Claude Skill

second-brain-lint

Audit a second-brain vault for structural problems and repair them: broken wikilinks, orphan pages, empty stubs, frontmatter that violates the schema, near-duplicate pages, and stale index entries. Use this skill whenever the user asks to lint, clean, audit or check the health of

LLM Mart · 0 points · 0 views 0 listing impressions 0 install-command copies
Virus-scanned Reviewed automatically before listing.

Full trust report

Download undefined-ui-second-brain-os-skills_second-brain-lint-347feee.zip · 1 KB
Part of undefined-ui/second-brain-os — 18 skills

Install

skills CLI npx skills add https://github.com/undefined-ui/second-brain-os/tree/main/skills/second-brain-lint
Claude Code claude plugin marketplace add https://llmmart.ai/marketplace.json && claude plugin install undefined-ui-second-brain-os@llmmart
Git git clone https://github.com/undefined-ui/second-brain-os.git

The skills CLI installs just this skill, for any of its supported agents. Claude Code installs the whole undefined-ui/second-brain-os collection as a plugin from our marketplace. Git is the plain clone.

Skill manifest

Lint the vault

Structural rot is silent. Nothing errors, nothing crashes, the vault just slowly stops answering questions well because a third of its pages are unreachable. Linting is the only thing that catches it before the owner does.

Core rule

Report before repairing. Fix the mechanical problems automatically, but never delete or merge a page without listing it first and getting a yes, because the owner may have written it by hand.

Workflow

  1. Inventory. Count pages by type. Build the link graph by parsing every [[wikilink]].
  2. Broken links. Links pointing at pages that do not exist. Distinguish two cases: a typo or rename, which you fix, and a genuine gap, which goes into index.md under Gaps.
  3. Orphans. Pages with no inbound links. For each, either find where it should be linked from and add the link, or flag it as a candidate for deletion.
  4. Stubs. Pages under roughly 40 words with no links. Usually a failed ingest. Flag for re-ingest from the original source in raw/.
  5. Schema. Missing or malformed frontmatter, wrong type, missing dates, tags outside the existing vocabulary.
  6. Near-duplicates. Pages with very similar titles or aliases. Propose merges; do not perform them unsolicited.
  7. Index drift. Pages missing from index.md, index entries pointing nowhere.
  8. Repair, then append the run to log.md.

Output format

Vault: <n> pages (<n> sources, <n> concepts, <n> entities, <n> synthesis)
Link graph: <n> links, average <n> per page, <n> orphans

Fixed automatically:
- <list>

Needs a decision:
- <list, each with the recommended action>

Health: orphan rate <n>%, broken link rate <n>%, stub rate <n>%

Calibration

An aggressive linter that merges pages on its own judgement destroys work. A timid one that only reports produces a list nobody acts on. The split above is the useful line: mechanical fixes happen, semantic decisions get proposed.

Do not run a full lint after every ingest. Weekly, or after an import of more than about twenty sources.

Files (second-brain-os)
  • SKILL.md 2.6 KB
    ---
    name: second-brain-lint
    description: >-
      Audit a second-brain vault for structural problems and repair them: broken
      wikilinks, orphan pages, empty stubs, frontmatter that violates the schema,
      near-duplicate pages, and stale index entries. Use this skill whenever the
      user asks to lint, clean, audit or check the health of their vault, says
      pages feel messy or links are broken, after a large bulk import, or when a
      scheduled maintenance run fires. Do NOT use for ingesting new sources, for
      answering questions from the vault, or for rewriting page content that is
      structurally fine but reads badly.
    ---
    
    # Lint the vault
    
    Structural rot is silent. Nothing errors, nothing crashes, the vault just
    slowly stops answering questions well because a third of its pages are
    unreachable. Linting is the only thing that catches it before the owner does.
    
    ## Core rule
    
    Report before repairing. Fix the mechanical problems automatically, but never
    delete or merge a page without listing it first and getting a yes, because the
    owner may have written it by hand.
    
    ## Workflow
    
    1. **Inventory.** Count pages by type. Build the link graph by parsing every
       `[[wikilink]]`.
    2. **Broken links.** Links pointing at pages that do not exist. Distinguish
       two cases: a typo or rename, which you fix, and a genuine gap, which goes
       into `index.md` under Gaps.
    3. **Orphans.** Pages with no inbound links. For each, either find where it
       should be linked from and add the link, or flag it as a candidate for
       deletion.
    4. **Stubs.** Pages under roughly 40 words with no links. Usually a failed
       ingest. Flag for re-ingest from the original source in `raw/`.
    5. **Schema.** Missing or malformed frontmatter, wrong `type`, missing dates,
       tags outside the existing vocabulary.
    6. **Near-duplicates.** Pages with very similar titles or aliases. Propose
       merges; do not perform them unsolicited.
    7. **Index drift.** Pages missing from `index.md`, index entries pointing
       nowhere.
    8. **Repair**, then append the run to `log.md`.
    
    ## Output format
    
    ```
    Vault: <n> pages (<n> sources, <n> concepts, <n> entities, <n> synthesis)
    Link graph: <n> links, average <n> per page, <n> orphans
    
    Fixed automatically:
    - <list>
    
    Needs a decision:
    - <list, each with the recommended action>
    
    Health: orphan rate <n>%, broken link rate <n>%, stub rate <n>%
    ```
    
    ## Calibration
    
    An aggressive linter that merges pages on its own judgement destroys work. A
    timid one that only reports produces a list nobody acts on. The split above is
    the useful line: mechanical fixes happen, semantic decisions get proposed.
    
    Do not run a full lint after every ingest. Weekly, or after an import of more
    than about twenty sources.
    

Comments (0)

Sign in to join the conversation.

No comments yet.

Reviews (0)

No reviews yet.

Related