* ci: run the unit_selection.sh shard files on every event instead of only fork pull requests
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* ci: rename fork-flag to unit-flag now that it applies on every event
* test: move tests/test_litellm root and small trees into tests/unit
Pure renames, no content changes. Follow-up commits in this PR fix
references, merge the three files that already existed in tests/unit,
keep live-provider tests in tests/test_litellm and wire CI.
* test: carry tests/test_litellm conftest isolation into tests/unit
Callback lists, routing fallbacks, cached HTTP clients, logger state, AWS,
proxy-URL and keychain env, and session-end client cleanup now reset for
unit tests too. The environment isolation owns its MonkeyPatch so a test's
own monkeypatch is undone before the model-cost teardown runs.
* test: merge, split and prune the moved root and small-tree tests
Merge batches/test_batch_utils.py and the chat_completions and messages
dispatch tests into the files that already existed in tests/unit. Keep
the live Gemini interactions tests, the async image-fetch format test and
the OpenAI embedding scorer test in tests/test_litellm since they need
real network or keys. Put test_router.py under tests/unit/test_router so
the existing package no longer shadows it. Delete eight tests the audit
found superseded by stronger ones kept in this move.
* ci: run the moved root and small-tree tests under their legacy flags
Add the misc and responses-caching-types flags to unit_selection.sh and
CircleCI, extend enterprise-routing and mcp-integration, and point the
legacy GHA shards, Makefile, redis-compat workflow, merge smoke manifest
and change classifier at the new paths.
* test: make the new tests/unit directories packages
tests/unit/test_package_layout.py requires every directory to carry an
__init__.py, and without one the moved and retained
test_litellm_responses_bridge.py modules collide on import.
* test: scope the unit socket block to tests/unit in shared sessions
The GHA shards collect the legacy test-path and the unit selection in one
pytest session. The unit conftest's loopback-only block leaked into legacy
modules that reach the network at import. The legacy conftest now lifts the
restriction at collect and setup time, and the unit conftest re-applies it
when collecting its own modules.
* test: move tests/test_litellm/llms into tests/unit/llms
Rename-only. Moves the provider tests and the fine-tuning fixtures they
load, mirroring the old paths. Follow-up commits merge, split and wire them.
* test: merge, split and prune the moved llms tests
Merges the Databricks chat transformation tests into the existing unit
file, keeps the tests that need real keys or the network in
tests/test_litellm, deletes the audited tests a stronger unit test
already covers, and points imports at tests.unit.llms.
* ci: run the moved llms tests under their legacy flags
The Vertex AI and All Other Providers shards keep their legacy test-path
for the retained files and add the llm-vertex-ai and llm-other-providers
unit selections. CircleCI gets matching unit jobs.
* test: make the tests/unit/llms directories packages
Adds __init__.py to the moved dirs and drops the legacy ones whose
directories no longer hold tests.
* test: drop script runners and path hacks the llms split left dangling
The __main__ runners in the split openai_like files and the Databricks e2e
runner called tests that now live in the other half of the split or were
deleted. The retained legacy halves also no longer need sys.path edits.
* test: give the shard-script tests their own GITHUB_OUTPUT
They only passed where the runner set it. The CircleCI unit job's env
allowlist drops it, so the script's redirect failed there.
* test: point the router and module-deletion checks at tests/unit
router_code_coverage and code_qa_check_tests only searched tests/test_litellm,
so the moved router tests no longer counted. The two silent-experiment tests
the audit deleted were the only direct callers of those methods; they are
replaced with tests that assert the forwarded shadow request and the
recursion guard.
* test: move tests/test_litellm integrations and secret_managers into tests/unit
Rename-only. Mirrors the old paths, including the directory conftests
and the prompt and JSON fixtures. Follow-up commits prune and wire them.
* test: prune and repoint the moved integrations tests
Deletes the 7 audited tests a stronger test in the same tree already
covers, imports the TLS sink helpers from their new conftest path, and
restores os.environ after each integrations test. Some presets write
OTEL_EXPORTER_OTLP_HEADERS straight into os.environ, and without the
legacy tree's test ordering that header leaked into the AgentOps tests.
* ci: run the moved integrations tests under their legacy flag
The integrations GHA shard and a new CircleCI job run the integrations
unit selection. secret_managers joins the misc selection.
* docs: point integrations and secret_managers references at tests/unit
* test: make the moved integrations directories packages
* test: keep the Databricks manual e2e runner and fix the SageMaker Nova run path
The Databricks e2e file is a manual script whose main() calls the tests
that were pruned, so pruning them broke the documented run. It is back to
its main version. The SageMaker Nova docstring now points at the file's
real location in tests/local_testing.
* test: keep the job's UNIT_FLAG out of the shard-script tests
---------
Co-authored-by: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* fix(spend): attribute CLI session spend to the per-user cli-session alias instead of the hashed session token
A CLI session token is a fresh random secret on every login, so since v1.99 each
login's spend rows carried a different sha256 hash as api_key and the usage APIs
could resolve neither key_alias nor user_email for them. Spend rows and logging
callbacks now attribute a session request to its stable alias,
cli-session-<user_id>, and the usage endpoints derive that alias and owner from
the key itself instead of scanning for a matching digest
* fix(spend): resolve the CLI session team from the user's first team in usage metadata
A cli-session key carries no team of its own in the DB, so the usage
breakdown showed team_id None for it and the export grouped it as
Unassigned. The login attaches the user's first team to the session, so
the recovery mirrors that rule for cli-session keys only.
* fix(spend): claim the session team only for a single-team user
The CLI login attaches a team on its own only when the user has exactly
one; a user in several teams picks one per login, so usage metadata for
the alias would otherwise name a team the login may not have used.
* test(pass_through): mark the mocked auth object as a plain key
The logged key follows the alias only for a session token; a bare
MagicMock reads as one, so the test names the field it relies on.
* fix(spend): attribute CLI session pass-through, queue, and managed batch spend to the cli-session alias
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* fix(spend): only treat the exact cli-session-<created_by> value as a batch key alias
A managed object row written by an older build can still carry the raw per-login
session token, which shares the cli-session- prefix. Matching on the prefix alone
would have surfaced that token as a trusted alias and persisted it verbatim in the
batch cost spend log, so the alias check now requires the exact per-user value and
every other prefixed value keeps going through redaction
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* fix(spend): log proxy executed batch rows under the cli-session alias instead of the session token
_row_metadata set user_api_key from the raw bearer token while user_api_key_hash carried the alias, so the spend log redaction rejected the alias as untrusted and hashed the random session token instead
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* fix(spend): attribute semantic search embedding spend to the cli-session alias
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* fix(spend): scope /key/spend/report for a CLI session to the cli-session alias
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* fix(spend): use the cli-session alias for websearch spend, prometheus failure labels and the parallel limiter
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* refactor(spend): drop explanatory docstrings on get_logged_api_key and attach_user_details
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* fix(spend): only recover cli-session usage keys whose suffix is a known user
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
---------
Co-authored-by: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* fix(websearch_interception): propagate parent session/trace ids into intercepted searches
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* fix(websearch_interception): let parent correlation win over configured search params and type test params
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* test(e2e): bill an intercepted web search under the parent request session
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* test(e2e): drop unrelated reformatting from the websearch session harness change
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* ci(e2e): keep the websearch interception session suite out of the stage-mirror gate
The stage-mirror stack runs no websearch_interception callback or search tool, so the suite is deselected there and the changed-tests gate fails on a file that executed nothing
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* ci(e2e): run the websearch interception session suite on the stage-mirror stack
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* docs(e2e): leave CONTRIBUTING.md untouched
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
---------
Co-authored-by: shivam <shivam@berri.ai>
Co-authored-by: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
Resolves handler.py against main's SearchOutcome refactor: the file is
main's version plus this PR's substantive hunks only (the RichWebSearchInput
import, the rich= wiring at the three _execute_search call sites, the
_rich_search_input and _provider_supports_rich_search helpers, and the
_execute_search forwarding), so the 88-column re-wrap noise the PR carried
is gone and the diff against main is the feature alone. RichWebSearchInput
sits beside main's new SearchSucceeded/SearchFailed types, and main's two
_execute_search test stubs accept the new rich argument.
CI's lint gate checks ruff format, not black; black's output differs on
a few line splits. No logic changes.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
- RichWebSearchInput fields are ReadOnly and constructed literally
- the pyright suppression now states why the str provider name is safe
- new tests drive _build_anthropic_request_patch and
_build_chat_completion_request_patch end to end so the tool-call ->
_rich_search_input wiring is covered, not just _execute_search
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
The intercepted web search tool only carries a single query string, so
search providers whose APIs take a natural-language objective plus
multiple keyword queries (documented best practice for Parallel AI's v1
search) always receive a degraded single-query request.
Widen the tool's input schema with optional objective and search_queries
fields (query stays required), and forward the richer shape from the
interception handler only to providers whose search config reports
supports_rich_search_input(). Every other provider, and every model that
keeps emitting just query, is byte-for-byte unchanged.
- BaseSearchConfig.supports_rich_search_input() defaults False;
ParallelAISearchConfig overrides True
- handler trims search_queries to five (the provider cap) and never
overrides an objective configured on the search tool's litellm_params
- mocked tests cover schema exposure, extraction validation, provider
gating, and the unchanged single-string path
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
This reverts commit 65a46a5f32 (#38113)
That change made two edits that combine into a hard failure for SDK
users. It dropped the null-router guard in
_select_search_tool_from_router, so a missing router now yields an empty
search_tools list instead of returning early, and it turned the no-match
case in _select_search_tool_from_list from a debug-logged fallback into
a raised ValueError. It also added a call site in
async_pre_call_deployment_hook that invokes the selection purely for the
side effect of raising, discarding the return value
Used together, any SDK caller that sets search_tool_name and sends a web
search tool now raises "Configured search tool '<name>' was not found"
on every request. There is no way to satisfy the check off the proxy,
because search_tools is only ever populated from the proxy router, so
the SDK path cannot register one
tests/pass_through_unit_tests/test_websearch_interception_e2e.py caught
this, but #38113 only updated the handler unit tests, so the break
landed on staging
Reverting restores the previous behavior while we work out a fix that
keeps the stricter validation on the proxy path, where a silently
substituted search provider is the real problem worth rejecting, without
turning the SDK path into an unconditional error
The ceiling used to go through `int(... or 3)`, so anything `int()` accepted
worked. Tightening the new shared validator to `isinstance(int)` turned a
config that boots today into a proxy that refuses to start, because
`max_agentic_loops: os.environ/MAX_AGENTIC_LOOPS` is resolved to a string
before it reaches either check, and a YAML-quoted "5" is a string too.
Accept ints, integral floats, and strings that parse to a whole number. Keep
refusing bools, fractional floats, words, and anything below 1.
The ceiling was only checked at the feature level, on
litellm_settings.websearch_interception_params. The per-deployment
litellm_params.max_agentic_loops, which wins over it, went straight into
int(kwargs.get("max_agentic_loops", 3) or 3), so a 0 was swallowed by the
falsy fallback and read as the default 3. Asking for the tightest ceiling
handed you the loosest one. A non-integer booted the proxy and then failed
every request to that model with "invalid literal for int() with base 10".
Both settings now share one validator, which names the field it rejected,
and the per-deployment value is checked while the model list is read at
startup so a bad value stops the proxy rather than surfacing per request.
The check sits in load_config rather than on LiteLLM_Params because the
proxy builds its router with ignore_invalid_deployments=True, where a
validation error drops the deployment silently instead of refusing to
start. This is the same placement the complexity_router_config plugin
check already uses.
Chat completions read the same key through a separate path that turned 0
into 1 and true into a ceiling of 1, so it now shares the validator too
and the key means one thing on both surfaces.
A capped turn on a streaming request is rebuilt into SSE by
FakeAnthropicMessagesStreamIterator. It emitted content_block_stop for
every block but content_block_start only for text, thinking,
redacted_thinking and tool_use, so a web search turn's server_tool_use
and web_search_tool_result blocks produced stops with no matching start.
Anthropic's SDK accumulator appends on content_block_start and then
indexes content[event.index] on content_block_delta, so the orphan stops
shifted every later index and client.messages.stream() raised IndexError
on the text block. Unknown block types now pass through with a start of
their own, which keeps position equal to index.
Also corrects two claims that said no current caller reaches the loop
with stream=True. AgenticStreamingIterator does, and it keeps raising,
because its events are already on the wire.
When the bounded loop cap or the repeated tool-call fingerprint guard refused a
rerun, the raise escaped the parent agentic frame and the client got the raw model
turn back: HTTP 200 carrying an unresolved tool_use block for the internal
litellm_web_search tool and stop_reason "tool_use". The client never declared that
tool, so it had no way to answer it and the conversation could not continue
The safety check now raises AgenticLoopSafetyError, a ValueError subclass, and
_call_agentic_completion_hooks catches it and returns a finalized response: the
blocks belonging to the refused tool calls are dropped, and stop_reason is closed
out to end_turn when nothing the client declared is still waiting. Refused blocks
are matched by the ids and names of the tool calls the rail refused rather than by
hardcoding the web search tool name
Only the non-streaming anthropic messages path ends the turn this way. A streaming
caller has already sent the original message by the time the hooks run, so a
finalized turn would arrive as a second message rather than replace the first, and
the responses surface carries a pydantic model this finalizer does not rewrite.
Both keep raising, exactly as they did before
Also adds max_agentic_loops to websearch_interception_params so the ceiling can be
set once for the whole feature. A per deployment litellm_params.max_agentic_loops
still wins over it, and the field stays on the proxy's untrusted root list so a
client cannot raise its own ceiling
* fix(websearch): restore snippet text in native web_search_tool_result blocks (LIT-5315)
The build_web_search_tool_result_block method copied url/title/page_age but
hardcoded encrypted_content to empty string, never reading SearchResult.snippet.
This left every native block content-free, forcing clients to web_fetch each
result to recover evidence—the reported symptom.
The Anthropic spec carries page text only in encrypted_content (an opaque
server-issued blob we cannot mint), so snippet is emitted as an additive key
alongside the spec fields. encrypted_content stays empty rather than holding
plaintext, which would assert encryption semantics that don't hold.
The anthropic SDK's BaseModel sets extra='allow', so the additive snippet key
survives SDK parsing. litellm has no typed model for web_search_result at all,
so nothing drops it internally. Turn-2 replay behavior is unaffected: the
empty encrypted_content already exists today.
Tests:
- Updated test_shape_with_results to assert snippet present
- Added test_snippet_carried_for_every_result to cover multi-result ordering
- Added test_missing_snippet_degrades_to_empty_string for edge case
- Mutation check: reverting source-only yields 3 test failures, restored to 117 passed
Fixes: LIT-5315
Co-Authored-By: Claude <noreply@anthropic.com>
* fix(websearch): make synthesized web_search blocks replayable by native clients
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* fix(websearch): flatten a resultless replayed search block so Bedrock accepts the next turn
The flatten added for LIT-5315 bails when the replayed web_search_tool_result
carries an empty content list, but that is exactly what the interceptor emits
when a search legitimately returns nothing and when a search raises. The block
survived into the outbound body, Bedrock rejected the tag, and the conversation
died on the following turn just as it did before the flatten existed.
An empty content list has no encrypted_content to respect and no evidence to
preserve, so it flattens safely, and its paired server_tool_use goes with it.
The rendered text now says so explicitly rather than emitting a bare header.
Adds the multi-turn replay coverage that existed nowhere: the outbound Bedrock
invoke body is asserted free of both block types, parametrized over the
results-present and resultless cases, and built from the interceptor's own
builder so the fixture cannot drift from what it emits.
Resolves LIT-5320
* test(websearch): pin flatten idempotency for the agentic-loop re-entry
The agentic loop re-enters the same /v1/messages entry point for its follow-up
call and hands it the original client history, so the flatten runs again over
already-flattened messages once per iteration. Bedrock always takes that path,
since its config reports web search as natively handled and the short-circuit
is skipped.
A pass that appended the rendered text instead of replacing the block would
duplicate the evidence on every iteration and re-ship the unsupported tag, and
no existing single-pass test sees it. Mutation checked: keeping the original
block alongside the rendered text fails this test on its own.
---------
Co-authored-by: Claude <noreply@anthropic.com>
Co-authored-by: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
Co-authored-by: Yassin Kortam <yassin@berri.ai>
An intercepted web search called litellm.asearch() with only the search tool's litellm_params, so the search request carried no owner. The proxy's spend hook skips any call with no key, user or team attached, so the search's provider cost never reached SpendLogs; it was missing from the Logs page and never counted against the caller's budget. The same path never ran the rate limiter either, so an intercepted search was free of the key's RPM/TPM limits.
The search now carries the originating key's attribution metadata (key hash, alias, user, team, org, plus model_group set to the resolved search tool) and runs the caller's rate limit checks before hitting the provider, matching what a direct /v1/search request gets. SDK calls with no proxy auth context are unchanged.
Resolves LIT-5033
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* fix(websearch): wire chat completion agentic loop to correct hooks
maybe_run_chat_completion_agentic_loop was calling async_should_run_agentic_loop (Anthropic format) and async_run_agentic_loop (Anthropic path) instead of the chat-completion variants. This meant WebSearchInterceptionLogger never intercepted chat completion requests — the LLM returned a litellm_web_search tool_call but the agentic loop never executed, so the raw tool_calls response was returned to the caller.
Fix: gate on async_should_run_chat_completion_agentic_loop override, call that hook and async_build_chat_completion_agentic_loop_plan / async_run_chat_completion_agentic_loop in the execution path.
Regression test added.
* fix(websearch): strip tool_choice from follow-up request
When the original request forces tool_choice to litellm_web_search,
the follow-up request after search execution inherited that tool_choice,
causing the model to call the search tool again instead of synthesizing
an answer from the results.
* fix(websearch): inject api_key into agentic hook kwargs for anthropic messages
Follow-up calls inside async_run_agentic_loop (e.g. websearch interception's
synthesis call after executing Exa/Perplexity searches) were missing api_key
because the named api_key param in async_anthropic_messages_handler was never
merged into the kwargs dict forwarded downstream. Result: every /v1/messages
websearch follow-up failed with "x-api-key header is required" and the caller
received the raw tool_use response instead of the synthesized answer.
* ci: trigger CI run
* fix(websearch): support unified agentic hooks alongside chat-completion-specific hooks
CodeInterpreterInterceptionLogger uses async_should_run_agentic_loop with
_agentic_loop_api_surface to handle both surfaces from one hook. The chat
completion loop must also check _gate_overridden so callbacks using the
unified hook pattern still fire for chat completions.
* fix(websearch): strip tool_choice from legacy chat completion follow-up call
The _execute_chat_completion_agentic_loop path merged original optional_params
(which includes forced tool_choice) into follow-up params without explicit
removal. _build_chat_completion_request_patch already excluded tool_choice from
its optional_params output, but dict.update() with a missing key leaves the
original value intact. Explicit pop after the merge removes it.
* fix(websearch): always strip tool_choice from plan-path follow-up params
The tool_choice removal was gated on patch.tools is not None. WebSearch sets
tools via patch.optional_params not patch.tools, so the gate was False and
forced tool_choice from the original request survived into the synthesis call.
Move the pop outside the patch.tools branch so it applies unconditionally.
failing test is not related to the pr
* fix(websearch): sync tool_choice when converting web_search tools
Claude Code forces native web search via tool_choice pointing at web_search
while websearch_interception renames the tool to litellm_web_search, causing
Anthropic 400s. Forward tool_choice into pre-request hooks and rewrite forced
tool_choice to match the converted tool name.
Co-authored-by: Cursor <cursoragent@cursor.com>
* fix(websearch): re-wrap agentic loop responses as SSE for streaming clients
When websearch interception converts stream=true to false for the agentic
loop, dict responses from the loop were returned as application/json even
though the client requested SSE. Wrap those responses in
FakeAnthropicMessagesStreamIterator so /v1/messages streaming callers
(e.g. Claude Code) receive text/event-stream after search completes.
Fixes#27721
Co-authored-by: Cursor <cursoragent@cursor.com>
* test(websearch): cover tool_choice sync and post-loop SSE wrap; fix UP006
Add regression tests for both websearch interception fixes: _sync_forced_tool_choice
repointing a forced web_search tool_choice to litellm_web_search (the 400 fix) and
_maybe_websearch_fake_stream_wrap re-wrapping agentic loop dict responses as SSE for
streaming clients (#27721). Switch the new helper annotations to builtin dict/list so
the ruff UP006 strict-rule ceiling stays within budget.
Co-authored-by: Cursor <cursoragent@cursor.com>
* fix(websearch): resolve merge conflict and unify fake stream wrapping
Remove the duplicate _maybe_websearch_fake_stream_wrap helper left by a bad merge that caused a SyntaxError in CI, and route all call sites through _maybe_wrap_in_fake_stream instead.
Co-authored-by: Cursor <cursoragent@cursor.com>
---------
Co-authored-by: Shivam Rawat <shivamrawat@Shivams-MacBook-Pro.local>
Co-authored-by: Cursor <cursoragent@cursor.com>
Co-authored-by: Shivam Rawat <shivamrawat@Shivams-MBP.localdomain>
* fix(websearch): wrap agentic loop response in fake stream for streaming requests
When websearch_interception converts stream=True to stream=False internally,
the agentic loop returns a plain dict. Previously this dict was returned
directly to the client expecting SSE events, resulting in empty streams.
Added _maybe_wrap_in_fake_stream() which checks the
websearch_interception_converted_stream flag and wraps dict responses in
FakeAnthropicMessagesStreamIterator. Applied to all return paths in
_call_agentic_completion_hooks:
- async_run_agentic_loop (legacy path)
- _execute_anthropic_agentic_plan (plan-based path)
- plan.response_override
- plan.terminate
Includes unit tests for _maybe_wrap_in_fake_stream().
* test(websearch): cover agentic-loop wrap paths; gate fake-stream on anthropic_messages surface
Guard _maybe_wrap_in_fake_stream on api_surface == anthropic_messages so the
responses API surface is never wrapped in an Anthropic SSE iterator, and type
logging_obj as Optional to match the None call sites. Adds regression tests
that drive the legacy, response_override, and terminate return paths of
_call_agentic_completion_hooks end to end.
* test(websearch): cover _execute_anthropic_agentic_plan and tail wrap paths
Drives the remaining two fake-stream return paths of
_call_agentic_completion_hooks (the _execute_anthropic_agentic_plan branch via
a stubbed handler, and the tail path when no agentic loop runs) so every
converted-stream return path is regression-tested.
---------
Co-authored-by: Clawd <fffff.c@gmail.com>
* fix(datadog): pass callback_specific_params so DatadogCostManagementLogger receives cost_tag_keys (#29590)
* fix(datadog): pass callback_specific_params so DatadogCostManagementLogger receives cost_tag_keys
* test(proxy): regression test that load_config forwards callback_specific_params
* fix(proxy): guard lakera_prompt_injection callback_specific_params against non-dict
Addresses review feedback: forwarding callback_settings as callback_specific_params
(so DatadogCostManagementLogger receives cost_tag_keys) exposed the
lakera_prompt_injection branch, which did lakeraAI_Moderation(**callback_specific_params
["lakera_prompt_injection"]) with no type guard. A config like
`callback_settings: {lakera_prompt_injection: "any-string"}` then hit `**"any-string"`
-> TypeError: argument after ** must be a mapping, not str.
Guard the lakera branch with isinstance(dict), matching the existing presidio and
datadog_cost_management branches (non-dict values fall back to {}). Add a regression
test asserting initialize_callbacks_on_proxy ignores a non-dict value instead of crashing.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* test: inject fake lakera_ai module to avoid importing the real one
CI fix for the lakera regression test: it stubbed litellm.proxy.proxy_server with
a SimpleNamespace and then monkeypatch.setattr'd the real lakera_ai module, which
forces importing it — and lakera_ai does `from litellm.proxy.proxy_server import
LiteLLM_TeamTable`, absent on the stub -> ImportError under proxy-infra tests.
Inject a fake lakera_ai module into sys.modules instead, so the callbacks branch's
`from ...lakera_ai import lakeraAI_Moderation` resolves to the stub without loading
the real module. The guard under test (isinstance(dict) in the lakera branch) is
unchanged.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* fix(callbacks): guard compression/websearch interceptors against non-dict callback_settings (#30153)
#29590 forwards the full callback_settings dict into initialize_callbacks_on_proxy, which activates the compression_interception and websearch_interception consumers. Their initialize_from_proxy_config read the callback_settings subkey without an isinstance(dict) guard, so a non-dict value such as `compression_interception: true` reached from_config_yaml(...).get(...) and aborted proxy startup with AttributeError. #29590 added that guard for lakera_prompt_injection but not for these two
Mirror the isinstance(dict) guard already used by the lakera, presidio, and datadog branches so a non-dict value is ignored and the callback initializes with defaults. A parametrized test feeds every callback_settings consumer a non-dict value through initialize_callbacks_on_proxy to catch a future consumer that forgets the guard
* fix(callbacks): normalize non-dict callback_specific_params to empty dict
A blank callback_settings: key in YAML loads as None, and
config.get('callback_settings', {}) returns None because dict.get only
falls back to the default when the key is absent. Forwarding that value
verbatim to initialize_callbacks_on_proxy made the first
'<name>' in callback_specific_params membership test raise
TypeError: argument of type 'NoneType' is not iterable, aborting proxy
startup. Same failure for any non-dict root such as callback_settings: true.
Normalize the value at the function boundary so both callsites (and any
future ones) initialize callbacks with their defaults instead of crashing.
---------
Co-authored-by: Hedi Daoud <150018939+hdaoud23@users.noreply.github.com>
Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* feat(custom_logger): add async_post_agentic_loop_response_hook
Lets a CustomLogger shape the response returned by the agentic-loop
follow-up call without bypassing the loop's safety / observability
machinery (depth tracking, fingerprinting, etc.). Default returns the
response unchanged.
Used by websearch_interception to inject Anthropic-native
web_search_tool_result blocks when the originating client requested a
native web_search_* tool.
* feat(llm_http_handler): call post-agentic-loop hook on the originating callback
In _execute_anthropic_agentic_plan, after anthropic_messages.acreate
returns, call the originating callback's
async_post_agentic_loop_response_hook so it can mutate the final
response (e.g. inject native tool_result blocks). Pass the callback
through from _call_agentic_completion_hooks.
Exceptions in the post-hook are caught and logged so a buggy callback
can't kill the request.
* feat(websearch_interception): add is_anthropic_native_web_search_tool
Identifies tools the Anthropic-native clients (Claude Desktop, the
Anthropic SDK, the Anthropic Console) use to request native search:
type starts with "web_search_" (e.g. web_search_20250305). Rejects the
LiteLLM standard tool, the OpenAI-function variant, the bare
"WebSearch" legacy name, and the bare "web_search" Claude Code shape.
This lets us decide per-request whether the client expects
web_search_tool_result content blocks in the response, without
renaming any existing constants or touching native-provider skip
logic.
* feat(websearch_interception): add build_web_search_tool_result_block
Produces the Anthropic-native web_search_tool_result content block
from a structured SearchResponse. Anthropic-native clients use this
block to populate citations / source links — the existing text-blob
flatten path only feeds readable evidence to the model and discards
the structure, so this builder gives us the missing piece.
Shape matches https://docs.anthropic.com/en/api/web-search-tool —
web_search_result items carry url, title, page_age, encrypted_content
(empty string when the search provider doesn't supply one).
* feat(websearch_interception): emit native web_search_tool_result blocks
When the originating client request carried a native Anthropic
web_search_* tool, the final response now also carries
web_search_tool_result content blocks alongside the model's text
answer — so Claude Desktop / Anthropic SDK clients can populate the
citations panel and replay conversation history with structured search
evidence.
Wiring:
- Pre-request hooks (both deployment + Anthropic path) set a flag on
kwargs when they see a native web_search_* tool, so the signal
survives the conversion-to-litellm_web_search step regardless of
which hook fires first.
- _execute_search now returns (text, SearchResponse) so the structured
results aren't lost when the text is flattened for the follow-up
model call.
- _build_anthropic_request_patch returns the parallel list of
SearchResponse objects.
- async_build_agentic_loop_plan pre-builds the web_search_tool_result
blocks (one per tool_use_id) and stashes them on plan.metadata when
the flag is set.
- async_post_agentic_loop_response_hook reads the metadata and
prepends the blocks to response.content.
- _execute_agentic_loop mirrors the injection for the legacy path so
both paths behave identically.
Clients that send the LiteLLM standard tool keep the existing
text-only behavior — no regression.
* test(websearch_interception): cover native web_search_tool_result emission
18 tests across:
- detector branches (native vs litellm-standard, OpenAI-function shape,
Claude Desktop builtin WebSearch, bare web_search, missing type)
- block-builder shape (results, none, empty)
- pre-request hook flag-setting (native sets, standard does not)
- async_build_agentic_loop_plan attaches blocks to plan.metadata when
the flag is present, leaves metadata untouched when absent
- post-hook injection into dict and object responses
- legacy _execute_agentic_loop mirrors the injection so both paths
return the same shape
* test(websearch_short_circuit): keep _execute_search mocks in sync with new tuple return
* test(websearch_thinking_constraint): keep _execute_search mocks in sync with new tuple return
* feat(websearch_interception): emit native blocks from try_short_circuit_search
The agentic-loop post-hook only fires when the model returns a tool_use
block. Cowork / Claude Desktop on Bedrock actually make TWO requests
per user turn: the main /v1/messages with their builtin tool, and a
separate standalone /v1/messages whose only tool is
web_search_20250305. That second request hits try_short_circuit_search
— no agentic loop, no post-hook — and was returning text-only, leaving
the citations panel empty.
When the short-circuit input carries a native web_search_* tool, build
a synthetic server_tool_use + web_search_tool_result pair (using the
structured SearchResponse already returned by _execute_search) so the
client gets the native shape it expects. The legacy text block is
preserved so non-native short-circuit callers (Claude Code,
github_copilot, etc.) see the same payload as before.
Failure path still emits the native block pair (with empty results)
plus the text-error block, so the client gets a well-formed response
rather than a malformed half-shape.
* test(websearch_native_blocks): cover short-circuit native-block emission
Three new cases on top of the existing 18:
- native web_search_20250305 short-circuit → [server_tool_use,
web_search_tool_result, text], ids paired, urls/titles carried.
- litellm_web_search short-circuit → text-only (no regression).
- native short-circuit on search failure → still emits the native
block pair (empty results) plus the text-error block, so the client
never sees a malformed half-shape.
* test(websearch_short_circuit): index assertions by block type, not by position
Native short-circuit responses now have [server_tool_use,
web_search_tool_result, text] when the input carries
web_search_20250305 — find the text block by type rather than relying
on content[0].
* fix(websearch_interception): gate legacy WebSearch name on schema absence
Clients like Cowork / Claude Desktop ship a client-side tool named
"WebSearch" with a full input_schema — they handle it themselves and
expect to make a separate native web_search_20250305 sub-request for
the actual search.
Today is_web_search_tool matches the bare name regardless of other
fields, which hijacks the client's tool server-side. The agentic loop
fires on the main request, the model never gets to emit the
client-side tool_use, and the separate native sub-request (where
citation data flows) is never made. Net: citations panel empty.
Real Anthropic client tools always carry input_schema (the API rejects
them otherwise), so a bare {name: "WebSearch"} with no schema is the
only thing that could be a legacy interception marker. Gate the match
on schema absence: legacy callers (if any) keep working, real
client-side WebSearch tools pass through untouched.
* fix(websearch_interception): drop "WebSearch" from response-detection lists
Post-conversion the model always sees ``litellm_web_search``, so the
"WebSearch" entry in the response-side tool_use detection lists was
dead at best. If a model ever did return ``tool_use(name="WebSearch")``
it would now (incorrectly) hijack the client's own ``WebSearch`` tool
again — same Cowork problem we just fixed on the input side. Drop it.
* test(websearch_native_blocks): cover the WebSearch legacy-name schema gate
Three new cases:
- {name: "WebSearch"} (bare interception marker) → still matched
- {name: "WebSearch", input_schema: {...}} (Cowork client tool) →
passes through untouched
- {name: "WebSearch", description: "..."} (no schema) → still matched
on the assumption it's a legacy marker rather than a malformed real
client tool.
---------
Co-authored-by: Ishaan Jaffer <ishaanjaffer0324@gmail.com>
* refactor: new agentic loop event hook
simplifies how to create logic for tool based multi llm calls
* fix: compress - make it work on anthropic input as well
* fix(compress.py): working prompt compression for claude code
ensures claude code messages can run through proxy easily
* docs: add agentic loop hook guide
* docs: add agentic_loop_hook to sidebar
* fix: fix multiple arguments error
* fix: fix tool call loop for compression on streaming /v1/messages
* fix: fix linting errors
* fix: fix ci/cd errors
* feat(litellm_pre_call_utils.py): use claude code session for litellm session id
allows claude code logs to be stitched together, making it easy to know they were all part of the same conversation
* fix: suppress incorrect mypy warning rE: module
* revert: drop PR's changes to litellm/proxy/_experimental/out/
Restores the 34 HTML files under _experimental/out/ to their pre-PR
paths (X/index.html -> X.html). All renames are R100 (content
unchanged); no other files are touched.
* fix: address greptile review comments on PR #25729
- Skip ``kwargs["tools"] = []`` injection when compression is a no-op —
Anthropic Messages rejects empty tool arrays on requests that did not
originally declare tools.
- Move agentic-loop safety guards (fingerprint cycle / max depth) out of
the per-callback try/except so they propagate instead of being swallowed
by the generic exception handler. Extracted _check_agentic_loop_safety.
- Gate generic ``x-<vendor>-session-id`` capture behind the
LITELLM_CAPTURE_VENDOR_SESSION_HEADERS env var (off by default) to
preserve backwards compatibility; explicit x-litellm-* headers are
unaffected.
- Fix monkeypatch target in pre-call-hook test to patch the actual
module-level binding
(litellm.integrations.compression_interception.handler.compress).
- Add regression tests for empty-tools skip and opt-in session capture.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
* revert: drop LITELLM_CAPTURE_VENDOR_SESSION_HEADERS flag
Generic x-<vendor>-session-id header capture is a new feature and only
runs *after* the explicit x-litellm-trace-id / x-litellm-session-id
checks, so it does not change behavior for any existing caller that was
already using the LiteLLM headers — no backwards-incompatibility to gate.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
* refactor(compress): replace input_type with CallTypes call_type
Drop the bespoke ``CompressionInputType`` literal and use the existing
``litellm.types.utils.CallTypes`` enum instead. ``litellm.compress()``
now takes ``call_type: Union[CallTypes, str]`` (default
``CallTypes.completion``) — no new concept to learn, and the enum is
already the way the rest of the codebase talks about request shapes.
Supported values: ``completion`` / ``acompletion`` (OpenAI chat-completions
shape) and ``anthropic_messages`` (Anthropic structured content blocks).
Updated: compress(), the compression_interception handler, tests, docs,
and the two eval scripts.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
The deployment hook now converts stream=True→False in wrapper_async's
scope so the streaming early-return path is skipped and logging executes.
logging_obj.stream is synced after the hook, and the original stream
intent is recovered for the short-circuit path.
Made-with: Cursor
* update bedrock models in tests
* updated more tests and model_prices_and_context_window
* fix model id and pricing
* replace more sonnet models
* update tests
* git push
* update pricing
* flaky total cost
* monkey patch
* relax the cost change
* fix and revert some changes
* revert the pricing
* chore: move cost/pricing changes to bedrock-cost-fixes branch
* chore: split Bedrock file-api beta stripping to separate branch
Removes strip_unsupported_file_api_betas_for_bedrock_invoke from this branch;
see litellm_bedrock_invoke_strip_file_api_betas for that fix.
Made-with: Cursor
- Skip short-circuit for providers that have a BaseAnthropicMessagesConfig
(bedrock, vertex_ai, azure_ai, anthropic) — they use the agentic loop
which includes a follow-up LLM synthesis step. Short-circuiting would
return raw search text instead of an LLM-synthesized answer.
- Add fallback to litellm.get_llm_provider() for custom_llm_provider
derivation when litellm_params is overwritten by kwargs.
- Add test for bedrock guard.
Addresses Greptile review comments #3 and #4.
- Replace hand-rolled _extract_search_query with existing
get_last_user_message from common_utils
- Use full UUID (str(uuid.uuid4())) to match codebase convention
- Move uuid import to module level per CLAUDE.md
Addresses Greptile review feedback:
- Save original stream flag before pre-request hooks convert it, so
streaming callers get SSE events instead of a plain dict
- Propagate custom_llm_provider derived inside _execute_pre_request_hooks
when it was not explicitly passed by the caller
- Add tests covering both scenarios
For providers like github_copilot that don't natively support web search,
Claude Code's search sub-conversations were falling through to the adapter
path which strips the web_search tool and has no stream reconversion.
Instead of routing search requests through the full LLM pipeline, detect
web-search-only requests early (all tools are web_search, simple prompt)
and execute the search directly via Tavily/Perplexity, returning a
synthetic Anthropic response. No adapter, no backend LLM call needed.
Fixes#21733
When extended thinking is enabled, the websearch interception agentic loop
builds a follow-up assistant message with only tool_use blocks. Anthropic's
API requires assistant messages to start with thinking/redacted_thinking
blocks when thinking is enabled, causing a 400 Bad Request.
Extract thinking blocks from the model's initial response, thread them
through the agentic loop, and prepend them to the follow-up assistant
message — matching the pattern used by anthropic_messages_pt in factory.py.
Fixes the error: "Expected 'thinking' or 'redacted_thinking', but found
'tool_use'"
The websearch interception handler was passing internal flags like
`_websearch_interception_converted_stream` to the follow-up LLM request.
This caused "Extra inputs are not permitted" errors from providers like
Bedrock that use strict Pydantic validation.
Fix: Filter out all kwargs starting with `_websearch_interception` prefix
before making the follow-up anthropic_messages.acreate() call.
