mk:memory — independently scanned and version-tracked by SaferSkills.
SaferSkills independently audited mk:memory (Agent Skill) and scored it 100/100 (green). The audit ran 55 deterministic rules across Security, Supply Chain, Maintenance, Transparency, and Community; it found 0 high-severity and 0 lower-severity findings. The full rule-by-rule trace and per-finding evidence are below. Free, methodology-open.
Findings & checks · 0 flagged
Every scanned point with the score it earned and what moved between them.
First recorded scan — no prior version to compare against.
The primary manifest — the file an agent reads to learn what this artifact does.
Reference guides for the memory system: session capture, pattern extraction, and cost tracking.
analyst agent tracks costs or extracts patternsOperates in Phase 0 (Orient) and Phase 6 (Reflect). Output supports the analyst agent.
| Reference | When to load | Content |
|---|---|---|
| [capture-architecture.md](./references/capture-architecture.md) | Before any agent-side memory write | The 2-path contract — ##prefix: is user-typed only; agents write via direct Edit |
| [session-capture.md](./references/session-capture.md) | Phase 6 | Capturing session learnings in 3 categories (patterns/decisions/failures) |
| [pattern-extraction.md](./references/pattern-extraction.md) | Phase 6 | Extracting high-frequency patterns for instruction-file promotion |
| [cost-tracking.md](./references/cost-tracking.md) | Phase 0, 6 | Token usage tracking, cost reporting, budget alerts |
| [consolidation.md](./references/consolidation.md) | Manual | Prune stale entries, merge duplicates, archive cost data (run when memory grows large) |
session-capture is not a CLI subcommand. At Phase 6 (Reflect) the agent reads references/session-capture.md and follows its 4 steps using Read / Write / Edit directly. There is no mewkit memory session-capture script — content extraction requires LLM analysis that a static CLI cannot produce.
--pruneArchive old standard-severity entries from topic files to lessons-archive.md.
Topic files subject to pruning:
fixes.md — bug-class session learningsreview-patterns.md — process/review learningsarchitecture-decisions.md — architectural decision recordsWhat gets pruned:
## headings with a date (YYYY-MM-DD, severity: standard) older than threshold (default: 90 days)What is exempt:
severity: critical or severity: security entries — never prunedHow to run:
/mk:memory --prune # default 90-day threshold
/mk:memory --prune --days 180 # custom threshold
/mk:memory --prune --dry-run # show what would be archived without movingMechanism (grep-based — no parser dependency):
## headings + date pattern (YYYY-MM-DD, severity: standard).claude/memory/lessons-archive.mdRecovery: Copy entries from lessons-archive.md back to the appropriate topic file.
Split JSON files (fixes.json, review-patterns.json, architecture-decisions.json) all use v2.0.0 schema with fields: version, scope, consumer, patterns[], metadata. Each pattern entry supports: id, type, category, severity, domain, applicable_when, context, pattern, frequency, lastSeen.
~30 seconds. Free. No account. Every finding cites a rule and a line of evidence.