Self-Hosted Memory Confidence Kit 20260820 v1.0
What it does
Self-Hosted Memory Confidence Kit helps teams decide whether a self-hosted memory setup has enough evidence to rely on. It checks backend identity, deployment mode, data model, write and read tests, retrieval quality, retention, privacy boundaries, backup, restore, migration, and rollback evidence.
The skill turns redacted notes into a practical confidence report for memory infrastructure. It is advisory, deterministic, local-only, and read-only.
Use cases
- Engineering teams validating a self-hosted memory backend before production use.
- Implementation teams collecting backup, restore, migration, and rollback evidence.
- Buyers comparing whether a memory stack is ready, incomplete, or blocked.
Requirements
- Python 3 with standard-library modules only.
- Redacted inline text or one explicit local notes file.
- No API keys required.
- Compatible with the standard OpenClaw skill layout and local script execution.
Example usage
python3 scripts/self_hosted_memory_confidence_kit.py --text "backend identifier documented; deployment mode documented; data model documented; write test documented; read test documented; retrieval quality documented; retention policy documented; privacy boundary documented; backup evidence documented; restore test documented; migration plan documented; rollback plan documented" --json
Expected output: JSON with result: "PASS", memory evidence coverage, and mutated_files: []. Weak or incomplete evidence returns REVIEW; serious risk phrases or low coverage return BLOCKED.
Price
$4.99