memory-audit SKILL.md skillA

memory-audit from kalabint/llm-tooling is an agent skill: a SKILL.md folder an AI agent loads when a task calls for it.

Install it, pinned

mdr add kalabint/llm-tooling/memory-audit@git:20260907.47ea7d2

Pins git:20260907.47ea7d2 in mdr.lock. Pin the fingerprint instead.

Audit grade: A, 17 of 17 checks passed
Kind: SKILL.md skill
Repository: kalabint/llm-tooling
Path: skills/memory-audit/SKILL.md
Versions recorded: 1
Latest version: git:20260907.47ea7d2, committed Sep 7, 2026
Size: 3,431 bytes
Repository stars: 6
License: AGPL-3.0

Audit the project's persistent agent memories for backfire-risk, stale/superseded state, contradictions, redundancy, and bloat. Lists problems for the user to disposition - does NOT auto-correct. Surfaced on a 14-day cadence by a SessionStart hook. Use when the user asks to audit/review/clean up memories, or when the overdue reminder fires.

Collected from public GitHub. Every version is saved exactly as it was and never changes. Install it at an exact version with the mdr tool, and every file is checked before it reaches your agent. The main file is checked against the SHA-256 fingerprint recorded here, and the others against the fingerprints GitHub keeps for its source commit. The automatic checks below grade the latest version, and the same checks always give the same file the same grade.

What the file says

# Memory audit

Keep the memory store **true and non-backfiring**. Memories are loaded into every session, so a wrong, stale, or over-prescriptive entry silently steers behaviour. This skill runs a periodic review.

Surfaced every 14 days by `~/.claude/hooks/memory-audit-due.sh` (SessionStart), gated on `~/.claude/skills/memory-audit/.last_audit`.

## Run order

1. **Pre-pass (cheap, mechanical).** This also resets the 14-day timer:
   ```bash
   python3 ~/.claude/skills/memory-audit/scan.py
   ```
   It lists the project's memory files with age/size and shortlists stale-marker / bloat / oldest candidates, and prints the MEMORY.md index size vs budget. Derives the memory dir from cwd; pass `--dir <path>` to override.

2. **Dispatch read-only auditors in parallel** (one message, multiple Agent calls, for context hygiene). Batch the files by type (user-profiles + profiling-feedback / feedback-rules / project / reference) so each worker applies a consistent lens. Every worker prompt MUST say: READ-ONLY, do not modify; read full content (not skim); flag only problems; return `FILENAME - [CATEGORY, severity] one-line issue ("quote")`; end with a CLEAN list.
…

Read the whole file at its exact version.

How to install

The exact contents, by fingerprint

mdr add kalabint/llm-tooling/memory-audit@sha256:a3aee2195dd142a7

The line at the top of this page pins the label git:20260907.47ea7d2, which is what most people want. This one pins the sha256 fingerprint of the contents instead, so it survives a label being moved. Either way the exact version is written to mdr.lock, and mdr install downloads and checks those contents again, so it installs them exactly or stops with an error.

Badge

[![mdr](https://markdownregistry.com/badge/art_3rbbdjs7ix4bkr2m.svg)](https://markdownregistry.com/a/art_3rbbdjs7ix4bkr2m)

1 badge image fetches in 30 days. What the badge shows.

Versions

The registry has recorded 1 version of this file, committed on Sep 7, 2026.

version	committed	commit	size	audit
git:20260907.47ea7d2 latest	2026-09-07	47ea7d2	3,431 B	A	view

Audit of the latest version

A 17 of 17 checks passed. Fixed automatic checks with no AI involved, so the same checks give the same answer every run.

pass: Frontmatter block present
pass: Frontmatter declares a name
pass: Frontmatter declares a description
pass: Size between 200 bytes and 200 KB (3431 bytes)
pass: No zero-width or bidi control characters
pass: No instruction hidden inside an HTML comment
pass: No link to an exfiltration or paste host
pass: No credential-shaped string
pass: No instruction to send local credentials anywhere
pass: No text hidden with inline styles
pass: No prompt-injection phrasing
pass: No curl or wget piped into a shell
pass: No recursive delete of root, home or parent
pass: No instruction to read or print local credentials
pass: No base64 blob over 200 characters
pass: No link to a raw IP address
pass: No script tag

Source

GitHub

kalabint/llm-tooling · 6 stars · license AGPL-3.0 · pushed 2026-10-20 · branch main

API

GET https://markdownregistry.com/api/v1/artifacts/art_3rbbdjs7ix4bkr2m
GET https://markdownregistry.com/api/v1/resolve?ref=kalabint/llm-tooling/memory-audit
GET https://markdownregistry.com/api/v1/blob/a3aee2195dd142a716ae7ff44ce0a1795ad3ec32c860f780582b635bcde0d2f8

Your AI assistant does the legwork. You hear about the deals worth your word. Hand yours the standing instructions at modelranch.com and it joins the network that reads files like this one.

Guides: the SKILL.md format and its required fields · what to check before you install a skill · how to pin a skill to an exact version · where to find agent skills · keeping a team on the same skills · which agent file does what.

More from kalabint/llm-tooling

ask-when-unclear SKILL.md skill

kalabint/llm-tooling · skills/ask-when-unclear/SKILL.md · If decoding an ambiguous instruction takes more than a few sentences of reasoning, stop and ask instead of guessing.