Handbook
Task catalog (generated draft)
Updated
Task catalog (generated draft)
| Group | task_id | uses_llm | schema_depth | has_examples | supports_fake_transport | live_optional | Contract |
|---|---|---|---|---|---|---|---|
| Games | board_game_choose_move |
yes | L2 | yes | yes | yes | board_game_choose_move |
| Games | board_game_coach_notes |
yes | L2 | yes | yes | yes | board_game_coach_notes |
| Games | board_game_rank_moves |
yes | L2 | yes | yes | yes | board_game_rank_moves |
| Games | game_move_explain |
yes | L2 | yes | yes | yes | game_move_explain |
| Games | game_move_parse |
yes | L2 | yes | yes | yes | game_move_parse |
| Games | game_move_rank |
yes | L2 | yes | yes | yes | game_move_rank |
| Generic analysis and planning | analyze_strategic_form |
yes | L2 | yes | yes | yes | analyze_strategic_form |
| Generic analysis and planning | automation_situation_arbitrate |
yes | L2 | yes | yes | yes | automation_situation_arbitrate |
| Generic analysis and planning | contradiction_scan |
yes | L2 | yes | yes | yes | contradiction_scan |
| Generic analysis and planning | decompose_problem |
yes | L2 | yes | yes | yes | decompose_problem |
| Generic analysis and planning | extract_quantities_and_units |
yes | L2 | yes | yes | yes | extract_quantities_and_units |
| Generic analysis and planning | extract_schema_from_text |
yes | L2 | yes | yes | yes | extract_schema_from_text |
| Generic analysis and planning | generate_script_prompt |
yes | L2 | yes | yes | yes | generate_script_prompt |
| Generic analysis and planning | llm_boolean_gate |
yes | L2 | yes | yes | yes | llm_boolean_gate |
| Generic analysis and planning | llm_enum_route |
yes | L2 | yes | yes | yes | llm_enum_route |
| Generic analysis and planning | llm_multi_label |
yes | L2 | yes | yes | yes | llm_multi_label |
| Generic analysis and planning | mechanism_compare_prose |
yes | L2 | yes | yes | yes | mechanism_compare_prose |
| Generic analysis and planning | outlook_apply_read_layout |
no | L2 | yes | yes | no | outlook_apply_read_layout |
| Generic analysis and planning | outlook_download_message_eml |
no | L2 | yes | yes | no | outlook_download_message_eml |
| Generic analysis and planning | outlook_ensure_mail_surface |
no | L2 | yes | yes | no | outlook_ensure_mail_surface |
| Generic analysis and planning | outlook_fetch_message_rfc822 |
no | L2 | yes | yes | no | outlook_fetch_message_rfc822 |
| Generic analysis and planning | outlook_list_folders |
no | L2 | yes | yes | no | outlook_list_folders |
| Generic analysis and planning | outlook_list_inbox_chunk |
no | L2 | yes | yes | no | outlook_list_inbox_chunk |
| Generic analysis and planning | outlook_list_inbox_dom_step |
no | L2 | yes | yes | no | outlook_list_inbox_dom_step |
| Generic analysis and planning | outlook_list_inbox_rest |
no | L2 | yes | yes | no | outlook_list_inbox_rest |
| Generic analysis and planning | outlook_list_messages_scroll |
no | L2 | yes | yes | no | outlook_list_messages_scroll |
| Generic analysis and planning | outlook_list_messages_viewport |
no | L2 | yes | yes | no | outlook_list_messages_viewport |
| Generic analysis and planning | outlook_normalize_item_url |
no | L2 | yes | yes | no | outlook_normalize_item_url |
| Generic analysis and planning | outlook_probe_mail_health |
no | L2 | yes | yes | no | outlook_probe_mail_health |
| Generic analysis and planning | outlook_read_conversation_thread |
no | L2 | yes | yes | no | outlook_read_conversation_thread |
| Generic analysis and planning | plan_decision_pack |
yes | L2 | yes | yes | yes | plan_decision_pack |
| Generic analysis and planning | playwright_extract_file |
no | L2 | yes | yes | no | playwright_extract_file |
| Generic analysis and planning | playwright_normalize_sharepoint_url |
no | L2 | yes | yes | no | playwright_normalize_sharepoint_url |
| Generic analysis and planning | playwright_read_items |
no | L2 | yes | yes | no | playwright_read_items |
| Generic analysis and planning | playwright_render_feed |
no | L2 | yes | yes | no | playwright_render_feed |
| Generic analysis and planning | playwright_sharepoint_page_health |
no | L2 | yes | yes | no | playwright_sharepoint_page_health |
| Generic analysis and planning | sharepoint_list_library_files_columns |
no | L2 | yes | yes | no | sharepoint_list_library_files_columns |
| Generic analysis and planning | sharepoint_list_library_files_rest |
no | L2 | yes | yes | no | sharepoint_list_library_files_rest |
| Generic analysis and planning | sharepoint_list_library_files_scroll |
no | L2 | yes | yes | no | sharepoint_list_library_files_scroll |
| Generic analysis and planning | sharepoint_list_library_files_viewport |
no | L2 | yes | yes | no | sharepoint_list_library_files_viewport |
| Generic analysis and planning | sharepoint_normalize_library_url |
no | L2 | yes | yes | no | sharepoint_normalize_library_url |
| Generic analysis and planning | sharepoint_probe_library_health |
no | L2 | yes | yes | no | sharepoint_probe_library_health |
| Generic analysis and planning | sharepoint_read_messages_index |
no | L2 | yes | yes | no | sharepoint_read_messages_index |
| Generic analysis and planning | sharepoint_verify_outlook_archive |
no | L2 | yes | yes | no | sharepoint_verify_outlook_archive |
| Generic analysis and planning | summarize_sections_schema |
yes | L2 | yes | yes | yes | summarize_sections_schema |
| Generic analysis and planning | teams_list_channel_posts_chunk |
no | L2 | yes | yes | no | teams_list_channel_posts_chunk |
| Generic analysis and planning | teams_list_channels |
no | L2 | yes | yes | no | teams_list_channels |
| Generic analysis and planning | teams_list_chat_messages_chunk |
no | L2 | yes | yes | no | teams_list_chat_messages_chunk |
| Generic analysis and planning | teams_list_chats |
no | L2 | yes | yes | no | teams_list_chats |
| Generic analysis and planning | teams_probe_health |
no | L2 | yes | yes | no | teams_probe_health |
| Generic analysis and planning | teams_read_thread_replies |
no | L2 | yes | yes | no | teams_read_thread_replies |
| Generic analysis and planning | teams_search_harvest_closed_day |
no | L2 | yes | yes | no | teams_search_harvest_closed_day |
| Generic analysis and planning | teams_search_messages_chunk |
no | L2 | yes | yes | no | teams_search_messages_chunk |
| Generic analysis and planning | teams_search_prepare_day_scope |
no | L2 | yes | yes | no | teams_search_prepare_day_scope |
| Generic analysis and planning | teams_search_remediate_drift |
no | L2 | yes | yes | no | teams_search_remediate_drift |
| Generic analysis and planning | timeline_extract |
yes | L2 | yes | yes | yes | timeline_extract |
| Generic analysis and planning | word_problem_to_calc_plan |
yes | L2 | yes | yes | yes | word_problem_to_calc_plan |
| Playwright / page intelligence | interaction_probe |
no | L2 | yes | yes | no | interaction_probe |
| Playwright / page intelligence | page_mechanics |
no | L2 | yes | yes | no | page_mechanics |
| Playwright / page intelligence | page_probe |
no | L2 | yes | yes | no | page_probe |
| Playwright / page intelligence | pw_chunk_classify |
yes | L2 | yes | yes | yes | pw_chunk_classify |
| Playwright / page intelligence | pw_extractor_synthesize_exemplar |
yes | L2 | yes | yes | yes | pw_extractor_synthesize_exemplar |
| Playwright / page intelligence | pw_extractor_synthesize_probe |
yes | L2 | yes | yes | yes | pw_extractor_synthesize_probe |
| Playwright / page intelligence | pw_grading_signal_infer |
yes | L2 | yes | yes | yes | pw_grading_signal_infer |
| Playwright / page intelligence | pw_incremental_diagnose |
yes | L2 | yes | yes | yes | pw_incremental_diagnose |
| Playwright / page intelligence | pw_mechanics_repair |
yes | L2 | yes | yes | yes | pw_mechanics_repair |
| Playwright / page intelligence | pw_network_api_route_infer |
yes | L2 | yes | yes | yes | pw_network_api_route_infer |
| Playwright / page intelligence | pw_page_kind_route |
yes | L2 | yes | yes | yes | pw_page_kind_route |
| Playwright / page intelligence | pw_quiz_mechanics_discover |
yes | L2 | yes | yes | yes | pw_quiz_mechanics_discover |
| Playwright / page intelligence | pw_selector_harden |
yes | L2 | yes | yes | yes | pw_selector_harden |
| Playwright / page intelligence | pw_static_mcq_mechanics_discover |
yes | L2 | yes | yes | yes | pw_static_mcq_mechanics_discover |
| RAG and evidence | answer_from_evidence |
yes | L2 | yes | yes | yes | answer_from_evidence |
| RAG and evidence | inject_citations |
yes | L2 | yes | yes | yes | inject_citations |
| RAG and evidence | rag_enough_context_gate |
yes | L2 | yes | yes | yes | rag_enough_context_gate |
| RAG and evidence | rag_query_plan |
yes | L2 | yes | yes | yes | rag_query_plan |
| Validation | validation_failure |
no | L2 | yes | yes | no | validation_failure |