{"id":20669,"plugin_id":"plugins_6aa73a9682ac8191bde61c8911b9a51a","kind":"skill","collection_source":null,"comparison_source":null,"observed_at":"2026-09-30T23:16:25.330Z","digest":"6e43e72a06dd7043bb6253e68d488d416f40f4467282fc1a9ef687f01139cfe0","against":null,"payload":{"description":"Recover authoritative current state before an agent changes code, data, configuration, or external state.","included_files":[],"name":"canonical-state-recovery","skill_md_contents":"---\nname: canonical-state-recovery\ndescription: Recover authoritative current state before an agent changes code, data, configuration, or external state.\n---\n\n# Canonical State Recovery\n\nUse before consequential work.\n\n1. Identify the authoritative system and exact scope.\n2. Record repository/project, branch, exact head/version, files, and environment.\n3. Separate CURRENT, HISTORICAL, CANDIDATE, and ARCHIVED state.\n4. Record decisions, blockers, negative memory, and next gate.\n5. Prefer fresh direct receipts over narratives or stale documents.\n6. Label missing evidence NOT_PROVEN or UNRESOLVED.\n7. Stop when authority, scope, or current state is ambiguous.\n\nRequired output: Objective, Authority, Current state, Decisions, Blockers, Next gate, Unresolved.\n\nNever promote prepared, historical, or candidate material to CURRENT without evidence.\n\nTruth rules:\nprepared != executed\nexecuted != verified\ncandidate != production\nhistorical != current\n"},"changes":[],"summary":"First saved snapshot. No earlier version is available for comparison.","summary_kind":"deterministic","summary_metadata":{}}