agentleFS
jakubtobiaszjakubtobiasz/second-brain-template/.agents/skills/repair/SKILL.md

repair

Skill0 starsChanged 9 days agoMITView on GitHub

Apply owner-approved fixes from a second brain lint report, one item at a time, with logging. Use when the owner says "fix 1-3", "repair the lint findings", "merge these duplicate pages", "fix the broken links", or approves items from a report in 50-outputs/lint/.

8 other projects use a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

6 rules. 0 are common, 1 is distinctive, and 5 are very specific to this project.

  • 0

    Common

    Most of the 6 edited versions we read keep these.

  • 1

    Distinctive

    This author's own choices. Worth a look.

  • 5

    Very specific

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

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

    Only fix what the owner approved.

    line 8
  2. Distinctive:

    Read every page involved before editing.

    line 13
  3. Very specific:

    Run python3 scripts/lint.py and confirm the fixed items are gone and nothing new broke.

    line 16
  4. Very specific:

    Run scripts/qmd-sync.sh (renames and deprecations must reach the search index).

    line 18
  5. Very specific:

    Use git mv old new, then update every inbound link: grep -rl "\[\[old-name" --include=*.md .

    line 36
  6. Very specific:

    Never delete pages: deprecate.

    line 41

“Common” is measured: kept by at least 6 in 10 of the 6 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.

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

Repair: apply approved lint fixes

Only fix what the owner approved. Use the latest report in 50-outputs/lint/ unless told otherwise.

Steps

  1. Open the report; list the approved items back in one line each ("Fixing: 1) broken link …, 2) …").
  2. For each item, in order:
    • Read every page involved before editing.
    • Apply the smallest fix that resolves it (patterns below).
    • Note what you did next to the item in the report: ✔ fixed: <how> (or ✘ skipped: <why>).
  3. Run python3 scripts/lint.py and confirm the fixed items are gone and nothing new broke.
  4. Append to log.md: ## [YYYY-MM-DD] repair | items 1-3 from YYYY-MM-DD-lint with created/updated pages.
  5. Run scripts/qmd-sync.sh (renames and deprecations must reach the search index).
  6. Suggest a commit: git commit -am "repair: <summary>".

Fix patterns

FindingFix
broken-linkFix the typo, or create a seed page for the target (frontmatter, one-line summary, link back)
not in indexAdd the index.md line
frontmatterAdd the missing fields; never invent sources. Downgrade to seed if there are none
orphanLink it from 1-2 genuinely related pages (with a reason), or propose deprecation
one-way-linkAdd the back-reference with a short "why" in ## Related
duplicateMerge: pick the canonical page (more links/content), move unique content + sources into it, add the other's title to aliases, set the other to status: deprecated + superseded_by: "[[canonical]]", repoint inbound links
stale-stateRe-check sources; update with a new as-of date, or mark "(unverified since YYYY-MM-DD)"
stale-seedGrow from existing sources, merge, or deprecate (ask which)
contradictionAdd a ## Contradictions section; don't pick a winner unless a newer/stronger source settles it
raw-modifiedDon't touch. Tell the owner: git checkout -- <file> restores it

Renames and moves

Use git mv old new, then update every inbound link: grep -rl "\[\[old-name" --include=*.md . → edit each. Rename = check in with the owner first (AGENTS.md §6 rule 7).

Hard constraints

  • One approved item at a time; anything beyond the approved list goes back to the owner.
  • Never delete pages: deprecate. Never edit 10-raw/ or 40-journal/.
  • More than 5 files affected by one item → show the plan and wait.

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.