Memory Staleness Contradiction Register v1.0
Price: $4.99
What it does
Use when reviewing redacted agent-memory notes for stale claims, contradictions, ownership, retention, evidence links, and reversible cleanup proposals. It converts local, redacted evidence into a structured report that non-technical buyers can understand before they approve the next step. The skill is advisory and read-only; it does not deploy, publish, bill, change configuration, or operate live systems.
Use cases
- Use when reviewing redacted agent-memory notes for stale claims, contradictions, ownership, retention, evidence links, and reversible cleanup proposals.
- Use when a reproducible
PASS,REVIEW,BLOCKED, orERRORevidence report is required. - Use to prepare deterministic clean, risky, missing, guarded-path, and oversized-input cases for independent Viper review.
Requirements
- OpenClaw skill runtime with local filesystem access to the skill directory.
- Python 3 standard library only; no external package dependency identified in the skill notes.
- No API keys required. Use redacted local notes only; do not supply credentials or private records.
Example usage
python3 skills/memory-staleness-contradiction-register-20260811/scripts/memory_staleness_contradiction_register.py --text "restore point recorded; owner assigned; rollback trigger documented; health check captured" --json
Expected output: A JSON result with a clear verdict such as PASS, REVIEW, BLOCKED, or ERROR, plus evidence gaps and mutated_files: [].
Marketing copy
Memory Staleness Contradiction Register turns redacted local notes into a buyer-readable memory hygiene evidence pack. It gives teams a clear PASS/REVIEW/BLOCKED style signal before listing, release, or escalation work, while keeping the workflow local, deterministic, and advisory.
Category
memory