MCPcopy Create free account

hub / github.com/crewAIInc/crewAI / functions

Functions19,663 in github.com/crewAIInc/crewAI

Functiontest_a_role_bearing_plain_attribute_is_cut_plainly_never_reshaped
A task may produce a JSON list whose items carry ``role`` (a roster, a chat log). On ``crewai.task.output`` that is data, not a GenAI conversa
lib/crewai/tests/telemetry/test_span_attribute_bound.py:348
Functiontest_a_role_key_is_built_from_the_declaration_a_model_key_mapped
Role wins, with the caller's declared settings: the llm a model key swapped on the way in is not what the role's model is built like.
lib/crewai/tests/test_llm_overlay.py:829
Functiontest_a_role_key_wins_over_model_keys_for_that_agent
()
lib/crewai/tests/test_llm_overlay.py:713
Functiontest_a_role_that_looks_like_a_model_key_is_never_one
()
lib/crewai/tests/test_llm_overlay.py:731
Functiontest_a_role_the_interpolation_leaves_unchanged_is_not_resolved_again
Construction's resolution stands: same model, same llm instance. Without this an agent built OUTSIDE the block would pick the mapped model up
lib/crewai/tests/test_llm_overlay.py:318
Methodtest_a_row_of_numbers_is_never_the_header
(self)
lib/crewai-files/tests/spreadsheets/test_reading.py:584
Methodtest_a_row_whose_only_cell_is_off_type_is_still_a_row
(self)
lib/crewai-files/tests/spreadsheets/test_reading.py:466
Functiontest_a_run_crewai_eval_started_is_traced_with_nobody_at_the_screen
The run `crewai eval` started to grade: running the command was the yes, whatever the terminal, the suite or the project's `.env` say.
lib/cli/tests/test_crew_run_tui.py:478
Functiontest_a_run_crewai_eval_started_is_uploaded_with_nobody_to_ask
`crewai eval` ran the crew to grade it, and the grade is read off the trace: running the command was the yes. A coding agent with no terminal
lib/crewai/tests/telemetry/test_deferred_trace_transitions.py:184
Functiontest_a_run_crewai_makes_on_model_text_reads_no_url
(downloads)
lib/crewai/tests/agents/test_agent_spreadsheets.py:925
Functiontest_a_run_from_hours_ago_is_not_waited_for
Nothing is in flight: AMP does not hold this run, and saying so at once is the useful answer.
lib/cli/tests/experimental/test_eval_crew.py:1481
Functiontest_a_run_inside_the_block_is_awaited_and_nothing_is_left_after
()
lib/crewai-core/tests/test_eval_handshake.py:21
Functiontest_a_run_is_recorded_atomically_and_read_back
(project)
lib/crewai/tests/telemetry/test_last_run.py:28
Functiontest_a_run_no_app_evaluated_is_graded_by_the_command
Not every run goes through the app: a conversational session never ends the way a crew does, and a flow with human feedback takes the terminal
lib/cli/tests/experimental/test_eval_crew.py:1374
Functiontest_a_run_nobody_is_evaluating_is_not_this_rules_to_decide
()
lib/crewai-core/tests/test_eval_handshake.py:115
Functiontest_a_run_nobody_is_waiting_on_evaluates_nothing
()
lib/cli/tests/test_crew_run_tui.py:755
Functiontest_a_run_started_by_crewai_eval_leaves_the_untraced_count_to_the_command
The command waiting behind this screen says "not traced" and counts it; counting it here as well would count one stop twice.
lib/cli/tests/test_crew_run_tui.py:2967
Functiontest_a_run_started_for_an_evaluation_evaluates_it_here
`crewai eval` runs the crew when nothing is traced. The evaluation it came for starts on this screen when the run ends — the app does not close
lib/cli/tests/test_crew_run_tui.py:642
Functiontest_a_run_that_exits_cleanly_or_is_interrupted_is_not_a_failure
A clean exit and Ctrl-C (130) leave the command as they always did.
lib/cli/tests/experimental/test_eval_crew.py:4688
Functiontest_a_run_that_exits_with_an_error_counts_as_failed_not_untraced
( project, monkeypatch, capfd, usage, code, app_saw_the_run )
lib/cli/tests/experimental/test_eval_crew.py:4582
Functiontest_a_run_that_failed_is_still_evaluated
Where a run went wrong is what somebody asking for an evaluation wants to know, and the trace it exported is still there.
lib/cli/tests/test_crew_run_tui.py:735
Functiontest_a_run_that_leaves_no_trace_behind_is_explained
(project, monkeypatch, capsys)
lib/cli/tests/experimental/test_eval_crew.py:1512
Functiontest_a_run_that_left_no_trace_counts_as_untraced_once
Whether the run app looked and found nothing, or no app got to the run, the command is the one that counts it; the app counts nothing here.
lib/cli/tests/experimental/test_eval_crew.py:2742
Functiontest_a_run_that_recorded_nothing_says_so
Nothing was traced, so nothing was evaluated and the app had nothing to show: this is the one thing the command still has to say.
lib/cli/tests/experimental/test_eval_crew.py:1463
Functiontest_a_run_the_app_graded_exits_1_only_when_it_failed
The run app graded the run on its own screen and left its id in the marker: the command grades nothing more, and exits 1 when the run failed,
lib/cli/tests/experimental/test_eval_crew.py:4654
Functiontest_a_run_the_command_ran_first_is_the_last_run_and_ran_crew_carries_nothing_new
No app got to the run (no terminal), so the command grades the run it just made: a local run, `last_run`. `eval_ran_crew:*` is unchanged.
lib/cli/tests/experimental/test_eval_crew.py:1041
Functiontest_a_run_with_its_keys_set_or_a_re_run_counts_no_key_event
( project, usage, commands, monkeypatch )
lib/cli/tests/test_run_usage.py:1227
Functiontest_a_run_with_no_new_inputs_reads_the_url_the_last_inputs_gave
(downloads, entry)
lib/crewai/tests/agents/test_agent_spreadsheets.py:748
Functiontest_a_run_with_no_trace_says_so_instead_of_evaluating
(monkeypatch)
lib/cli/tests/test_crew_run_tui.py:710
Functiontest_a_run_with_nothing_traced_counts_why_the_crew_ran_and_how_it_ended
Three counts: why the crew ran, that the evaluation began, and its gate — so "reached a result" is counted rather than read off a missing stop.
lib/cli/tests/experimental/test_eval_crew.py:2718
Functiontest_a_same_provider_swap_keeps_the_declared_configuration
The mapped model is built like the declared llm, not from a bare string. A new instance of the same class, so everything derived from the model i
lib/crewai/tests/test_llm_overlay.py:369
Functiontest_a_save_cut_short_after_the_report_is_tried_again
report.md alone is not the evaluation saved: no marker, so the next read saves it again instead of pointing at half of it.
lib/cli/tests/experimental/test_eval_crew.py:4086
Functiontest_a_save_only_store_says_it_cannot_load
()
lib/crewai/tests/test_artifacts.py:920
Functiontest_a_save_that_fails_is_one_line_and_never_the_verdicts
The exit code stays the gate's: a failed gate is still 1, a passed one 0.
lib/cli/tests/experimental/test_eval_crew.py:3993
Functiontest_a_save_that_skipped_a_file_says_so_and_is_tried_again
Missing checks or a missing trace must not read as the full data.
lib/cli/tests/experimental/test_eval_crew.py:4068
Functiontest_a_saved_login_amp_rejects_is_replaced_and_the_deploy_retried
( monkeypatch, deploy_usage )
lib/cli/tests/test_crew_run_tui.py:2703
Functiontest_a_saved_login_that_cannot_be_read_counts_as_such
(project, monkeypatch, usage)
lib/cli/tests/experimental/test_eval_crew.py:2587
Methodtest_a_scan_status_amp_does_not_define_is_shown_as_unknown
( self, skill_command: SkillCommand, capsys: pytest.CaptureFixture[str] )
lib/cli/tests/skills/test_main.py:608
Methodtest_a_scoped_hook_returning_false_blocks_the_call
(self)
lib/crewai/tests/hooks/test_llm_hooks.py:654
Functiontest_a_second_file_with_the_same_name_is_answered_from_its_own_rows
(downloads, entry)
lib/crewai/tests/agents/test_agent_spreadsheets.py:1130
Functiontest_a_second_kickoff_never_reads_the_last_runs_plan
( downloads, monkeypatch, first, second )
lib/crewai/tests/agents/test_agent_spreadsheets.py:733
Functiontest_a_second_run_soon_after_reads_the_file_without_downloading_it_again
(downloads)
lib/crewai/tests/agents/test_agent_spreadsheets.py:1028
Methodtest_a_server_error_reports_api_5xx
( self, mock_input, mock_repository, mock_fetch_env )
lib/cli/tests/deploy/test_deploy_main.py:1190
Functiontest_a_service_that_does_not_export_yet_costs_nothing_and_says_nothing
(project, monkeypatch, capsys)
lib/cli/tests/experimental/test_eval_crew.py:3979
Functiontest_a_session_without_an_answer_says_why
( tmp_path, monkeypatch, returncode, stdout, stderr, reason )
lib/devtools/tests/test_agents_md_eval.py:360
Functiontest_a_setting_cannot_print_over_the_hosts_it_names
A control character in a setting could move the cursor back over the host and print the plain prompt in its place.
lib/cli/tests/test_api_keys.py:754
Functiontest_a_setting_the_provider_derived_from_the_model_is_not_carried
Anthropic fills `max_tokens` with the model's output cap when the caller did not set it; pinning one model's cap on another is a 400 waiting to
lib/crewai/tests/test_llm_overlay.py:408
Functiontest_a_shallow_span_that_does_not_parse_still_formats_json_inside_it
()
lib/cli/tests/test_crew_run_tui.py:2521
Methodtest_a_sheet_profile
(self)
lib/crewai-files/tests/spreadsheets/test_answers.py:170
Methodtest_a_sheet_that_declares_a_dtd_is_refused
( self, prolog, calamine_must_not_open )
lib/crewai-files/tests/spreadsheets/test_reading.py:329
Functiontest_a_shell_key_a_json_crews_env_file_replaces_is_asked_for
A JSON crew's run applies the `.env` line over the shell's key, so ask, naming the host the project's `.env` sends requests to.
lib/cli/tests/test_api_keys.py:559
Functiontest_a_single_message_matches_the_string_form
()
lib/crewai/tests/agents/test_agent_kickoff_message_history.py:181
Functiontest_a_span_with_every_content_attribute_at_the_bound_fits_one_wharf_request
Wharf refuses a request over 3,072,000 encoded bytes and the exporter drops a span that alone is over it. An LLM call carries the most content
lib/crewai/tests/telemetry/test_span_attribute_bound.py:317
Functiontest_a_span_with_many_large_attributes_is_shrunk_not_dropped
Each attribute fits its own bound, the span does not fit a request: the largest are cut, each marked, until it fits; nothing else changes.
lib/crewai/tests/telemetry/test_grant_export_bounds.py:136
Functiontest_a_spreadsheet_past_the_deadline_is_named_as_too_slow
(downloads, monkeypatch)
lib/crewai/tests/agents/test_agent_spreadsheets.py:1076
Functiontest_a_start_the_run_app_was_refused_is_not_tried_again_by_the_command
`crewai eval` runs the crew, the app it opens tries the evaluation and AMP refuses the start. The app has shown the refusal and counted it once; t
lib/cli/tests/experimental/test_eval_crew.py:2660
Functiontest_a_start_without_an_evaluation_id_counts_as_an_invalid_response
(project, monkeypatch, usage)
lib/cli/tests/experimental/test_eval_crew.py:2538
Functiontest_a_stop_first_seen_when_the_comparison_ends_is_still_said_and_its_words_stay_in_their_quote
(deployed, monkeypatch, capsys)
lib/cli/tests/experimental/test_eval_crew.py:1916
Functiontest_a_stop_outside_any_evaluation_carries_no_id
The run app's Evaluate, declined before anything started: nothing to join.
lib/cli/tests/experimental/test_eval_crew.py:1125
Functiontest_a_stop_the_service_cannot_place_is_said_after_the_wait_with_what_is_known
(deployed, monkeypatch, capsys)
lib/cli/tests/experimental/test_eval_crew.py:1865
Functiontest_a_stopped_setup_reaches_a_coding_agent_at_once_with_where_it_broke_and_the_error_quoted
(deployed, monkeypatch, capsys)
lib/cli/tests/experimental/test_eval_crew.py:1835
Methodtest_a_stray_value_far_down_one_column_still_reads
(self)
lib/crewai-files/tests/spreadsheets/test_reading.py:210
Methodtest_a_streak_completed_mid_batch_counts_once
(self, features: list[str])
lib/crewai/tests/tools/test_tool_failure.py:2493
Methodtest_a_streak_completed_mid_batch_still_forces_the_answer
The other call in the batch that hits the limit must not undo the stop.
lib/crewai/tests/tools/test_tool_failure.py:2132
Methodtest_a_streamed_tool_call_learns_it_too
(self, monkeypatch)
lib/crewai/tests/llms/openai/test_tools_reasoning_effort_retry.py:354
Functiontest_a_string_guardrail_never_fetches_a_url_in_the_output
(downloads)
lib/crewai/tests/agents/test_agent_spreadsheets.py:660
Functiontest_a_structured_litellm_call_is_seen_once
(seen_agents, monkeypatch)
lib/crewai/tests/hooks/test_model_call_hook_reach.py:110
Functiontest_a_subclass_of_llm_keeps_its_model
Only `LLM` itself routes; a subclass is the caller's own choice of class.
lib/crewai/tests/test_llm_overlay.py:742
Functiontest_a_succeeding_uv_script_returns
(monkeypatch)
lib/cli/tests/test_run_crew.py:1381
Methodtest_a_success_body_that_is_not_a_creation_reports_invalid_creation_response
A 2xx JSON body without a deployment uuid is a broken creation contract, not a success, and must not crash.
lib/cli/tests/deploy/test_deploy_main.py:1320
Methodtest_a_successful_create_reports_no_failure
( self, mock_input, mock_repository, mock_fetch_env )
lib/cli/tests/deploy/test_deploy_main.py:1412
Methodtest_a_sum_leaves_the_total_row_out_and_says_so
(self)
lib/crewai-files/tests/spreadsheets/test_answers.py:608
Methodtest_a_sum_over_text_says_what_it_left_out
(self)
lib/crewai-files/tests/spreadsheets/test_answers.py:556
Methodtest_a_summary_returned_alongside_a_private_result_is_published
Identity, not content: a different return value is still promoted.
lib/crewai/tests/test_flow_conversation.py:3618
Functiontest_a_symlink_already_in_the_folder_cannot_lead_a_file_out
(project, monkeypatch, capsys)
lib/cli/tests/experimental/test_eval_crew.py:3010
Methodtest_a_target_with_xml_escapes_is_refused
(self)
lib/crewai-files/tests/spreadsheets/test_reading.py:771
Functiontest_a_task_output_arrives_whole_under_both_keys
()
lib/crewai/tests/telemetry/test_span_attribute_bound.py:174
Functiontest_a_task_restored_from_a_checkpoint_never_reads_its_plan
( downloads, monkeypatch, tmp_path )
lib/crewai/tests/agents/test_agent_spreadsheets.py:767
Functiontest_a_task_with_a_spreadsheet_url_queries_every_row
(downloads, entry)
lib/crewai/tests/agents/test_agent_spreadsheets.py:322
Functiontest_a_task_with_no_agent_still_reports_a_failure_type
The no-agent branch raises inside the same try, so it must report too.
lib/crewai/tests/telemetry/test_task_failure_instrumentation.py:451
Functiontest_a_task_without_a_spreadsheet_url_is_unchanged
(downloads, monkeypatch)
lib/crewai/tests/agents/test_agent_spreadsheets.py:447
Functiontest_a_template_key_stops_matching_once_the_role_is_interpolated
The overlay maps the role's current text: the template is a key, the interpolated text is not, so a kickoff inside the block puts the declared
lib/crewai/tests/test_llm_overlay.py:278
Functiontest_a_template_role_that_is_itself_a_key_still_matches_at_construction
()
lib/crewai/tests/test_llm_overlay.py:219
Methodtest_a_terminal_event_without_a_start_is_inert
A listener attached mid-run has no stamp; that is not an error.
lib/crewai/tests/telemetry/test_crew_completed_span.py:169
Functiontest_a_terminal_that_cannot_be_handed_over_shows_the_refusal_as_before
( rejected_login, monkeypatch )
lib/cli/tests/test_crew_run_tui.py:3182
Functiontest_a_terminal_whose_output_is_captured_gets_the_line_instead
`crewai eval` sends a run's output to a log when nobody is watching it.
lib/cli/tests/test_api_keys.py:876
Functiontest_a_token_a_dotenv_could_set_approves_nothing
The variable alone is what a project's `.env` could carry; without the file `crewai eval` writes, it counts for nothing.
lib/crewai/tests/telemetry/test_deferred_trace_transitions.py:291
Functiontest_a_tool_call_sequence_survives
A `tool` message needs its preceding assistant `tool_calls` message. Filtering on truthy content dropped the assistant turn, leaving a sequence
lib/crewai/tests/agents/test_agent_kickoff_message_history.py:205
Methodtest_a_tool_call_then_the_follow_up_turn
(self, server, provider)
lib/crewai/tests/llms/openai_compatible/test_openai_compatible.py:596
Functiontest_a_tool_project_gets_no_evaluation_guidance
(mock_subprocess, tool_command)
lib/cli/tests/tools/test_main.py:88
Functiontest_a_tool_result_arrives_whole
(size: int)
lib/crewai/tests/telemetry/test_span_attribute_bound.py:124
Functiontest_a_tool_result_over_the_bound_is_cut_with_the_marker_and_its_size
()
lib/crewai/tests/telemetry/test_span_attribute_bound.py:133
Methodtest_a_total_row_is_flagged_by_its_label_or_its_sum
(self)
lib/crewai-files/tests/spreadsheets/test_answers.py:103
Methodtest_a_total_row_is_named_in_the_profile
(self)
lib/crewai-files/tests/spreadsheets/test_answers.py:203
Functiontest_a_trace_that_cannot_be_finished_never_replaces_the_runs_error
Finishing a failed run's trace is best effort: whatever sharing it or shutting it down raises is logged, and the run's own error is what escapes.
lib/crewai/tests/telemetry/test_session_trace_export.py:1043
Functiontest_a_trace_that_reaches_amp_is_counted_once_by_tier
Declined consent, a refused grant, or a rejected export never counts.
lib/crewai/tests/telemetry/test_session_trace_export.py:879
Functiontest_a_traced_run_that_exits_with_an_error_is_still_graded
( project, monkeypatch, capfd, usage )
lib/cli/tests/experimental/test_eval_crew.py:4634
Functiontest_a_traced_run_whose_script_failed_is_still_graded
`crewai run` exits with a failed script's code; `crewai eval` still grades that traced run, then exits 1 because the run failed.
lib/cli/tests/experimental/test_eval_crew.py:4360
Methodtest_a_transport_failure_reports_network_error_and_propagates
No response, so no status; the exception reaches the caller unchanged.
lib/cli/tests/deploy/test_deploy_main.py:1233
Functiontest_a_truncated_ephemeral_trace_is_not_counted
It is uploaded, but like the `crewai eval` record it is not a whole run.
lib/crewai/tests/telemetry/test_session_trace_export.py:950
← previousnext →10,501–10,600 of 19,663, ranked by callers