Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Run a health check on the LLM Wiki vault — mechanical checks (orphans, broken links, stale pages, missing frontmatter, log gap, duplicates) plus semantic checks (contradictions, cross-reference gaps, concepts missing their own page). Outputs a markdown report with suggested actions. Usage /wiki-lint [--stale-days N] [--log-gap-days N]
.claude/skills/alirezarezvani-wiki-lint/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-13 | ✗→✓ | ▲ Improved | 16% | 0% |
| case-14 | ✗→✓ | ▲ Improved | 3% | 0% |
| case-07 | ✗→✓ | ▲ Improved | -21% | 0% |
| case-06 | ✗→✓ | ▲ Improved | 62% | 0% |
| case-10 | ✗→✓ | ▲ Improved | -9% | 0% |
<!-- canonical copy: engineering/llm-wiki/commands/wiki-lint.md — keep in sync (root copy uses repo-root-relative script paths) -->
Health-check the wiki. Surfaces orphan pages, broken wikilinks, stale claims, missing frontmatter, contradictions, and structural drift. Reports, doesn't silently fix — you decide what to change.
Run this weekly, after batch ingests, and always before sharing the wiki.
/wiki-lint
/wiki-lint --stale-days 60
/wiki-lint --log-gap-days 7engineering/llm-wiki/skills/llm-wiki/scripts/lint_wiki.py — orphans, broken links, stale pages, missing frontmatter, duplicate titles, log gapengineering/llm-wiki/skills/llm-wiki/scripts/graph_analyzer.py — hubs, sinks, connected components, graph statsA markdown report grouped by severity:
markdown# Wiki lint — <date> **Total pages:** N **Components:** N **Last log:** <date> ## Found - ⚠️ <N> contradictions (list) - <N> orphans - <N> broken links - <N> stale pages - ... ## Suggested actions 1. Investigate contradiction between [[sources/a]] and [[sources/b]] 2. Create concept page for "<name>" 3. Fix broken link in [[concepts/x]] 4. Re-ingest [[sources/c]] — stale + contradicted 5. ...
Then appends a lint entry to log.md.
Dispatches the wiki-linter sub-agent. See agents/wiki-linter.md.
engineering/llm-wiki/skills/llm-wiki/scripts/lint_wiki.pyengineering/llm-wiki/skills/llm-wiki/scripts/graph_analyzer.pyengineering/llm-wiki/skills/llm-wiki/scripts/append_log.py| Trigger | Pass | |---|---| | Weekly | Mechanical only — fast | | After batch ingest | Full (mechanical + semantic) | | Monthly | Full + structural review | | Before sharing | Full + extra review |
→ engineering/llm-wiki/skills/llm-wiki/SKILL.md → engineering/llm-wiki/skills/llm-wiki/references/lint-workflow.md
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-13 | fail→pass | 10,694 | 7,613 | -29% | 1 | 1 | 0% | 1,742 | 2,013 | +16% | 0 | 0 | — |
case-14 | fail→pass | 11,903 | 7,710 | -35% | 1 | 1 | 0% | 1,956 | 2,016 | +3% | 0 | 0 | — |
case-07 | fail→pass | 12,185 | 5,595 | -54% | 1 | 1 | 0% | 2,170 | 1,717 | -21% | 0 | 0 | — |
case-01 | fail→fail | 15,415 | 16,368 | +6% | 1 | 1 | 0% | 2,511 | 3,682 | +47% | 0 | 0 | — |
case-12 | pass→pass | 9,723 | 4,580 | -53% | 1 | 1 | 0% | 1,590 | 1,595 | +0% | 0 | 0 | — |
case-02 | fail→fail | 18,529 | 7,253 | -61% | 1 | 1 | 0% | 3,396 | 1,025 | -70% | 0 | 0 | — |
case-03 | fail→fail | 12,268 | 8,873 | -28% | 1 | 1 | 0% | 2,317 | 2,356 | +2% | 0 | 0 | — |
case-04 | pass→fail | 7,466 | 3,369 | -55% | 1 | 1 | 0% | 1,147 | 1,293 | +13% | 0 | 0 | — |
case-05 | pass→pass | 10,064 | 2,551 | -75% | 1 | 1 | 0% | 1,684 | 1,069 | -37% | 0 | 0 | — |
case-06 | fail→pass | 7,075 | 9,251 | +31% | 1 | 1 | 0% | 1,413 | 2,283 | +62% | 0 | 0 | — |
case-08 | pass→pass | 9,931 | 1,798 | -82% | 1 | 1 | 0% | 1,403 | 954 | -32% | 0 | 0 | — |
case-09 | fail→fail | 12,509 | 9,761 | -22% | 1 | 1 | 0% | 2,208 | 2,284 | +3% | 0 | 0 | — |
case-10 | fail→pass | 6,907 | 1,830 | -74% | 1 | 1 | 0% | 1,142 | 1,043 | -9% | 0 | 0 | — |
case-11 | fail→pass | 10,555 | 1,805 | -83% | 1 | 1 | 0% | 1,761 | 981 | -44% | 0 | 0 | — |
case-15 | fail→pass | 17,483 | 2,309 | -87% | 1 | 1 | 0% | 3,244 | 1,099 | -66% | 0 | 0 | — |
case-16 | fail→pass | 8,558 | 2,532 | -70% | 1 | 1 | 0% | 1,353 | 1,037 | -23% | 0 | 0 | — |
case-17 | pass→pass | 7,637 | 5,136 | -33% | 1 | 1 | 0% | 1,250 | 1,517 | +21% | 0 | 0 | — |
case-18 | fail→fail | 13,412 | 11,429 | -15% | 1 | 1 | 0% | 2,352 | 2,848 | +21% | 0 | 0 | — |
case-19 | fail→pass | 5,832 | 1,345 | -77% | 1 | 1 | 0% | 1,028 | 866 | -16% | 0 | 0 | — |
case-20 | pass→pass | 18,156 | 11,620 | -36% | 1 | 1 | 0% | 3,226 | 2,962 | -8% | 0 | 0 | — |
case-21 | pass→pass | 6,780 | 8,483 | +25% | 1 | 1 | 0% | 1,330 | 2,439 | +83% | 0 | 0 | — |
case-22 | pass→pass | 4,887 | 3,909 | -20% | 1 | 1 | 0% | 834 | 1,342 | +61% | 0 | 0 | — |
DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 22 cases were attempted, and 21 counted toward the lift figure. The other 1 produced results that are not comparable between the two arms, so they are excluded from the headline rather than averaged into it. The headline lift of +36 percentage points is the difference between those two pass rates over the 21 comparable cases. 2 cases got worse with the skill loaded, and they are included in that figure.
Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.
Other measured skills in the registry, with their headline benchmark lift.