Analyze auto-memory for promotion candidates, stale entries
/si:memory-review — Analyze Auto-Memory

Analyze auto-memory for promotion candidates, stale entries, consolidation opportunities, and health metrics. Use when the user runs /si:memory-review or asks what has been learned and what should be promoted or pruned.
--- name: "memory-review" description: "Analyze auto-memory for promotion candidates, stale entries, consolidation opportunities, and health metrics. Use when the user runs /si:memory-review or asks what has been learned and what should be promoted or pruned." --- # /si:memory-review — Analyze Auto-Memory Performs a comprehensive audit of Claude Code's auto-memory and produces actionable recommendations. ## Usage ``` /si:memory-review # Full review /si:memory-review --quick # Summary only (counts + top 3 candidates) /si:memory-review --stale # Focus on stale/outdated entries /si:memory-review --candidates # Show only promotion candidates ```
Overview
Performs a comprehensive audit of Claude Code's auto-memory and produces actionable recommendations.
Manual install (npx, ZIP)
npx skills add alirezarezvani/claude-skills --skill memory-review -g -a claude-codemkdir -p ~/.claude/skills && curl -sL https://codeload.github.com/alirezarezvani/claude-skills/tar.gz/19392f7a0826 \
| tar -xz -C ~/.claude/skills --strip-components=4 claude-skills-19392f7a0826/engineering-team/self-improving-agent/skills/memory-reviewNew-Item -ItemType Directory -Force "$HOME\.claude\skills" | Out-Null; Invoke-WebRequest "https://m.agentalley.io/zip/260f9e316d327b0119a9/memory-review.zip" -OutFile "$env:TEMP\memory-review.zip"; Expand-Archive "$env:TEMP\memory-review.zip" -DestinationPath "$HOME\.claude\skills" -ForceConnects with Agent Alley MCP · agentalley.io/mcp
Files included
SKILL.md4.2 KB
Read the files here
alirezarezvani/
Shared references2SKILL.md reads these from elsewhere in the repo. Not in the ZIP.
Get them on GitHub ↗Show the full text127 lines
/si:memory-review — Analyze Auto-Memory
Performs a comprehensive audit of Claude Code's auto-memory and produces actionable recommendations.
Usage
/si:memory-review # Full review
/si:memory-review --quick # Summary only (counts + top 3 candidates)
/si:memory-review --stale # Focus on stale/outdated entries
/si:memory-review --candidates # Show only promotion candidates
What It Does
Step 1: Locate memory directory
# Find the project's auto-memory directory
MEMORY_DIR="$HOME/.claude/projects/$(pwd | sed 's|/|%2F|g; s|%2F|/|; s|^/||')/memory"
# Fallback: check common path patterns
# ~/.claude/projects/<user>/<project>/memory/
# ~/.claude/projects/<absolute-path>/memory/
# List all memory files
ls -la "$MEMORY_DIR"/
If memory directory doesn't exist, report that auto-memory may be disabled. Suggest checking with /memory.
Step 2: Read and analyze MEMORY.md
Read the full MEMORY.md file. Count lines and check against the 200-line startup limit.
Analyze each entry for:
Recurrence indicators
- Same concept appears multiple times (different wording)
- References to "again" or "still" or "keeps happening"
- Similar entries across topic files
Staleness indicators
- References files that no longer exist (
findto verify) - Mentions outdated tools, versions, or commands
- Contradicts current CLAUDE.md rules
- References files that no longer exist (
Consolidation opportunities
- Multiple entries about the same topic (e.g., three lines about testing)
- Entries that could merge into one concise rule
Promotion candidates — entries that meet ALL criteria:
- Appeared in 2+ sessions (check wording patterns)
- Not project-specific trivia (broadly useful)
- Actionable (can be written as a concrete rule)
- Not already in CLAUDE.md or
.claude/rules/
Step 3: Read topic files
If MEMORY.md references or the directory contains additional files (debugging.md, patterns.md, etc.):
- Read each one
- Cross-reference with MEMORY.md for duplicates
- Check for entries that belong in the main file (high value) vs. topic files (details)
Step 4: Cross-reference with CLAUDE.md
Read the project's CLAUDE.md (if it exists) and compare:
- Are there MEMORY.md entries that duplicate CLAUDE.md rules? (→ remove from memory)
- Are there MEMORY.md entries that contradict CLAUDE.md? (→ flag conflict)
- Are there MEMORY.md patterns not yet in CLAUDE.md that should be? (→ promotion candidate)
Also check .claude/rules/ directory for existing scoped rules.
Step 5: Generate report
Output format:
📊 Auto-Memory Review
Memory Health:
MEMORY.md: {{lines}}/200 lines ({{percent}}%)
Topic files: {{count}} ({{names}})
CLAUDE.md: {{lines}} lines
Rules: {{count}} files in .claude/rules/
🎯 Promotion Candidates ({{count}}):
1. "{{pattern}}" — seen {{n}}x, applies broadly
→ Suggest: {{target}} (CLAUDE.md / .claude/rules/{{name}}.md)
2. ...
🗑️ Stale Entries ({{count}}):
1. Line {{n}}: "{{entry}}" — {{reason}}
2. ...
🔄 Consolidation ({{count}} groups):
1. Lines {{a}}, {{b}}, {{c}} all about {{topic}} → merge into 1 entry
2. ...
⚠️ Conflicts ({{count}}):
1. MEMORY.md line {{n}} contradicts CLAUDE.md: {{detail}}
💡 Recommendations:
- {{actionable suggestion}}
- {{actionable suggestion}}
When to Use
- After completing a major feature or debugging session
- When
/si:memory-statusshows MEMORY.md is over 150 lines - Weekly during active development
- Before starting a new project phase
- After onboarding a new team member (review what Claude learned)
Tips
- Run
/si:memory-review --quickfrequently (low overhead) - Full review is most valuable when MEMORY.md is getting crowded
- Act on promotion candidates promptly — they're proven patterns
- Don't hesitate to delete stale entries — auto-memory will re-learn if needed
| 1 | |
| 2 | name "memory-review" |
| 3 | description "Analyze auto-memory for promotion candidates, stale entries, consolidation opportunities, and health metrics. Use when the user runs /si:memory-review or asks what has been learned and what should be promoted or pruned." |
| 4 | |
| 5 | |
| 6 | # /si:memory-review — Analyze Auto-Memory |
| 7 | |
| 8 | Performs a comprehensive audit of Claude Code's auto-memory and produces actionable recommendations. |
| 9 | |
| 10 | ## Usage |
| 11 | |
| 12 | |
| 13 | /si:memory-review # Full review |
| 14 | /si:memory-review --quick # Summary only (counts + top 3 candidates) |
| 15 | /si:memory-review --stale # Focus on stale/outdated entries |
| 16 | /si:memory-review --candidates # Show only promotion candidates |
| 17 | |
| 18 | |
| 19 | ## What It Does |
| 20 | |
| 21 | ### Step 1: Locate memory directory |
| 22 | |
| 23 | |
| 24 | # Find the project's auto-memory directory |
| 25 | MEMORY_DIR="$HOME/.claude/projects/$(pwd | sed 's|/|%2F|g; s|%2F|/|; s|^/||')/memory" |
| 26 | |
| 27 | # Fallback: check common path patterns |
| 28 | # ~/.claude/projects/<user>/<project>/memory/ |
| 29 | # ~/.claude/projects/<absolute-path>/memory/ |
| 30 | |
| 31 | # List all memory files |
| 32 | ls -la "$MEMORY_DIR"/ |
| 33 | |
| 34 | |
| 35 | If memory directory doesn't exist, report that auto-memory may be disabled. Suggest checking with `/memory`. |
| 36 | |
| 37 | ### Step 2: Read and analyze MEMORY.md |
| 38 | |
| 39 | Read the full `MEMORY.md` file. Count lines and check against the 200-line startup limit. |
| 40 | |
| 41 | Analyze each entry for: |
| 42 | |
| 43 | **Recurrence indicators** |
| 44 | Same concept appears multiple times (different wording) |
| 45 | References to "again" or "still" or "keeps happening" |
| 46 | Similar entries across topic files |
| 47 | |
| 48 | **Staleness indicators** |
| 49 | References files that no longer exist (`find` to verify) |
| 50 | Mentions outdated tools, versions, or commands |
| 51 | Contradicts current CLAUDE.md rules |
| 52 | |
| 53 | **Consolidation opportunities** |
| 54 | Multiple entries about the same topic (e.g., three lines about testing) |
| 55 | Entries that could merge into one concise rule |
| 56 | |
| 57 | **Promotion candidates** — entries that meet ALL criteria: |
| 58 | Appeared in 2+ sessions (check wording patterns) |
| 59 | Not project-specific trivia (broadly useful) |
| 60 | Actionable (can be written as a concrete rule) |
| 61 | Not already in CLAUDE.md or `.claude/rules/` |
| 62 | |
| 63 | ### Step 3: Read topic files |
| 64 | |
| 65 | If `MEMORY.md` references or the directory contains additional files (`debugging.md`, `patterns.md`, etc.): |
| 66 | Read each one |
| 67 | Cross-reference with MEMORY.md for duplicates |
| 68 | Check for entries that belong in the main file (high value) vs. topic files (details) |
| 69 | |
| 70 | ### Step 4: Cross-reference with CLAUDE.md |
| 71 | |
| 72 | Read the project's `CLAUDE.md` (if it exists) and compare: |
| 73 | Are there MEMORY.md entries that duplicate CLAUDE.md rules? (→ remove from memory) |
| 74 | Are there MEMORY.md entries that contradict CLAUDE.md? (→ flag conflict) |
| 75 | Are there MEMORY.md patterns not yet in CLAUDE.md that should be? (→ promotion candidate) |
| 76 | |
| 77 | Also check `.claude/rules/` directory for existing scoped rules. |
| 78 | |
| 79 | ### Step 5: Generate report |
| 80 | |
| 81 | Output format: |
| 82 | |
| 83 | |
| 84 | 📊 Auto-Memory Review |
| 85 | |
| 86 | Memory Health: |
| 87 | MEMORY.md: {{lines}}/200 lines ({{percent}}%) |
| 88 | Topic files: {{count}} ({{names}}) |
| 89 | CLAUDE.md: {{lines}} lines |
| 90 | Rules: {{count}} files in .claude/rules/ |
| 91 | |
| 92 | 🎯 Promotion Candidates ({{count}}): |
| 93 | 1. "{{pattern}}" — seen {{n}}x, applies broadly |
| 94 | → Suggest: {{target}} (CLAUDE.md / .claude/rules/{{name}}.md) |
| 95 | 2. ... |
| 96 | |
| 97 | 🗑️ Stale Entries ({{count}}): |
| 98 | 1. Line {{n}}: "{{entry}}" — {{reason}} |
| 99 | 2. ... |
| 100 | |
| 101 | 🔄 Consolidation ({{count}} groups): |
| 102 | 1. Lines {{a}}, {{b}}, {{c}} all about {{topic}} → merge into 1 entry |
| 103 | 2. ... |
| 104 | |
| 105 | ⚠️ Conflicts ({{count}}): |
| 106 | 1. MEMORY.md line {{n}} contradicts CLAUDE.md: {{detail}} |
| 107 | |
| 108 | 💡 Recommendations: |
| 109 | - {{actionable suggestion}} |
| 110 | - {{actionable suggestion}} |
| 111 | |
| 112 | |
| 113 | ## When to Use |
| 114 | |
| 115 | After completing a major feature or debugging session |
| 116 | When `/si:memory-status` shows MEMORY.md is over 150 lines |
| 117 | Weekly during active development |
| 118 | Before starting a new project phase |
| 119 | After onboarding a new team member (review what Claude learned) |
| 120 | |
| 121 | ## Tips |
| 122 | |
| 123 | Run `/si:memory-review --quick` frequently (low overhead) |
| 124 | Full review is most valuable when MEMORY.md is getting crowded |
| 125 | Act on promotion candidates promptly — they're proven patterns |
| 126 | Don't hesitate to delete stale entries — auto-memory will re-learn if needed |
| 127 |
Details
- Updated
- Aug 30, 2026
- License
- MIT
- Stars on the repo
- ★ 26,349
- Checked
- Source
- alirezarezvani/claude-skills ↗
- Creator
- alirezarezvani · 363 listings
Similar skills
See all →
SkillExport the Obsidian wiki's knowledge graph to structuredWiki Export — Knowledge Graph Export
SkillImport a wiki knowledge graph into the current vaultWiki Import — Reconstruct Pages from an Export
SkillYour team's shared, AI-ready knowledge base/company-brain — Team-shared AI-ready knowledge base
SkillGive Claude a permanent, searchable long-term memoryAI Brain — persistent memory for Claude via Obsidian
SkillFind prior Claude Code and Codex sessions with an indexed/recall: session memory engine
MCPA basic implementation of persistent memory using a localKnowledge Graph Memory ServerMore from alirezarezvani
SkillResume a paused experiment/ar:ar-resume — Resume Experiment
SkillShow experiment dashboard with results, active loops/ar:ar-status — Experiment Dashboard
SkillStart an autonomous experiment loop with user-selected/ar:loop — Autonomous Experiment Loop
SkillRun a single experiment iteration/ar:run — Single Experiment Iteration
SkillSet up a new autoresearch experiment interactively/ar:setup — Create New Experiment
Skill/cs:boardroom <brief/cs:boardroom — Multi-Role Boardroom Deliberation