review

Audit Claude Code auto-memory and generate a structured optimization report.

Updated Jun 2, 2026
One-click install
npx skills add https://github.com/ano4l/SiteRent --skill review-ano4l
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: review
Source: https://github.com/ano4l/SiteRent/tree/main/skills/self-improving-agent/skills/review
Command: npx skills add https://github.com/ano4l/SiteRent --skill review-ano4l

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Helps teams audit Claude Code's auto-memory to identify promotion candidates, stale entries, consolidation opportunities, and overall health.

Core Features & Use Cases

  • Locate memory directory and read MEMORY.md to assess startup size, line count, and structure.
  • Analyze entries for recurrence, staleness, and opportunities for consolidation; identify promotion candidates.
  • Cross-reference MEMORY.md with topic files and CLAUDE.md to ensure alignment and detect conflicts.
  • Generate a structured optimization report with actionable recommendations.

Quick Start

Run /si:review to perform a full auto-memory audit and generate recommendations.

Frequently Asked Questions about review

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I audit Claude Code's auto-memory for stale entries and consolidation opportunities?

Auditing auto-memory involves reading MEMORY.md and topic files to identify stale entries, recurring patterns, and consolidation opportunities. The review skill cross-references these with CLAUDE.md to detect conflicts and generate a structured optimization report for memory hygiene.

What is the best way to identify promotion candidates in Claude Code memory files?

Identifying promotion candidates in memory files requires analyzing entries for recurrence and relevance against existing rules. The review skill validates MEMORY.md contents to flag high-value entries suitable for promotion, producing actionable recommendations for memory retention.

How does memory health analysis work across multiple project directories?

Memory health analysis works by locating the memory directory and reading MEMORY.md to assess startup size and line count. It then cross-references topic files and CLAUDE.md across projects to ensure alignment, detect conflicts, and evaluate overall memory structure.

Can I use auto-memory optimization for large MEMORY.md files with outdated references?

Yes, auto-memory optimization handles large MEMORY.md files by analyzing entry staleness and outdated references. The review skill flags outdated content for removal or consolidation, ensuring the memory structure remains actionable and aligned with current project rules.

What are the limitations of automated memory consolidation in Claude Code?

Automated memory consolidation relies on validating entries against existing CLAUDE.md rules and topic files. Limitations include potential conflicts during cross-referencing and the need for manual review of the generated promotion and consolidation report before applying structural changes.