fix: add custom_body parameter to endpoint_func in create_pass_through_route#20849
Conversation
…h_route
The bedrock_proxy_route calls `endpoint_func(custom_body=data)` to
pass a pre-parsed, SigV4-signed request body. However, the
`endpoint_func` closure created by `create_pass_through_route` does
not accept a `custom_body` keyword argument, causing:
TypeError: endpoint_func() got an unexpected keyword argument 'custom_body'
Add `custom_body: Optional[dict] = None` to both `endpoint_func`
definitions (adapter-based and URL-based). In the URL-based path,
when `custom_body` is provided by the caller, use it instead of
re-parsing the body from the raw request.
Fixes #16999
|
The latest updates on your projects. Learn more about Vercel for GitHub.
|
Greptile OverviewGreptile SummaryThis PR fixes a runtime The change fits into the proxy passthrough architecture by allowing higher-level routes (like Confidence Score: 4/5
|
| Filename | Overview |
|---|---|
| litellm/proxy/pass_through_endpoints/pass_through_endpoints.py | Adds custom_body kwarg to create_pass_through_route endpoint closures and prefers the provided body over parsed request body in the URL-based passthrough path; adapter-based path currently ignores the new parameter. |
Sequence Diagram
sequenceDiagram
participant Client
participant Proxy as Proxy(FastAPI)
participant Bedrock as bedrock_proxy_route
participant Factory as create_pass_through_route
participant Parse as _parse_request_data_by_content_type
participant PT as pass_through_request
participant Target as External Target URL
Client->>Proxy: "POST /bedrock/{endpoint}"
Proxy->>Bedrock: "bedrock_proxy_route(request)"
Bedrock->>Bedrock: "parse JSON body"
Bedrock->>Bedrock: "SigV4 sign headers/url"
Bedrock->>Factory: "create_pass_through_route(target=prepped.url, headers=prepped.headers)"
Factory-->>Bedrock: "endpoint_func"
Bedrock->>Factory: "endpoint_func(custom_body=data)"
alt "URL-based target"
Factory->>Parse: "parse query/body from raw request"
Factory->>Factory: "if custom_body != None use it"
Factory->>PT: "pass_through_request(custom_body=final_custom_body)"
PT->>Target: "HTTP request"
Target-->>PT: "response"
PT-->>Bedrock: "response"
else "Adapter-based target"
Factory->>Proxy: "chat_completion_pass_through_endpoint(...)"
Note over Factory,Proxy: "custom_body currently not forwarded"
Proxy-->>Bedrock: "response"
end
Bedrock-->>Client: "response"
Additional Comments (1)
In the adapter-based branch, |
Review FeedbackThanks for the fix — this addresses a real Greptile's finding is correct: adapter-based path silently ignores
|
|
You're right — Current state:
Options:
Since Bedrock targets currently resolve to URLs (not adapters), option 1 is the safest short-term fix. But if there's any chance adapter targets will need |
|
@themavik can you please make the requested changes? @shin-bot-litellm for monitoring |
Address reviewer feedback on PR #20849: - Document why the adapter-based endpoint_func accepts custom_body for signature compatibility but does not forward it (the underlying chat_completion_pass_through_endpoint does not support it). - Add test_create_pass_through_route_custom_body_url_target: verifies that when a caller (e.g. bedrock_proxy_route) supplies custom_body, it takes precedence over the body parsed from the raw request. - Add test_create_pass_through_route_no_custom_body_falls_back: verifies that the default path (no custom_body) correctly uses the request-parsed body, preserving existing behavior. Both tests are fully mocked following the project's CONTRIBUTING.md guidelines and the patterns established in the existing test file. Co-authored-by: Cursor <cursoragent@cursor.com>
|
Addressed the review feedback:
|
|
@greptile-apps review this PR please |
Greptile OverviewGreptile SummaryFixes a
Confidence Score: 4/5
|
| Filename | Overview |
|---|---|
| litellm/proxy/pass_through_endpoints/pass_through_endpoints.py | Added custom_body parameter to both adapter-based and URL-based endpoint_func closures, and updated body selection logic to prefer caller-supplied body over request-parsed body. Fixes TypeError crash on Bedrock knowledge base retrieve requests. |
| tests/test_litellm/proxy/pass_through_endpoints/test_pass_through_endpoints.py | Added two mock-only tests: one verifying custom_body is forwarded and takes precedence over the parsed request body, and one verifying fallback to parsed body when custom_body is not provided. |
…ctions (#21192) * Access groups UI * new badge changes * adding tests * fix: add custom_body parameter to endpoint_func in create_pass_through_route (#20849) * fix: add custom_body parameter to endpoint_func in create_pass_through_route The bedrock_proxy_route calls `endpoint_func(custom_body=data)` to pass a pre-parsed, SigV4-signed request body. However, the `endpoint_func` closure created by `create_pass_through_route` does not accept a `custom_body` keyword argument, causing: TypeError: endpoint_func() got an unexpected keyword argument 'custom_body' Add `custom_body: Optional[dict] = None` to both `endpoint_func` definitions (adapter-based and URL-based). In the URL-based path, when `custom_body` is provided by the caller, use it instead of re-parsing the body from the raw request. Fixes #16999 * Add tests for custom_body handling in create_pass_through_route Address reviewer feedback on PR #20849: - Document why the adapter-based endpoint_func accepts custom_body for signature compatibility but does not forward it (the underlying chat_completion_pass_through_endpoint does not support it). - Add test_create_pass_through_route_custom_body_url_target: verifies that when a caller (e.g. bedrock_proxy_route) supplies custom_body, it takes precedence over the body parsed from the raw request. - Add test_create_pass_through_route_no_custom_body_falls_back: verifies that the default path (no custom_body) correctly uses the request-parsed body, preserving existing behavior. Both tests are fully mocked following the project's CONTRIBUTING.md guidelines and the patterns established in the existing test file. Co-authored-by: Cursor <cursoragent@cursor.com> --------- Co-authored-by: themavik <themavik@users.noreply.github.com> Co-authored-by: Cursor <cursoragent@cursor.com> * change to model name for backwards compat * addressing comments * allow editing of access group names * fix: populate identity fields in proxy admin JWT early-return path (#21169) * fix: populate identity fields in proxy admin JWT early-return path When is_proxy_admin is True, the UserAPIKeyAuth early-return now includes user_id, team_id, team_alias, team_metadata, org_id, and end_user_id resolved from the JWT. Previously only user_role and parent_otel_span were set, causing blank Team Name and Internal User in Request Logs UI. * test: add unit tests for proxy admin JWT identity fields * bump: version 0.4.36 → 0.4.37 * migration + build files * Add pyroscope for observability (#21167) * Pyroscope: require PYROSCOPE_APP_NAME and PYROSCOPE_SERVER_ADDRESS, add UTF-8 locale hint - No defaults for PYROSCOPE_APP_NAME or PYROSCOPE_SERVER_ADDRESS; fail at startup if unset when Pyroscope is enabled - Set LANG/LC_ALL to C.UTF-8 when unset to reduce malformed_profile (invalid UTF-8) rejections - Startup message suggests PYTHONUTF8=1 if server rejects profiles - Simplify LITELLM_ENABLE_PYROSCOPE in config_settings; document Pyroscope env vars as required with no default - Add pyroscope_profiling to sidebar (Alerting & Monitoring) - pyproject.toml: pyroscope-io as required dep on non-Windows (marker), in proxy extra * proxy: add PYROSCOPE_SAMPLE_RATE env, use verbose logging, fix int type - Add optional PYROSCOPE_SAMPLE_RATE env (integer, no default) - Pass sample_rate to pyroscope.configure() as int for pyroscope-io - Replace print with verbose_proxy_logger (info/warning) - Document PYROSCOPE_SAMPLE_RATE in config_settings.md * Address Greptile PR feedback: Pyroscope optional, docs, tests, docstring - pyproject.toml: mark pyroscope-io as optional=true (proxy extra only) - Add docs/my-website/docs/proxy/pyroscope_profiling.md (fix broken sidebar link) - Add tests/test_litellm/proxy/test_pyroscope.py for _init_pyroscope() - proxy_server: fix _init_pyroscope docstring (required server/app name, sample rate as int) * Update litellm/proxy/proxy_server.py Co-authored-by: greptile-apps[bot] <165735046+greptile-apps[bot]@users.noreply.github.com> --------- Co-authored-by: greptile-apps[bot] <165735046+greptile-apps[bot]@users.noreply.github.com> * fix(model_info): Add missing tpm/rpm for Gemini models (#21175) Several Gemini models (TTS, native-audio, robotics, gemma) were missing tpm/rpm values, causing test_get_model_info_gemini to fail. Added conservative default values (tpm=250000, rpm=10) for preview models. gemini-2.5-flash-preview-tts gets tpm=4000000, rpm=10. Co-authored-by: OpenClaw <openclaw@users.noreply.github.com> * fix(ci): Fix ruff lint error - unused import in vertex_ai_ingestion (#21178) Co-authored-by: shin-bot-litellm <shin-bot-litellm@users.noreply.github.com> * fix(ci): Fix mypy type errors across 6 files (#21179) - vertex_ai/gemini: fix TypedDict assignment via explicit dict cast - mcp_server: convert MutableMapping scope to dict for type safety - pass_through_endpoints: simplify custom_body logic to fix type narrowing - vector_store_endpoints: add Any annotation for dynamic hook return - responses transformation: use dict() for Reasoning and setattr for dynamic field - zscaler_ai_guard: add assert for api_base None check Co-authored-by: shin-bot-litellm <shin-bot-litellm@users.noreply.github.com> * fix(ci): Fix E2E login button selector - use exact match (#21176) * fix(ci): Fix ruff lint error - unused import Remove unused 'cast' import in vertex_ai_ingestion.py (ruff F401) * fix(ci): Fix E2E login button selector - use exact match Login button selector now matches both 'Login' and 'Login with SSO', causing strict mode violation. Use { exact: true } to match only 'Login'. --------- Co-authored-by: OpenClaw <openclaw@users.noreply.github.com> * fix(mypy): Fix type errors across multiple files (#21180) - vertex_ai/gemini/transformation.py: Fix TypedDict assignment via dict alias - mcp_server/server.py: Convert ASGI scope to dict for type compatibility - pass_through_endpoints.py: Add explicit Optional[dict] type annotation - vector_store_endpoints/endpoints.py: Add Any type for dynamic proxy hook - responses transformation.py: Use dict(Reasoning()) and setattr for compatibility - zscaler_ai_guard.py: Add assert for api_base nullability Co-authored-by: OpenClaw <openclaw@users.noreply.github.com> * [Guardrails] Add guardrail pipeline support for conditional sequential execution (#21177) * Add pipeline type definitions for guardrail pipelines PipelineStep, GuardrailPipeline, PipelineStepResult, PipelineExecutionResult with validation for actions (allow/block/next/modify_response) and modes. * Export pipeline types from policy_engine types package * Add optional pipeline field to Policy model * Add pipeline executor for sequential guardrail execution * Parse pipeline config in policy registry * Add pipeline validation in policy validator * Add pipeline resolution and managed guardrail tracking * Resolve pipelines and exclude managed guardrails in pre-call * Integrate pipeline execution into proxy pre_call_hook * Add test guardrails for pipeline E2E testing * Add example pipeline config YAML * Add unit tests for pipeline type definitions * Add unit tests for pipeline executor * Update litellm/proxy/policy_engine/pipeline_executor.py Co-authored-by: greptile-apps[bot] <165735046+greptile-apps[bot]@users.noreply.github.com> * Update litellm/proxy/utils.py Co-authored-by: greptile-apps[bot] <165735046+greptile-apps[bot]@users.noreply.github.com> --------- Co-authored-by: greptile-apps[bot] <165735046+greptile-apps[bot]@users.noreply.github.com> * Add pipeline flow builder UI for guardrail policies (#21188) * Add pipeline type definitions for guardrail pipelines PipelineStep, GuardrailPipeline, PipelineStepResult, PipelineExecutionResult with validation for actions (allow/block/next/modify_response) and modes. * Export pipeline types from policy_engine types package * Add optional pipeline field to Policy model * Add pipeline executor for sequential guardrail execution * Parse pipeline config in policy registry * Add pipeline validation in policy validator * Add pipeline resolution and managed guardrail tracking * Resolve pipelines and exclude managed guardrails in pre-call * Integrate pipeline execution into proxy pre_call_hook * Add test guardrails for pipeline E2E testing * Add example pipeline config YAML * Add unit tests for pipeline type definitions * Add unit tests for pipeline executor * Add pipeline column to LiteLLM_PolicyTable schema * Add pipeline field to policy CRUD request/response types * Add pipeline support to policy DB CRUD operations * Add PipelineStep and GuardrailPipeline TypeScript types * Add Zapier-style pipeline flow builder UI component * Integrate pipeline flow builder with mode toggle in policy form * Add pipeline display section to policy info view * Add unit tests for pipeline in policy CRUD types * Refactor policy form to show mode picker first with icon cards * Add full-screen FlowBuilderPage component for pipeline editing * Wire up full-screen flow builder in PoliciesPanel with edit routing * Restyle flow builder to match dev-tool UI aesthetic * Restyle flow builder cards to match reference design * Update step card to expanded layout with stacked ON PASS / ON FAIL sections * Add end card to flow builder showing return to normal control flow * Add PipelineTestRequest type for test-pipeline endpoint * Export PipelineTestRequest from policy_engine types * Add POST /policies/test-pipeline endpoint * Add testPipelineCall networking function * Add PipelineStepResult and PipelineTestResult types * Add test pipeline panel to flow builder with run button and results display * Fix pipeline executor: inject guardrail name into metadata so should_run_guardrail allows execution * Update litellm/proxy/policy_engine/pipeline_executor.py Co-authored-by: greptile-apps[bot] <165735046+greptile-apps[bot]@users.noreply.github.com> * Update litellm/proxy/utils.py Co-authored-by: greptile-apps[bot] <165735046+greptile-apps[bot]@users.noreply.github.com> * Update litellm/proxy/policy_engine/policy_endpoints.py Co-authored-by: greptile-apps[bot] <165735046+greptile-apps[bot]@users.noreply.github.com> * Update litellm/proxy/policy_engine/pipeline_executor.py Co-authored-by: greptile-apps[bot] <165735046+greptile-apps[bot]@users.noreply.github.com> --------- Co-authored-by: greptile-apps[bot] <165735046+greptile-apps[bot]@users.noreply.github.com> * fix(responses-bridge): extract list-format system content into instructions When system message content is a list of content blocks (e.g. [{"type": "text", "text": "..."}]) instead of a plain string, the responses API bridge was passing it through as a role: system message in the input items. APIs like ChatGPT Codex reject this with "System messages are not allowed". This happens when requests come through the Anthropic /v1/messages adapter, which converts system prompts into list-format content blocks in the OpenAI chat completions format. Fix: extract text from list content blocks and concatenate into the instructions parameter, matching the existing behavior for string system content. * test: add tests for system message extraction in responses bridge Add three tests for convert_chat_completion_messages_to_responses_api: - String system content → instructions - List-format content blocks → instructions (the bug this PR fixes) - Multiple system messages (mixed string and list) concatenated * fix: add warning log for unexpected system content types Address review feedback: add an else clause that logs a warning for any system content that is neither str nor list, rather than silently dropping it. --------- Co-authored-by: yuneng-jiang <yuneng.jiang@gmail.com> Co-authored-by: The Mavik <179817126+themavik@users.noreply.github.com> Co-authored-by: themavik <themavik@users.noreply.github.com> Co-authored-by: Cursor <cursoragent@cursor.com> Co-authored-by: Ishaan Jaff <ishaanjaffer0324@gmail.com> Co-authored-by: Alexsander Hamir <alexsanderhamirgomesbaptista@gmail.com> Co-authored-by: greptile-apps[bot] <165735046+greptile-apps[bot]@users.noreply.github.com> Co-authored-by: shin-bot-litellm <shin-bot-litellm@berri.ai> Co-authored-by: OpenClaw <openclaw@users.noreply.github.com> Co-authored-by: shin-bot-litellm <shin-bot-litellm@users.noreply.github.com>
…h_route (BerriAI#20849) * fix: add custom_body parameter to endpoint_func in create_pass_through_route The bedrock_proxy_route calls `endpoint_func(custom_body=data)` to pass a pre-parsed, SigV4-signed request body. However, the `endpoint_func` closure created by `create_pass_through_route` does not accept a `custom_body` keyword argument, causing: TypeError: endpoint_func() got an unexpected keyword argument 'custom_body' Add `custom_body: Optional[dict] = None` to both `endpoint_func` definitions (adapter-based and URL-based). In the URL-based path, when `custom_body` is provided by the caller, use it instead of re-parsing the body from the raw request. Fixes BerriAI#16999 * Add tests for custom_body handling in create_pass_through_route Address reviewer feedback on PR BerriAI#20849: - Document why the adapter-based endpoint_func accepts custom_body for signature compatibility but does not forward it (the underlying chat_completion_pass_through_endpoint does not support it). - Add test_create_pass_through_route_custom_body_url_target: verifies that when a caller (e.g. bedrock_proxy_route) supplies custom_body, it takes precedence over the body parsed from the raw request. - Add test_create_pass_through_route_no_custom_body_falls_back: verifies that the default path (no custom_body) correctly uses the request-parsed body, preserving existing behavior. Both tests are fully mocked following the project's CONTRIBUTING.md guidelines and the patterns established in the existing test file. Co-authored-by: Cursor <cursoragent@cursor.com> --------- Co-authored-by: themavik <themavik@users.noreply.github.com> Co-authored-by: Cursor <cursoragent@cursor.com>
Summary
Fixes #16999
The
bedrock_proxy_routecallsendpoint_func(custom_body=data)to pass a pre-parsed, SigV4-signed request body to the pass-through handler. However, theendpoint_funcclosure created bycreate_pass_through_routedoes not accept acustom_bodykeyword argument, causing:This crashes all Bedrock knowledge base retrieve requests going through the v1 passthrough.
Root Cause
create_pass_through_routedefines two variants ofendpoint_func:CustomLogger)Neither accepts
custom_bodyas a parameter. The bedrock proxy route at line 1050 passescustom_body=dataas a keyword argument, which fails.Fix
custom_body: Optional[dict] = Noneto bothendpoint_funcsignatures.custom_bodyis provided by the caller, use it directly instead of re-parsing the body from the raw request. This is the correct behavior since the bedrock proxy has already parsed and signed the body.Changes
litellm/proxy/pass_through_endpoints/pass_through_endpoints.pyTest Plan
TypeError.custom_body.custom_bodyis not provided.