# State Update Skill Maintain accurate CURRENT_STATE.json for agent handoff and session continuity. ## Usage ``` /state-update [context] ``` **Context:** Brief description of what triggered the update (capture, proving, fix, gap, session-end) ## Workflow ### 1. Read Current State Always read CURRENT_STATE.json before modifying: ```bash cat CURRENT_STATE.json ``` Verify JSON is valid before proceeding. ### 2. Apply Updates Based on Context #### After Evidence Capture Update evidence section: ```json { "evidence": { "status": "FRESH", "sessionId": "", "captureDate": "", "capturedSwaps": { ... } } } ``` #### After Proving Run Update venue baseline: ```json { "venueStatus": { "": { "truePassRate": "%", "baselineDate": "", "baselineDetails": { ... } } } } ``` #### After Fix Applied Mark issue as FIXED: ```json { "venueStatus": { "": { "issues": [ { "id": "", "status": "FIXED", "fixedDate": "", "fix": "" } ], "completedFixes": [ "" ] } } } ``` #### After Gap Found Add new issue: ```json { "venueStatus": { "": { "issues": [ { "id": "", "title": "", "description": "", "status": "identified", "priority": "high|medium|low", "file": "", "lines": "" } ] } } } ``` #### Before Session End Update session metadata: ```json { "lastUpdated": "", "updatedBy": "Claude ", "sessionSummary": "<1-2 sentence summary of what was accomplished>" } ``` Update nextActions if priorities changed. ### 3. Validate Update After any modification: - Ensure JSON is valid (parseable) - Ensure required fields are present - Ensure no workaround references added - Ensure terminology follows guidelines (no "simulation", "phase", etc.) ## Required Fields | Field | Required | Description | |-------|----------|-------------| | lastUpdated | Yes | Date of last update | | updatedBy | Yes | Who updated (model name) | | sessionSummary | Yes | Brief summary | | activeLayer | Yes | Current layer (1-4) | | currentVenue | Yes | Current venue being worked | | venueStatus | Yes | Status per venue | | nextActions | Yes | Prioritized action list | ## Terminology Enforcement **Use:** - Local cache state - Proving - Math engines - Layer 1 fix - Gap - Pool decoder / Instruction decoder **Avoid:** - Simulation - Phase - Workaround - Bandaid ## Key Files | File | Purpose | |------|---------| | `CURRENT_STATE.json` | Primary state file | | `MENTAL_MAP.md` | Architecture reference | | `LAYERS.md` | Layer definitions |

State Update Skill

Scan your own AI components