Code
Hub
Workspaces
Following
Trending
Connect
MCP
copy
Create free account
hub
/
github.com/crewAIInc/crewAI
/ functions
Functions
19,663 in github.com/crewAIInc/crewAI
⨍
Functions
19,663
◇
Types & classes
3,456
↳
Endpoints
994
↓ 2 callers
Function
_origin
(url: str | None)
lib/cli/src/crewai_cli/experimental/eval_crew.py:729
↓ 2 callers
Function
_parse_decision
Return the validated decision, or ``None`` and the failure reason.
scripts/jev_evaluation_runner.py:473
↓ 2 callers
Method
_parse_frontmatter
Extract YAML frontmatter fields from a SKILL.md string. Reuses crewai.skills.parser when available, with a minimal fallback for envir
lib/cli/src/crewai_cli/skills/main.py:497
↓ 2 callers
Function
_parse_iso
Parse an ISO 8601 timestamp to an epoch float; ``0.0`` on failure.
lib/cli/src/crewai_cli/model_catalog.py:564
↓ 2 callers
Method
_parse_planning_response
Parses the planning response to extract the plan and readiness. Args: response: The LLM response. Returns: T
lib/crewai/src/crewai/utilities/reasoning_handler.py:589
↓ 2 callers
Function
_parse_template_segments
(value: str)
lib/crewai/src/crewai/flow/expressions.py:119
↓ 2 callers
Function
_parse_xml
(xml: bytes, on_start: Callable[[str, dict[str, str]], None])
lib/crewai-files/src/crewai_files/spreadsheets/reading.py:460
↓ 2 callers
Function
_patch_conversational_flow_project
A flow project whose kickoff script yields a conversational flow. Returns the list that records whether the chat TUI or the kickoff subproces
lib/cli/tests/test_run_crew.py:1618
↓ 2 callers
Function
_patch_conversational_kickoff
(monkeypatch: pytest.MonkeyPatch)
lib/cli/tests/test_kickoff_flow.py:92
↓ 2 callers
Method
_persist_backend_for
( self, persist_definition: FlowPersistenceDefinition )
lib/crewai/src/crewai/flow/runtime/__init__.py:3158
↓ 2 callers
Function
_pipe_to_stdin
Pipe ``text`` to the real ``click.prompt``; a hidden prompt fails the test, because outside pytest it reads the terminal and never sees the pipe.
lib/cli/tests/test_create_crew.py:1639
↓ 2 callers
Function
_pop_outcome
Raise a scripted error or return a scripted successful result.
lib/crewai/tests/llms/test_retry.py:153
↓ 2 callers
Method
_post_agent_execution
(self, agent: BaseAgent)
lib/crewai/src/crewai/task.py:1027
↓ 2 callers
Method
_post_kickoff
(self, result: CrewOutput)
lib/crewai/src/crewai/crew.py:1111
↓ 2 callers
Function
_prepare_agent_cards_dict
Prepare agent cards dictionary from result and existing cards. Shared logic for both sync and async response handlers.
lib/crewai/src/crewai/a2a/wrapper.py:981
↓ 2 callers
Function
_prepare_auth_headers
Prepare authentication headers and TLS verification settings. Args: auth: Optional authentication scheme. timeout: Request timeou
lib/crewai/src/crewai/a2a/utils/agent_card.py:58
↓ 2 callers
Method
_prepare_completion_params
Prepare parameters for Azure AI Inference chat completion. Args: messages: Formatted messages for Azure tools: Tool d
lib/crewai/src/crewai/llms/providers/azure/completion.py:671
↓ 2 callers
Method
_prepare_completion_params
Prepare parameters for OpenAI chat completion.
lib/crewai/src/crewai/llms/providers/openai/completion.py:2257
↓ 2 callers
Function
_prepare_delegation_context
Prepare delegation context from agent response and task. Shared logic for both sync and async delegation. Returns: DelegationContext
lib/crewai/src/crewai/a2a/wrapper.py:1046
↓ 2 callers
Method
_prepare_event
Register source, set scope/sequence metadata, and record the event. This method mutates ContextVar state (scope stack, last_event_id)
lib/crewai/src/crewai/events/event_bus.py:537
↓ 2 callers
Method
_prepare_event_data
( self, tool: Any, tool_calling: ToolCalling | InstructorToolCalling )
lib/crewai/src/crewai/tools/tool_usage.py:1060
↓ 2 callers
Method
_prepare_feedback_iteration
Reset flow completion state before rerunning with feedback.
lib/crewai/src/crewai/agents/agent_executor.py:3596
↓ 2 callers
Function
_prepare_llm_call
Shared pre-call logic: run before hooks and resolve messages. Yields for the duration of the LLM call so the LLM layer knows the hooks alread
lib/crewai/src/crewai/utilities/agent_utils.py:476
↓ 2 callers
Method
_prepare_output
Ensures output path is ready for writing.
lib/crewai-tools/src/crewai_tools/tools/files_compressor_tool/files_compressor_tool.py:102
↓ 2 callers
Function
_prepare_project_for_deploy
Validate deploy inputs before creating a missing lockfile.
lib/cli/src/crewai_cli/deploy/main.py:275
↓ 2 callers
Function
_prepare_project_import_path
()
lib/cli/src/crewai_cli/kickoff_flow.py:29
↓ 2 callers
Function
_prepare_search_params
Prepare search parameters for Qdrant query_points. Args: collection_name: Name of the collection to search. query_embedding: Embe
lib/crewai/src/crewai/rag/qdrant/utils.py:155
↓ 2 callers
Method
_prepare_task_execution
Prepare common setup for task execution shared by sync and async paths. Handles reasoning, prompt building, and memory retrieval. Ar
lib/crewai/src/crewai/agent/core.py:714
↓ 2 callers
Function
_print_brief
The markdown brief AMP wrote for a coding agent, when nobody is watching. It is what the report page's "Copy for your coding agent" button copies
lib/cli/src/crewai_cli/experimental/eval_crew.py:2326
↓ 2 callers
Function
_print_evaluation_line
The evaluation's link, once the app that showed it has gone. An evaluation started inside the app is read there; the terminal is what is left
lib/cli/src/crewai_cli/run_crew.py:1097
↓ 2 callers
Function
_print_info
(meta: dict[str, Any])
lib/cli/src/crewai_cli/checkpoint_cli.py:413
↓ 2 callers
Function
_proactively_summarize_context
Try to summarize oversized context without blocking a valid LLM call.
lib/crewai/src/crewai/utilities/agent_utils.py:678
↓ 2 callers
Method
_process_async_tasks
( self, futures: list[tuple[Task, Future[TaskOutput], int]], was_replayed: bool = Fals
lib/crewai/src/crewai/crew.py:2044
↓ 2 callers
Method
_process_completion_response
Process completion response with usage tracking, tool execution, and events. Args: response: Chat completion response from Azure
lib/crewai/src/crewai/llms/providers/azure/completion.py:874
↓ 2 callers
Function
_process_query_results
Process ChromaDB query results and convert to SearchResult format. Args: collection: The ChromaDB collection (sync or async) that was que
lib/crewai/src/crewai/rag/chromadb/utils.py:268
↓ 2 callers
Function
_process_response_result
Process LLM response and determine next action. Shared logic for both sync and async handlers. Returns: Tuple of (final_result, next
lib/crewai/src/crewai/a2a/wrapper.py:881
↓ 2 callers
Method
_process_response_with_tools
Process response, execute function calls, and finalize completion. Args: response: The completion response contents:
lib/crewai/src/crewai/llms/providers/gemini/completion.py:946
↓ 2 callers
Method
_process_schema_type
Process a JSON schema and return appropriate Python type.
lib/crewai-tools/src/crewai_tools/adapters/enterprise_adapter.py:110
↓ 2 callers
Function
_process_search_results
Process Qdrant search response into SearchResult format. The threshold is applied to the normalized score, the same 0-1 scale the results rep
lib/crewai/src/crewai/rag/qdrant/utils.py:209
↓ 2 callers
Method
_process_stream_chunk
Process a single streaming chunk. Args: chunk: The streaming chunk response full_response: Accumulated response text
lib/crewai/src/crewai/llms/providers/gemini/completion.py:1066
↓ 2 callers
Method
_process_streaming_update
Process a single streaming update chunk. Args: update: Streaming update from Azure API full_response: Accumulated res
lib/crewai/src/crewai/llms/providers/azure/completion.py:996
↓ 2 callers
Function
_project_endpoints
Where the project's `.env`, JSON crew or agents.yaml send model requests that the machine's own environment does not, each as `host (setting)`.
lib/cli/src/crewai_cli/api_keys.py:322
↓ 2 callers
Function
_project_files
The project's own files, by path relative to it: hidden folders, caches, virtual environments, `crewai eval export` folders and `eval.jsonc` (the
lib/cli/src/crewai_cli/experimental/eval_crew.py:1030
↓ 2 callers
Function
_project_root
(project_root: Path | None)
lib/crewai/src/crewai/project/json_loader.py:805
↓ 2 callers
Function
_project_run_command
Return a runner command using the project environment.
lib/cli/src/crewai_cli/run_crew.py:596
↓ 2 callers
Function
_project_script_target
(script_name: str)
lib/cli/src/crewai_cli/kickoff_flow.py:17
↓ 2 callers
Function
_project_state
Each project file with its modification time and size.
lib/cli/src/crewai_cli/experimental/eval_crew.py:1072
↓ 2 callers
Function
_project_tracing_value
`CREWAI_TRACING_ENABLED` as the run will see it: the project's `.env` (which wins, as the run loads it) or else the environment, lowercased.
lib/cli/src/crewai_cli/experimental/eval_crew.py:288
↓ 2 callers
Function
_prompt_can_be_shown
Could a question about traces be put to somebody at all? Three cases where it cannot: a suite under test, a host that asked for tracing messa
lib/crewai/src/crewai/events/listeners/tracing/utils.py:531
↓ 2 callers
Method
_prompt_input_async
Show rich panel and prompt for input without blocking the event loop. Args: crew: The crew instance for context. Returns
lib/crewai/src/crewai/core/providers/human_input.py:372
↓ 2 callers
Function
_prompt_text
( label: str, default: str = "", *, spacing_before: bool = True, )
lib/cli/src/crewai_cli/create_json_crew.py:312
↓ 2 callers
Function
_provider_api_key
First non-empty API key found among the provider's env vars.
lib/cli/src/crewai_cli/model_catalog.py:76
↓ 2 callers
Function
_pydantic_valid_event
Return True if *data* validates as an A2UIEvent.
lib/crewai/tests/a2a/extensions/test_a2ui_schema_conformance.py:43
↓ 2 callers
Function
_pydantic_valid_message
Return True if *data* validates as an A2UIMessage.
lib/crewai/tests/a2a/extensions/test_a2ui_schema_conformance.py:34
↓ 2 callers
Function
_python_ref_path
(value: Any, source: str | Path)
lib/crewai/src/crewai/project/json_loader.py:694
↓ 2 callers
Function
_query_value
(url: str, params: tuple[str, ...])
lib/crewai-files/src/crewai_files/spreadsheets/__init__.py:171
↓ 2 callers
Function
_raise_for_upstream_error
Raise when a 200 response carries an upstream error instead of choices. Gateways commit `200 OK` as soon as a provider accepts the request, so a
lib/crewai/src/crewai/llms/providers/openai/completion.py:131
↓ 2 callers
Function
_raw_create_double
Double for `chat.completions.with_raw_response.create`. `text` has to be real JSON carrying `choices` so the gateway error-envelope guard see
lib/crewai/tests/llms/openai/test_openai.py:503
↓ 2 callers
Function
_react_text_executor
Wire an executor for the ReAct text tool route (execute_tool_action).
lib/crewai/tests/agents/test_react_tool_hook_contract.py:35
↓ 2 callers
Function
_read_agent_definition
(path: Path)
lib/crewai/src/crewai/project/json_loader.py:241
↓ 2 callers
Function
_read_definition
(path: Path)
lib/cli/src/crewai_cli/agent_tools.py:157
↓ 2 callers
Function
_read_key
()
lib/cli/src/crewai_cli/tui_picker.py:31
↓ 2 callers
Function
_record
(record_id: str, scope: str)
lib/crewai/tests/memory/test_lancedb_scoped_count.py:23
↓ 2 callers
Method
_record_to_point
Convert a MemoryRecord to a Qdrant Point.
lib/crewai/src/crewai/memory/storage/qdrant_edge_storage.py:192
↓ 2 callers
Method
_record_to_row
(self, record: MemoryRecord)
lib/crewai/src/crewai/memory/storage/lancedb_storage.py:272
↓ 2 callers
Function
_record_usage
Count an evaluation that is actually starting, whether the caller was logged in, and which run it grades. TARGET is `last_run` when `crewai e
lib/cli/src/crewai_cli/experimental/eval_crew.py:135
↓ 2 callers
Function
_recorded_run
This execution's own record, if its trace reached AMP. crewAI records a run when its spans are exported, so the record answers the one questi
lib/cli/src/crewai_cli/crew_run_tui.py:509
↓ 2 callers
Method
_redeploy
( self, endpoint: str, *, project_id: str | None, force: bool )
lib/crewai-core/src/crewai_core/plus_api.py:482
↓ 2 callers
Function
_referenced_input_names
All ``{placeholder}`` names referenced by a crew's agents and tasks.
lib/cli/src/crewai_cli/run_crew.py:196
↓ 2 callers
Method
_refresh_tree
(self)
lib/cli/src/crewai_cli/checkpoint_tui.py:335
↓ 2 callers
Function
_register_custom_provider
A custom vector store: its own config class plus a registered factory.
lib/crewai/tests/knowledge/test_knowledge_storage_providers.py:219
↓ 2 callers
Method
_remember_no_reasoning_effort_model
Record that this model rejected `reasoning_effort`. Called once the retry without the parameter has succeeded, so a retry that fails
lib/crewai/src/crewai/llms/providers/openai/completion.py:2161
↓ 2 callers
Method
_remember_tools_need_responses
Send this model's later calls on this endpoint to the Responses API. Not with ``custom_openai=True``: `_effective_api` never upgrades those,
lib/crewai/src/crewai/llms/providers/openai/completion.py:2098
↓ 2 callers
Method
_render
Render the tool name and description in plain text.
lib/crewai/src/crewai/tools/tool_usage.py:855
↓ 2 callers
Method
_render_log_panel
(self)
lib/cli/src/crewai_cli/crew_run_tui.py:2256
↓ 2 callers
Method
_replace
( self, sandbox: Any, paths: list[str], pattern: str, replacement: str
lib/crewai-tools/src/crewai_tools/tools/daytona_sandbox_tool/daytona_file_tool.py:472
↓ 2 callers
Method
_replace_step
Replace current step in-place (no archive). Used for tool results.
lib/cli/src/crewai_cli/crew_run_tui.py:2454
↓ 2 callers
Function
_report_empty_llm_answer
( printer: Printer, verbose: bool, *, retrying: bool )
lib/crewai/src/crewai/utilities/agent_utils.py:512
↓ 2 callers
Function
_report_url
The report link, if it is one a browser may be sent to. It is printed as a link and opened without a click, and it comes from whichever AMP a
lib/cli/src/crewai_cli/experimental/eval_crew.py:1507
↓ 2 callers
Method
_request_kwargs
Validate the LLM, dispatch before-call hooks and build the request.
lib/crewai/src/crewai/utilities/internal_instructor.py:172
↓ 2 callers
Method
_require_cel_source
(value: str, *, source: str | None = None)
lib/crewai/src/crewai/flow/expressions.py:455
↓ 2 callers
Method
_reset_current
Reset the current agent and task tracking state.
lib/crewai/src/crewai/experimental/evaluation/evaluation_listener.py:212
↓ 2 callers
Method
_reset_memory_system
Reset a single memory system. Args: system: The memory system instance to reset. name: Display name of the memory sys
lib/crewai/src/crewai/crew.py:2462
↓ 2 callers
Method
_reset_turn_execution_state
Clear per-execution tracking so the next turn re-runs the graph.
lib/crewai/src/crewai/flow/conversational_mixin.py:1245
↓ 2 callers
Method
_reset_user_settings
Reset all user settings to default values.
lib/crewai-core/src/crewai_core/settings.py:264
↓ 2 callers
Function
_resolve_against_base
Resolve *path* the way the sandbox does, anchoring relatives to *base_dir*. ``validate_file_path`` and ``format_path_for_display`` both join a re
lib/crewai-tools/src/crewai_tools/tools/file_read_tool/file_read_tool.py:16
↓ 2 callers
Method
_resolve_answer_from_history_prompt
Return the deprecated history-only route's effective prompt.
lib/crewai/src/crewai/flow/conversational_mixin.py:1293
↓ 2 callers
Function
_resolve_api_key
Pick ``api_key``, then ``GOOGLE_API_KEY``, then ``GEMINI_API_KEY``. The placeholder is skipped so a leftover one cannot hide a real key further
lib/crewai/src/crewai/llms/providers/gemini/completion.py:44
↓ 2 callers
Method
_resolve_as_url
Resolve a URL source as UrlReference. Args: file: The file with URL source. Returns: UrlReference with the U
lib/crewai-files/src/crewai_files/resolution/resolver.py:147
↓ 2 callers
Method
_resolve_duration
Validate and clamp the requested duration to ``max_seconds``. ``BaseTool.run`` skips ``args_schema`` validation when called with posi
lib/crewai-tools/src/crewai_tools/tools/wait_tool/wait_tool.py:157
↓ 2 callers
Function
_resolve_embedder_python_refs
( value: Any, source: str | Path, project_root: Path | None, )
lib/crewai/src/crewai/project/json_loader.py:1609
↓ 2 callers
Function
_resolve_explain_command
Resolve the underlying command from an EXPLAIN [ANALYZE] [VERBOSE] statement. Returns the real command (e.g., 'DELETE') if ANALYZE is present, el
lib/crewai-tools/src/crewai_tools/tools/nl2sql/nl2sql_tool.py:185
↓ 2 callers
Function
_resolve_hooks
Resolve the ordered hooks for a point: global first, then scoped.
lib/crewai/src/crewai/hooks/dispatch.py:201
↓ 2 callers
Method
_resolve_index_name
(self, index_name: str | None, idx_type: str)
lib/crewai-tools/src/crewai_tools/tools/oracle_vector_search_tool/vector_search.py:659
↓ 2 callers
Method
_resolve_inline
Resolve a file as inline content. Args: file: The file to resolve (used for logging). provider: Provider name.
lib/crewai-files/src/crewai_files/resolution/resolver.py:396
↓ 2 callers
Method
_resolve_input_provider
Resolve the input provider using the priority chain. Resolution order: 1. ``self.input_provider`` (per-flow override) 2. ``fl
lib/crewai/src/crewai/flow/runtime/__init__.py:3479
↓ 2 callers
Function
_resolve_instance_ref
(ref: str, *, field: str)
lib/crewai/src/crewai/flow/runtime/__init__.py:302
↓ 2 callers
Function
_resolve_persistence
(value: Any)
lib/crewai/src/crewai/flow/runtime/__init__.py:290
↓ 2 callers
Function
_resolve_project_path
( value: str, project_root: Path | None, source: str | Path, )
lib/crewai/src/crewai/project/json_loader.py:1805
↓ 2 callers
Function
_resolve_ref
Resolve a $ref to its actual schema. Args: ref: The $ref string (e.g., "#/$defs/MyType"). root_schema: The root schema containing
lib/crewai/src/crewai/utilities/pydantic_schema_utils.py:1079
← previous
next →
2,501–2,600 of 19,663, ranked by callers