mirror of
https://github.com/crewAIInc/crewAI.git
synced 2026-09-24 20:00:30 +00:00
dcb02da8e4ec032dec032c82f78e232cb0a9c0ae
6 Commits
| Author | SHA1 | Message | Date | |
|---|---|---|---|---|
Lucas Gomide
|
e9d4c57b2a |
fix: run model call hooks on every path and propagate a deny (#7111)
* fix: let a hook deny reach the caller as a deny
A hook that raised `HookAborted` on `pre_model_call` never reached the code
making the call: the LLM layer caught it and returned `False`, which providers
translated into `ValueError("LLM call blocked by before_llm_call hook")`,
dropping the reason and the source and making a policy decision
indistinguishable from a provider outage. Every internal model call then
absorbed that error through the `except Exception` that keeps a provider hiccup
from failing a run, so memory analysis fell back to defaults and the converter
and reasoning handler retried the call that was just denied. The abort now
propagates out of the LLM layer while the boolean convention keeps its
documented `ValueError` via `LegacyHookBlocked`, and the fail-open handlers
around internal model calls re-raise it instead of degrading.
* fix: dispatch model call hooks on the paths that skipped them
A model call was only checked when the executor loop drove it: the
`from_agent is not None` short-circuit in `base_llm` silenced the hooks
for agent planning and step observation, no provider `acall` dispatched
them at all, and `InternalInstructor` bypassed `llm.call` entirely. This
replaces that short-circuit with an explicit
`model_call_hooks_already_dispatched` window so the enclosing caller
claims the dispatch, adds the pre-call dispatch to every provider's
`acall`, and runs the hooks around the Instructor client call. A denial
now emits a denied event instead of being logged and reported as a
provider failure.
* fix: report a boolean-convention deny as a deny, not an outage
A `before_llm_call` hook that blocks by returning `False` reached the five
native providers as a plain `ValueError`, which fell through to their generic
`except Exception` and was logged and emitted as `OpenAI API call failed: ...`
— the same deny raised as `HookAborted` was already labelled correctly, so the
two dialects disagreed on whether a policy decision was a provider outage. The
LLM layer now converts it into `LLMCallBlockedError`, still a `ValueError` so
the fail-open handlers around internal model calls keep absorbing it, but its
own type so a provider can report the decision it is. Since a block is raised
rather than returned, the thirteen callers that turned the return flag into a
raise by hand drop that line, and `_prepare_llm_call` raises the same type.
* fix: keep a denied plan from letting the agent run unplanned
`AgentExecutor.generate_plan` wraps `handle_agent_reasoning()` in a bare
`except Exception`, so guarding the reasoning handler alone still left the
deny absorbed one frame up: the executor logged "Error during planning" and
the agent proceeded with no plan. It now re-raises `HookAborted` like the
other planning boundaries, and the accompanying test also covers the
boolean convention still degrading at a fail-open site.
* fix: stop a denied knowledge query from running the task without knowledge
`handle_knowledge_retrieval` and its async twin wrap the query rewrite in
their own `except Exception`, so guarding `_get_knowledge_search_query`
alone still let `execute_task` continue on the unaugmented prompt after a
deny. Both now emit the terminal `KnowledgeSearchQueryFailedEvent` and
re-raise `HookAborted`, matching the second-frame guard already added to
`AgentExecutor.generate_plan`. Also documents the abort contract on
`PlannerObserver.observe`.
* fix: stop nine callers from re-swallowing a model call deny
CodeRabbit caught the replan path re-swallowing a deny, so an AST sweep of
every caller of a guarded function found the same defeat in nine places:
classic and replan planning, memory recall and memory save on both `Agent`
and `LiteAgent`, the base executor's save, and `LLMGuardrail.__call__`,
which turned a refused call into validation feedback. Each now re-raises
`HookAborted` after emitting whatever terminal event it owes, while every
other failure keeps degrading as before — the knowledge guards move to that
same idiom instead of duplicating their emit.
* fix: pair a denied guardrail with the event it started
Re-raising from `LLMGuardrail` left `process_guardrail` between its started
and completed events, so a denied validation read as one still in flight
rather than a policy decision. It now emits `LLMGuardrailCompletedEvent`
with the deny reason before the abort leaves, matching what every other
guarded site in this change already does.
* fix: stop retrying a task after a hook denied its model call
`Agent.execute_task` funnels every exception into `_handle_execution_error`,
which re-runs the whole task up to `max_retry_limit` times, so a policy deny
read as a transient blip: a crew whose first model call was denied retried and
returned a normal answer. `HookAborted` now joins `_passthrough_exceptions`,
the tuple already reserved for deliberate stops. The new boundary tests drive
the public entry points instead of the frame that makes the call, and count
model calls so a deny that gets retried fails the assertion — ten of the twelve
fail against `main`.
* fix: stop a denied plan step from being reported as a failed step
Making model call hooks reachable on agent-bearing calls put a deny inside
`StepExecutor.execute`, whose broad `except Exception` turned it into
`StepResult(success=False)` and let the plan carry on; `HookAborted` now
joins `ToolExecutionFailedError` in the passthrough handlers there, and
`execute_todos_parallel` re-raises a deny that `return_exceptions=True`
would otherwise record as one failed todo. `_emit_call_denied_event` also
renders the source through the now-public `source_name`, so a hook that
names itself with a callable reads as its name instead of a repr.
---------
Co-authored-by: Vidit Ostwal <110953813+Vidit-Ostwal@users.noreply.github.com>
|
||
|
Lucas Gomide
|
7d21283630 |
feat: add generic interception-hook dispatcher (#6516)
* feat: add generic interception-hook dispatcher Introduces `crewai/hooks/dispatch.py` as a single engine behind every interception point: a hook receives a typed context, may mutate or replace its `payload`, or raise `HookAborted(reason, source)` to stop the operation. The full `InterceptionPoint` catalog is frozen from day zero, with global and contextvar-scoped registries, an `@on` decorator, a no-op fast path, and a `HookDispatchedEvent` for telemetry. The four existing `before/after_llm_call` and `before/after_tool_call` hooks become adapters over the dispatcher, so the legacy dialect and `return False` semantics keep working unchanged while gaining the new contract. * fix: harden interception dispatcher against review findings Corrects several dispatcher edge cases surfaced in review. `_default_reducer` now reports a modification only when a `payload` is actually applied, the `agents=` filter falls back to `agent_role` for contexts without an `agent` object, and `unregister` resolves the filter wrapper stashed by `on` so a filtered hook can be removed. The tool-hook runners honor the executing agent's `verbose` flag instead of silently swallowing hook errors, and the ReAct tool path now runs `POST_TOOL_CALL` on blocked calls to match the native paths. Also adds abort-telemetry coverage and replaces the flaky absolute no-op timing budget with a relative one. * fix: honor scoped hooks on direct llm calls and register @on crew methods Direct agent-less LLM calls short-circuited on the empty global hook list, so hooks registered only for the current `scoped_hooks()` context never ran; the direct-call helpers now defer to `dispatch`, which resolves scoped hooks behind its own no-op fast path. `CrewBase` likewise only scanned the legacy `is_*_hook` markers, so `@on(InterceptionPoint.X)` methods were silently dropped — it now registers them on the dispatcher with filters applied and `self` bound. Also tightens result typing across the tool-call seams so `mypy` stays green. * refactor: scope InterceptionPoint to the points this layer wires The dispatcher only fires the model- and tool-call boundaries, so `InterceptionPoint` now lists just those four rather than the full future catalog. New points are introduced alongside the seams that dispatch them, keeping every layer free of enum members with no live consumer. The dispatcher unit tests that borrowed unused points as generic examples are remapped onto the four kept points. * test: pin per-hook fail-open at the LLM and tool seams The dispatcher swallows a hook's exception per hook rather than around the whole loop, so one buggy hook no longer silently skips every hook registered after it. These seam-level tests pin that behavior through `_setup_before_llm_call_hooks` and `run_before/after_tool_call_hooks`, and confirm an intentional `return False` block still short-circuits later hooks. * fix: run execution-scoped hooks on the agent executor model seams `_setup_before/after_llm_call_hooks` only ran the executor's snapshot hook lists, so hooks registered via `scoped_hooks()` never fired on `PRE/POST_MODEL_CALL` during normal agent execution, while the tool seams (which go through `dispatch`) merged them. The seams now append the current scope's hooks after the snapshot via `get_scoped_hooks`, matching dispatch's global-then-scoped ordering, and a scoped-only registration no longer short-circuits the seam. |
||
|
Lucas Gomide
|
6452608724 |
fix: after_llm_call hooks no longer break native tool execution (#6531)
* fix: don't clobber native tool-call responses in after-LLM hooks Registering any `after_llm_call` hook broke native tool execution: the executor invokes `_setup_after_llm_call_hooks` on the intermediate response that carries the model's tool calls, the non-str payload was stringified for the response-rewrite pass, and the executor then treated that string as a final answer instead of executing the tools. Structured payloads (neither `str` nor `BaseModel`) now pass through untouched, mirroring the isinstance guard `_invoke_after_llm_call_hooks` already applies on the direct-call path; hooks still fire on the follow-up textual response. Fixes #6529 * style: shorten the tool-call guard comment |
||
|
Greyson LaLonde
|
fd10c64148 | chore(crewai): drop self-explanatory comments | ||
|
Lorenze Jay
|
c456e5c5fa |
Lorenze/ensure hooks work with lite agents flows (#3981)
* liteagent support hooks * wip llm.call hooks work - needs tests for this * fix tests * fixed more * more tool hooks test cassettes |
||
|
Lorenze Jay
|
528d812263 |
Lorenze/feat hooks (#3902)
* feat: implement LLM call hooks and enhance agent execution context - Introduced LLM call hooks to allow modification of messages and responses during LLM interactions. - Added support for before and after hooks in the CrewAgentExecutor, enabling dynamic adjustments to the execution flow. - Created LLMCallHookContext for comprehensive access to the executor state, facilitating in-place modifications. - Added validation for hook callables to ensure proper functionality. - Enhanced tests for LLM hooks and tool hooks to verify their behavior and error handling capabilities. - Updated LiteAgent and CrewAgentExecutor to accommodate the new crew context in their execution processes. * feat: implement LLM call hooks and enhance agent execution context - Introduced LLM call hooks to allow modification of messages and responses during LLM interactions. - Added support for before and after hooks in the CrewAgentExecutor, enabling dynamic adjustments to the execution flow. - Created LLMCallHookContext for comprehensive access to the executor state, facilitating in-place modifications. - Added validation for hook callables to ensure proper functionality. - Enhanced tests for LLM hooks and tool hooks to verify their behavior and error handling capabilities. - Updated LiteAgent and CrewAgentExecutor to accommodate the new crew context in their execution processes. * fix verbose * feat: introduce crew-scoped hook decorators and refactor hook registration - Added decorators for before and after LLM and tool calls to enhance flexibility in modifying execution behavior. - Implemented a centralized hook registration mechanism within CrewBase to automatically register crew-scoped hooks. - Removed the obsolete base.py file as its functionality has been integrated into the new decorators and registration system. - Enhanced tests for the new hook decorators to ensure proper registration and execution flow. - Updated existing hook handling to accommodate the new decorator-based approach, improving code organization and maintainability. * feat: enhance hook management with clear and unregister functions - Introduced functions to unregister specific before and after hooks for both LLM and tool calls, improving flexibility in hook management. - Added clear functions to remove all registered hooks of each type, facilitating easier state management and cleanup. - Implemented a convenience function to clear all global hooks in one call, streamlining the process for testing and execution context resets. - Enhanced tests to verify the functionality of unregistering and clearing hooks, ensuring robust behavior in various scenarios. * refactor: enhance hook type management for LLM and tool hooks - Updated hook type definitions to use generic protocols for better type safety and flexibility. - Replaced Callable type annotations with specific BeforeLLMCallHookType and AfterLLMCallHookType for clarity. - Improved the registration and retrieval functions for before and after hooks to align with the new type definitions. - Enhanced the setup functions to handle hook execution results, allowing for blocking of LLM calls based on hook logic. - Updated related tests to ensure proper functionality and type adherence across the hook management system. * feat: add execution and tool hooks documentation - Introduced new documentation for execution hooks, LLM call hooks, and tool call hooks to provide comprehensive guidance on their usage and implementation in CrewAI. - Updated existing documentation to include references to the new hooks, enhancing the learning resources available for users. - Ensured consistency across multiple languages (English, Portuguese, Korean) for the new documentation, improving accessibility for a wider audience. - Added examples and troubleshooting sections to assist users in effectively utilizing hooks for agent operations. --------- Co-authored-by: Greyson LaLonde <greyson.r.lalonde@gmail.com> |