# Rlm Audit

> Audit RLM cache coverage - compare manifest against filesystem

- Skill: `richfrem/rlm-audit` (Agent Skill, multi-file: 5 files)
- Install (CLI): `npx skillmds@latest add richfrem/rlm-audit`
- Raw SKILL.md: https://api.skillmd.com/api/skills/richfrem/rlm-audit/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: AI & ML
- Author: richfrem (https://skillmd.com/u/richfrem)
- Updated: 2026-09-10
- Page: https://skillmd.com/skills/richfrem/rlm-audit

---


# rlm-audit

## Purpose
Systematically audit the RLM (Recursive Language Model) semantic cache to identify gaps between the project manifest and the actual summary files stored on disk. 

## Capabilities
- **Coverage Analysis**: Calculates the percentage of project documentation currently summarized.
- **Gap Identification**: Detects files that are included in the manifest but missing from the cache.
- **Exporting**: Generates CSV lists of missing files for batch processing.
- **Validation**: Ensures the cache directory structure mirrors the source repository.

## Usage
Run the audit for a specific profile to see what's missing:

```bash
python scripts/audit_cache.py --profile wiki --report audit_report.txt --csv missing_files.csv
```

## Related
- **rlm-distill-agent**: Use to fill the gaps identified by this audit.
- **rlm-cleanup-agent**: Use to remove stale entries from the cache.