{"id":19365,"plugin_id":"plugins_6a97f99a25c08191a853f2805bcc6efb","kind":"skill","collection_source":null,"comparison_source":null,"observed_at":"2026-09-30T23:15:31.789Z","digest":"04b5750c7086e5b33669c18c980cc90f5f9dfa492516cc82f8bd7de9c409ac14","against":null,"payload":{"description":"AI DevKit · Supervise multi-agent workflows over repeated passes: poll progress, unblock waiting agents, coordinate dependencies, relay outputs, resolve conflicts, and verify completion. Use only for ongoing multi-agent coordination, not one-off list/detail/send/start/kill actions.","included_files":[{"relative_path":"agents/openai.yaml","size_in_bytes":406}],"name":"agent-orchestration","skill_md_contents":"---\nname: agent-orchestration\ndescription: \"AI DevKit · Supervise multi-agent workflows over repeated passes: poll progress, unblock waiting agents, coordinate dependencies, relay outputs, resolve conflicts, and verify completion. Use only for ongoing multi-agent coordination, not one-off list/detail/send/start/kill actions.\"\n---\n\n# Agent Orchestration\n\nUse only for multi-agent supervision: coordinating dependencies, polling progress, unblocking waiting agents, relaying outputs, resolving conflicts, and verifying completion across agents. For one-off list/detail/send/start/kill work, use `$agent-management` or `$agent-communication`.\n\nUse `$agent-management` for safe agent selection and lifecycle actions. Use `$agent-communication` for list/detail/send mechanics. Use `$verify` before accepting any agent's completion claim.\n\n## Rules\n\n- Own the loop until assigned work is complete, blocked, or stopped.\n- Run `agent list --json` before each pass; never assume names/statuses.\n- Inspect waiting, idle, unknown, missing, or stale agents before acting.\n- Send self-contained instructions and avoid duplicate follow-ups.\n- Sequence agents that touch the same files; relay only relevant upstream output.\n- Escalate only for repeated failures, unresolved conflicts, product/business decisions, or destructive/shared/production/security-sensitive actions.\n\n## Loop\n\nIf the goal or agent ownership is unclear, run one scan/detail pass. Ask the user once only if context is still insufficient.\n\n1. Scan agents.\n2. Assess agents needing attention with `detail --tail 10`.\n3. Act: approve, clarify, correct, delegate, relay, verify, or escalate.\n4. Report one brief status line.\n5. Sleep 10-60s and repeat.\n\n## Completion\n\nFinish when all assigned work is verified, blocked with a clear reason, or stopped by the user. Summarize per-agent outcomes, verification, unresolved issues, and next step.\n"},"changes":[],"summary":"First saved snapshot. No earlier version is available for comparison.","summary_kind":"deterministic","summary_metadata":{}}