From memesh
Reviews MeMesh memory database: computes health score, detects stale/conflicting/redundant memories, analyzes work patterns, suggests CLI cleanup actions.
How this skill is triggered — by the user, by Claude, or both
Slash command
/memesh:memesh-reviewThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Review the memory database and provide actionable cleanup recommendations.
Review the memory database and provide actionable cleanup recommendations.
Use CLI (works everywhere) or MCP tools (if available). See the memesh skill for auto-detect instructions.
# Get system health
memesh status
# Get all recent memories (structured output for analysis)
memesh recall --limit 50 --json
# Get memories by type for quality analysis
memesh recall --tag "type:decision" --json
memesh recall --tag "type:lesson_learned" --json
memesh recall --tag "type:session_keypoint" --json
If MCP user_patterns tool is available, also run it for work pattern analysis:
user_patterns: {}
From the recalled data, compute and present:
## Memory Health Report
### Overview
- Total entities: N
- Last 30 days active: N (N%)
- Knowledge types: N decisions, N patterns, N lessons, N auto-tracked
### Health Score: N/100
- Activity: N% (accessed in last 30 days)
- Quality: N% (high confidence, well-tagged)
- Freshness: N% (new this week)
- Self-Improvement: N% (lessons learned ratio)
### Quality Issues Found
**Stale (not accessed 30+ days, low confidence)**
- "entity-name" — confidence: N% — Suggest: archive?
**Verbose (5+ observations, needs consolidation)**
- "entity-name" (N observations) — Suggest: `memesh consolidate --name "entity-name"`
**Potential conflicts**
- "entity-A" vs "entity-B" — contradicting decisions
**Noise ratio**
- N% auto-tracked (session_keypoint, commit) vs N% intentional knowledge
- If noise > 80%: recommend more deliberate `memesh remember` usage
### Recommended Actions
1. `memesh forget --name "old-design"` (superseded)
2. `memesh consolidate --name "auth-history"` (12 obs → ~3)
3. `memesh remember ...` (knowledge gap in [area])
Present the report first. Ask which actions to execute. Then run the commands:
memesh forget --name "outdated-entity"
memesh consolidate --name "verbose-entity"
memesh remember --name "missing-knowledge" --type decision --obs "..."
memesh recall --limit 5 --json # confirm changes took effect
For comprehensive documentation synchronization and lint checks, use:
@sa:comprehensive-code-review
This skill includes:
MeMesh-specific automation:
# Quick verification (memesh-specific checks)
./scripts/verify-docs-sync.sh
# Exit code 0 = all checks pass
# Lint check
npm run lint # 0 errors expected, ~83 warnings (technical debt)
memesh remember for decisionsmemesh serve first)npx claudepluginhub pcircle-ai/memesh-llm-memory --plugin memeshAudits Claude Code auto-memory (MEMORY.md, topic files) for promotion candidates to CLAUDE.md, stale entries, duplicates, consolidation opportunities, and health metrics.
Manages persistent AI memory across coding sessions: remember decisions/patterns, recall context, learn from bugs/mistakes. Auto-triggers on memory prompts, key events like decisions/fixes via Claude Code hooks.
Audits memory quality across six dimensions (purity, freshness, coverage, clarity, relevance, structure) with prioritized findings and actionable recommendations.