agentleFS
NishikantaRayNishikantaRay/Attic/skills/attic-review/SKILL.md

attic-review

Skill5 starsChanged 2 months agoMITView on GitHub

Check which stashed items may no longer match the code, and resolve them. Use when the user says "review the attic", "what's stale", "is the attic still accurate", "check the attic against the code", "verify this finding", or after a big refactor, rebase or merge that may have invalidated earlier findings. This is about whether stashed knowledge is still TRUE; use attic-doctor for structural damage and attic-prune for items that are merely old.

The file

We split this skill into the rules it gives an agent

3 rules. About 0 are common, 0 are distinctive, and 3 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈0

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 3

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Read the item (/attic-recall <slug>), then check its claim against the code as it is now.

    line 48
  2. Very specific:

    Never run verify without having read the code it refers to.

    line 64
  3. Very specific:

    If a finding turns out to be wrong, the useful move is usually to stash the correction (which appends a dated update to the same slug) and …

    line 68

≈ “Common” is an estimate: at least 1 in 5 of the 372,429 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Review the attic for stale knowledge

Argument: $ARGUMENTS (a slug to verify directly, or empty to list).

Listing what needs a look

node "${CLAUDE_SKILL_DIR}/../attic/scripts/attic.js" review

Each item comes back with a status and the reason for it:

StatusMeans
possibly-staleA file the item cites has changed since it was recorded.
needs-reviewA cited file is gone, or the item is a workaround, or someone flagged it.
unknownNo provenance to check it against.

Items stashed before 1.6 have no provenance and are hidden by default, since listing an entire existing attic as suspect is noise. --all includes them.

What these statuses do NOT mean

possibly-stale means a file moved under this finding, not that the finding is wrong. A rename, a reformat or an unrelated edit in the same file all produce it. Say "may need checking", never "this is out of date" — you do not know that yet, and the whole point of the status is to prompt a look rather than to replace one.

Resolving an item

Read the item (/attic-recall <slug>), then check its claim against the code as it is now. Only after actually looking:

# Checked, still true. Re-stamps confidence and moves the revision baseline.
node "${CLAUDE_SKILL_DIR}/../attic/scripts/attic.js" verify <slug> \
  --note "what you checked"

# Checked, no longer true. Flags it; the text is preserved.
node "${CLAUDE_SKILL_DIR}/../attic/scripts/attic.js" verify <slug> --stale \
  --note "what changed"

# Obsolete. Leaves the index, stays recallable.
node "${CLAUDE_SKILL_DIR}/../attic/scripts/attic.js" archive <slug>

Never run verify without having read the code it refers to. It records that someone checked; running it to clear a warning makes that record a lie, and a verified stamp nobody earned is worse than no stamp at all.

If a finding turns out to be wrong, the useful move is usually to stash the correction (which appends a dated update to the same slug) and then verify, rather than to archive and lose the history of what was believed and why.

Reply

One line per item: the handle, what you checked, and the outcome. If nothing needed review, say so in one line.

Download now (ZIP)

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
What did you use it for? e.g. Reviewing pull requests in a Rails app

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.