{"wiki":{"id":2515,"slug":"pm-s1","entity_type":"droid","entity_id":"pm-s1","title":"Tracker","prose_md":"## About\n\nTracker turns completed planner executions into clear task records that [[agt_011]] can review. It counts each agent’s passed, failed, and pending checks, then writes the results to [[carols-org]].\n\nOn demand, it extracts execution data, resolves agent IDs, creates per-agent records, and rolls up multi-level results. It avoids duplicate tasks, preserves the planner’s assignments and statuses, and skips unresolved agents. Tracker’s exact name is unique, but it belongs to the reusable intelligence family, which has 25 droids run across several agents. It is not pure software: it calls Claude using the opus model to turn execution results into agent task summaries.","namesake_json":"{\"engine\": \"claude\", \"model\": \"opus\", \"claude_purpose\": \"turn execution results into agent task summaries\", \"family\": \"intelligence\", \"family_size\": 25, \"is_unique\": true}","profile_pic_path":"","source_hash":"fb9a33da1c15a4607b4ad850568b093bf0f6e1880ec8ccb4bf1e968a0bf4c35e","status":"being_built","last_generated_at":"2026-08-04 01:39:23","created_at":"2026-08-01 03:28:43","updated_at":"2026-08-04 01:39:23"},"facts":{"id":"pm-s1","name":"Tracker","machine_name":"PM-S1","owner":"agt_011","function":"Syncs planner executions into agent_tasks in Org","process_type":"triggered","schedule":"On demand","process_name":"tracker","avatar_color":"#f59e0b","created_for":"Elrond is responsible for strategic oversight of how initiatives execute. PM-S1 serves this by ensuring that once work runs, the results are captured as task records in Org, giving Elrond visibility into what each agent accomplished.","purpose":"When an execution finishes, PM-S1 extracts the results from the planner, tallies how many checks each agent passed, failed, or left pending, and writes task records to Org. This transforms execution data into clear, trackable task records that Elrond can review.","duties":"- Extract execution records and check results from the planner database.\n- Tally results by assigned agent: passes, failures, and pending items.\n- Look up agent names in the registry and get their IDs.\n- Create a task record for each agent with their count and overall result (pass/fail/pending).\n- For multi-level executions, roll up sub-execution results into summary tasks.\n- Run in batch mode to sync all executions at once, updating all summary counts.","constraints":"- Does not create a new task if one already exists for that execution and agent.\n- Does not modify agent assignments — only records assignments the planner made.\n- Does not change execution results or status; copies them as-is to Org.\n- Silently skips agents if their name cannot be resolved in the registry.","status":"running","gender":"female","archetype":"intelligence","building_block":"support","service_override":null,"enabled":1,"task_key":"initiatives.blanket_planner","model_free":0},"page":{"type":"droid","page_class":"main","class_label":"Main page","kind_label":"Droid","kind_gloss":"","listed":true}}