessionRole) so every page fetches with admin visibility, with
userRoleLabel/isViewOnly preserving the account-menu label and the
playground cost guard. The server remains the write authority.
* refactor(agents): remove side-effectful health_check param from GET /v1/agents
Addresses a security review finding on the admin viewer read parity change:
listing agents with health_check=true made the proxy issue a server-side GET
to every agent URL, so a read-scoped caller could trigger request fan-out
beyond their object permissions. The list endpoint is now a pure read for
every role.
Removes the query param, the URL probing helper and its timeouts, the
AgentHealthCheck httpx provider tag, and the dashboard's Health Check
toggle. Requests still passing health_check=true get the full list back
with the param ignored.
* fix(proxy): keep credential encryption check proxy_admin only
The residual scan behind GET /credentials/migrate-encryption/check loads
every model, credential, MCP, team, and verification-token row and runs a
decryption attempt on each stored value. Extending it to proxy_admin_viewer
let a read-only account repeatedly trigger deployment-wide scans, so the
route keeps its original full-admin gate.
* fix(agents): restore health_check, keep list fast path proxy_admin only
Restores the agent health_check feature exactly as before this PR: the
query param, the URL probing helper, the httpx provider tag, and the
dashboard toggle all return, so existing callers keep the filtering
contract. The viewer expansion is instead reverted at its source: the
GET /v1/agents admin fast path stays PROXY_ADMIN only, so a
proxy_admin_viewer goes through the object-permission scoped branch as
before and cannot fan out health checks beyond their allowlist. The
viewer read of a single agent stays viewer-inclusive since it has no
side effects.
Requests migrated from NIM/vLLM servers carry extras that flow through the
extra_body passthrough verbatim, but the Fireworks chat completions API
either names them differently or does not accept them at all. Add
FireworksAIConfig.map_extra_body_params, invoked from the fireworks chat
dispatch, which renames truncate_prompt_tokens to prompt_truncate_len,
maps chat_template_kwargs.enable_thinking to reasoning_effort, converts
guided_json/guided_grammar/guided_choice to response_format, and drops
the remaining extras (min_tokens, stop_token_ids, skip_special_tokens,
guided_regex, etc.) with a debug log. Alias and competing-constraint
combinations raise BadRequestError. Unrecognized extras keep passing
through untouched, as do fireworks-native params like top_k.
Resolve litellm/types/google_genai/main.py and litellm/types/utils.py by
keeping this branch's modernized annotations on top of staging's removal
of inert type: ignore comments. Rebuild ruff-strict-budget.json from
measured merged-tree counts where de-excluding litellm/types adds
violations, keeping the stricter of the two sides' limits everywhere
else so no rule gains headroom. Fix the four type-discipline additions
the merge surfaced: freeze GEMINI_1_5_ACCEPTED_FILE_TYPES, drop a
callback_args parameter rebind in guardrails, and give the two remaining
mutations reasoned suppressions
