vault-audit
Skill0 starsChanged 36 days agoMITView on GitHub
Audits a markdown knowledge base (Obsidian or any notes vault) for drift against its own rules file — stale task/follow-up items, folder-structure drift, and unconsolidated duplicate notes. Use on "check the vault", "audit my notes", "vault snapshot", or when you want a status read on your task/tracking files against what's actually true. Read-only against the vault; writes only one dated audit file.
7 other projects use a version of this.
Adopted by
Most recently taken up by
- xpoes123/claude-skillstheir own version, first seen Oct 20, 2026
- kriscard/Skillstheir own version, first seen Oct 13, 2026
- gnekt/My-Brain-Is-Full-Crewtheir own version, first seen Oct 8, 2026
- Vifert/second-brain-templatetheir own version, first seen Sep 28, 2026
- NovaKepler513/claude-skillstheir own version, first seen Jul 2, 2026
- YiweiCreates/claude-skillstheir own version, first seen Jul 2, 2026
Other versions7 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 7
- gnekt/My-Brain-Is-Full-Crew★ 3.9ksame name
- pass-agent/loomkin★ 179same name
- kriscard/Skills★ 15same name
- Vifert/second-brain-template★ 6same name
- NovaKepler513/claude-skills★ 0same name
- xpoes123/claude-skills★ 0same name
- YiweiCreates/claude-skills★ 0same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository4 other files it gives its agents.
- buildSkill
- outreach-criticSkill
- overnight-explorerSkill
- plan-soloSkill
The file
We split this skill into the rules it gives an agent
2 rules. 0 are common, 2 are distinctive, and 0 are very specific to this project.
0
Common
Most of the 5 edited versions we read keep these.
2
Distinctive
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 14
Locate the vault: If the current working directory contains a governance/rules file that describes the vault's structure, use it as the …
- Distinctive: line 27
After writing the file, report in one line where it landed and give a one-line summary of what it found (or that nothing needed flagging).
“Common” is measured: kept by at least 6 in 10 of the 5 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Vault Audit
Scans a markdown knowledge base against its own governance rules and writes one dated status file. Read-only against the rest of the vault — the only thing it writes is the audit file itself.
This skill assumes a vault that keeps (a) a governance/rules file at its root that describes the intended structure, and (b) one or more running lists of open items (tasks, follow-ups). Adapt the specific filenames below to your own setup; the method is what generalises.
Instructions
- Locate the vault. If the current working directory contains a governance/rules file that describes the vault's structure, use it as the vault root. Otherwise ask for the vault path before doing anything else — don't guess.
- Read the governance files: the rules file plus any open-item lists (e.g. a tasks file and a follow-ups file) at vault root.
- Check structural drift: list the actual top-level folders and compare against the intended set in the rules file. Flag anything new/unlisted. Flag any note sitting directly at vault root that the rules say shouldn't be there.
- Check the task list for staleness: for each open task, skim its linked note/area folder for evidence it's actually done, superseded, or abandoned (e.g. dated context that's clearly long past with no update since). Don't guess — only flag when the linked note gives real evidence either way. If it's genuinely unclear, say so rather than picking a side.
- Check the follow-up list for staleness the same way — flag anything that looks answered/resolved in its linked note but is still listed as open.
- Spot-check one or two folders for notes that duplicate content already covered elsewhere, or raw material sitting around that should have been consolidated by now, if the rules call for one synthesised note per topic.
- Write the output as
Vault Audit — YYYY-MM-DD.mdinside a dedicated audits folder (create it on first run if it doesn't exist) — not at vault root, so dated files don't pile up there. Headings, in this order:- Structure — folder-list drift, stray root files
- Tasks check — tasks that look done/stale, with the evidence and a pointer to where you found it
- Follow-ups check — same, for follow-ups
- Hygiene notes — duplication/consolidation issues spotted
- Any section with nothing to report should say so plainly (e.g. "no drift found") — don't pad it with filler.
- Rules: only read, never invent. If a file referenced in the rules doesn't exist, or content is ambiguous, say "not there" / "unclear — worth checking" rather than guessing. Don't edit the task or follow-up lists yourself — the audit reports, it doesn't act; checking boxes or archiving items is a separate, explicit follow-up step.
- After writing the file, report in one line where it landed and give a one-line summary of what it found (or that nothing needed flagging).
Running it again
Invoke any time — each run creates a new dated file; old runs stay in place as a history of vault health over time. Don't overwrite or delete them.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

