* feat(litellm_content_filter.py): add support for content filtering categories
make it easy for proxy admin to prevent messages about violence, self harm or illegal weapons going through litellm
* feat: initial commit adding bias detection
allows admin to block inappropriate content about sexual orientation, etc.
* refactor: simplify content_filter.py
use a more exhaustive set of keywords, instead of guessing at potential phrases user can use
* feat(content_filter.py): add new denied topics for in-built content filter guardrails
allow user to automatically block content relating to certain categories from being sent to the LLML
* refactor(content-filter): document new params to litellm content filter
* feat(ui/): litellm content filter - select content categories on ui
* docs: update documentation
* docs(litellm_content_filter.md): document new content filters
* feat: initial commit adding support for inappropriate images via litellm content filter
* feat(content_filter.py): support blocking images containing blocked content
prevent images which contain disallowed content from being sent to the llm api
* docs(litellm_content_filter.md): document new image capabilities of litellm_content_filter
* fix: fix expected error code
- Add 'mask' action to SUPPORTED_ON_FLAGGED_ACTIONS
- Automatically sanitizes sensitive content using masked_session_messages
- Allows requests to proceed with masked content instead of blocking
- Add MCP call support
- Add pre_mcp_call and during_mcp_call to supported_event_hooks
- Verify mcp_call is supported in call_type Literal types
- Control exception details based on config
- Conditionally include scanners/evidence in exceptions based on
include_scanners and include_evidence settings
- Reduces payload size when detailed exception info isn't needed
- Add comprehensive test coverage
- Tests for masking functionality
- Tests for conditional exception details
- Tests for MCP call support
- Update documentation
- Add Mask section explaining masking functionality
- Clarify exception details control
All changes maintain backward compatibility.
* prioritized bedrock guardrail blocking by removing early return based on masking flags
When mask_request_content: true or mask_response_content: true, the method immediately returning False.
The Result: Even when Bedrock Guardrails returned action: "BLOCKED" for dangerous content, LiteLLM would not raise an exception and allowing the content through the response.
So removed that early condition which will return true for the blocked actions based on guardrails.
* Added test case for bedrock guardrail block content precedence
* fix(unified_guardrails.py): send all chunks on completion of final stream
* feat(generic_guardrail_api.py): handle tool call response on streaming LLM responses
* fix(anthropic/chat/guardrail_translation): initial commit adding anthropic tool response streaming guardrails
enables guardrail checks on tool response from llm's to work via `/v1/messages`
* feat(anthropic/): working guardrail checks on tool response from LLMs
ensures guardrail checks on anthropic /v1/messages works as expected
* feat(responses/guardrail_translation): support tool call response guardrails on streaming for /v1/responses
ensures complete coverage of tool call responses
* refactor(openai.py): refactor to use consistent pydantic model for responses api tool response on streaming
enables non-openai model tool call response to work correctly with guardrail checks on /v1/responses
* test: update tests
* fix: fix linting error
* fix: fix failing tests
* fix: fix import errors
* fix(openai/chat/guardrail_transformation): fix final chunk returned on streaming
* attempt to implement the passthrough feature
* Formatting and small change
* Fix formatting
* feat: grayswan guardrail overwrite ModelResponse in passthrough mode
* fix missing exception error catching on certain
endpoints
* fix wrong call site
* fix: patch anthropic endpoint internal error on streaming obj
* fix grayswan testcase
* feat: update the violation response to more natural
* Formatting
* move passthrough exception definition to custom_guardrail.
* Enhancement: show whether the blocked at input or output
* update exception name
* fix a typo in testing unit.
---------
Co-authored-by: Xiaohan Fu <xiaohan@grayswan.ai>
- Skip empty/whitespace text before calling Presidio API
- Handle error dict responses gracefully (e.g., {'error': 'No text provided'})
- Add defensive error handling for invalid result items
- Add comprehensive test coverage for empty content scenarios
Fixes crash in tool/function calling where assistant messages have empty content.
* fix(unified_guardrail.py): support during_call event type for unified guardrails
allows guardrails overriding apply_guardrails to work 'during_call'
* feat(generic_guardrail_api.py): support new 'tool_calls' field for generic guardrail api
returns the tool calls emitted by the LLM API to the user
* fix(generic_guardrail_api.py): working anthropic /v1/messages tool call response
send llm tool calls to guardrail api when called via `/v1/messages` API
* fix(responses/): run generic_guardrail_api on responses api tool call responses
* fix: fix tests
* test: fix tests
* fix: fix tests
Fixes#17517
- Fixed bug where only the first matching blocked keyword was masked
- Now iterates through ALL blocked keywords and masks each one
- Added 3 regression tests for multiple keyword masking
* fix(unified_guardrail.py): correctly map a v1/messages call to the anthropic unified guardrail
* fix: add more rigorous call type checks
* fix(anthropic_endpoints/endpoints.py): initialize logging object at the beginning of endpoint
ensures call id + trace id are emitted to guardrail api
* feat(anthropic/chat/guardrail_translation): support streaming guardrails
sample on every 5 chunks
* fix(openai/chat/guardrail_translation): support openai streaming guardrails
* fix: initial commit fixing output guardrails for responses api
* feat(openai/responses/guardrail_translation): handler.py - fix output checks on responses api
* fix(openai/responses/guardrail_translation/handler.py): ensure responses api guardrails work on streaming
* test: update tests
* test: update tests
* fix: support multiple kinds of input to the guardrail api
* feat(guardrail_translation/handler.py): support extracting tool calls from openai chat completions for guardrail api's
* feat(generic_guardrail_api.py): support extracting + returning modified tool calls on generic_guardrails_api
allows guardrail api to analyze tool call being sent to provider - to run any analysis on it
* fix(guardrails.py): support anthropic /v1/messages tool calls
* feat(responses_api/): extract tool calls for guardrail processing
* docs(generic_guardrail_api.md): document tools param support
* docs: generic_guardrail_api.md
improve documentation
* fix(unified_guardrail.py): correctly map a v1/messages call to the anthropic unified guardrail
* fix: add more rigorous call type checks
* fix(anthropic_endpoints/endpoints.py): initialize logging object at the beginning of endpoint
ensures call id + trace id are emitted to guardrail api
* feat(anthropic/chat/guardrail_translation): support streaming guardrails
sample on every 5 chunks
* fix(openai/chat/guardrail_translation): support openai streaming guardrails
* fix: initial commit fixing output guardrails for responses api
* feat(openai/responses/guardrail_translation): handler.py - fix output checks on responses api
* fix(openai/responses/guardrail_translation/handler.py): ensure responses api guardrails work on streaming
* test: update tests
* test: update tests
* test: update tests
* fix(bedrock_guardrails.py): fix post call streaming iterator logic
* fix: fix return
* fix(bedrock_guardrails.py): fix
* fix(initial-commit): adding a way to get the right response type based on the api route
* feat(unified_guardrail.py): support streaming guardrails
* test: update tests
* fix: fix linting errors
* test: update tests
* refactor(generic_guardrail_api.py): refactor to update to new guardrail api logic
* refactor: refactor llm api integrations to support passing in text as a list[str] instead of one at a time
* refactor: fix linting errors
* refactor: pass request type to guardrail api
allows request vs. response processing to occur
* feat: pass user api key dict information to the guardrail api
* fix: pass user api key dict information to the guardrail api
* feat: pass litellm call id + trace id, if present
* docs: update docs
* store and fetch lasso-conversation id from cache
* include gateway/v# in the baseUrl to allow simpler version migrations in the future
* add tests for cached conversation ID
* add prompt security guardrails provider
* cosmetic
* small
* add file sanitization and update context window
* add pdf and OOXML files support
* add system prompt support
* add tests and documentation
* remove print
* fix PLR0915 Too many statements (96 > 50)
* cosmetic
* fix mypy error
* Fix failed tests due to naming conflict of responses directory with same-named pip package
* Fix mypy error: use 'aembedding' instead of 'embeddings' for async embedding call type
* Fix: Install enterprise package into Poetry virtualenv for tests
The GitHub Actions workflow was installing litellm-enterprise to system Python
using 'python -m pip install -e .', but tests run in Poetry's virtualenv using
'poetry run pytest'. This caused ImportError for enterprise package types.
Changed to 'poetry run pip install -e .' so the package is available in the
same virtualenv where pytest executes.
Fixes enterprise test collection errors in GitHub Actions CI.
* Move Prompt Security guardrail tests to tests/test_litellm/
Per reviewer feedback, move test_prompt_security_guardrails.py from
tests/guardrails_tests/ to tests/test_litellm/proxy/guardrails/ so
it will be executed by GitHub Actions workflow test-litellm.yml.
This ensures the Prompt Security integration tests run in CI.
---------
Co-authored-by: Ori Tabac <oritabac@prompt.security>
Co-authored-by: Vitaly Neyman <vitaly@prompt.security>
* attempt to implement the passthrough feature
* Formatting and small change
* Fix formatting
* Format test file
---------
Co-authored-by: Xiaohan Fu <xiaohan@grayswan.ai>
* Use auth key name if there are no app id in in headers or in extra_data
* use key alias instead of key name
* Fix
* last priority key alias
* Fix
* Add tests
* feat(guardrails): Add deduplication and session tracking
- Implement deduplication logic to prevent duplicate scans (via call_id; add _check_and_mark_scanned) caused by LiteLLM callback system
- Add session tracking using litellm_trace_id as AI Session ID for Prisma AIRS SCM logging
- Extract helper methods _extract_prompt_from_request maintainability
- Use httpxSpecialProvider import (LoggingCallback -> GuardrailCallback)
- Add comprehensive tests for deduplication and session tracking (7 new tests)
- Update documentation with multi-turn conversation tracking examples
* docs: update PANW Prisma AIRS multi-turn conversation example to use industry-standard terminology
- Clearer example for conversation tracking
- Updated terminology from 'AI Session ID' to 'Prisma AIRS AI Session ID' for clarity
* fix: remove unused asyncio import
* fix: correct mypy type ignore comment
* noma support v2 api and images with during call
* supporting streams and images with texts
* Supporting text now
* annonymization works
* removing function
* fixing noma.py
* all old tests pass
* adding new tests
* removing changes
* Fixing application id headers
* fix whitespace
* deleting unused imports
* UI Config Guardrails should not be editable and guardrails info definition location fix
* Remove unused import
* Added literals for guardrail definition location
* fix(presidio.py): handle content as a list of texts
covers openai + anthropic messages api
* fix(presidio.py): safe get messages
* test: add unit testing for presidio guardrails
* fix(unified_guardrail.py): initial commit
* fix(enkryptai.py): implement apply_guardrail to enkrypt guardrail
* fix(unified_guardrail.py): support unified guardrail on input
* feat(unified_guardrail.py): add post call success hook implementation
allows us to just have 1 place to handle llm translation to guardrail api spec
* refactor: refactor initial unified guardrail component
* refactor: more refactoring
* feat(responses/): add guardrails to responses api
allows existing guardrails to work for new llm endpoints
* docs(adding_guardrail_support.md): document new guardrail endpoint support
* test: add unit tests
* feat(image_generation/): add guardrail support for image generation endpoint
* feat(openai/text_completion): support guardrails on `/v1/completions` API
* docs: document guardrails support on new endpoints
* docs: clarify when guardrails run
* feat(openai/speech): add guardrail support for input
* docs(rerank/): add guardrail support on input query
* fix: fix ruff check
