Agent Memory Lifecycle Manager 20260712 v1.0

What It Does

Agent Memory Lifecycle Manager 20260712 turns memory and learning-management notes into an auditable lifecycle report. It checks whether agent memory practices include capture rules, review cadence, owner approval, retention period, expiry date, rollback point, change evidence, sensitive-data boundaries, audit trail, and learning summaries.

The skill is designed for teams that need control over what agents learn over time. It reports PASS / REVIEW / BLOCKED locally and never edits memory, config, cron, billing, deployment, or live services.

Use Cases

Organizations that need evidence-based governance for agent memory and learning.
Operators preparing memory updates for review, retention, expiry, and rollback decisions.
QA teams checking whether memory practices have owner approval and audit evidence.

Requirements

Python 3 with standard library only.
No API keys required.
Local, non-sensitive memory lifecycle notes.
OpenClaw version: compatible with local skill execution in current OpenClaw skill format.

Example Usage


python3 scripts/agent_memory_lifecycle_manager.py --text "Capture rule, review cadence, owner approval, retention period, expiry date, rollback point, change evidence, sensitive-data boundary, audit trail, and learning summary are documented." --json

Expected output:


{
  "ok": true,
  "status": "PASS",
  "score": 100,
  "signals_found": ["capture rule", "review cadence", "owner approval"],
  "risks_found": [],
  "mutated_files": []
}

Price

$4.99