agentsclimarketplace

Maintain

Skill Oshayr/LLM-Wiki/skills/maintain

Wiki maintenance — lint broken links, merge near-duplicates, upgrade confidence, flag stale pages via freshness tiers, gap analysis, concept synthesis. Use on: 'wiki maintenance', 'wiki cleanup', 'fix wiki', 'wiki health', 'check wiki', 'consolidate wiki'.From its SKILL.md

Install
npx -y skills add Oshayr/LLM-Wiki --skill maintain

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

SKILL.md

3.5 KB, as published. Nobody here has run it

Wiki Maintain

Comprehensive wiki maintenance: lint, deduplicate, upgrade, and analyze.

Resolve .wiki/ from plugin install scope. If not found, say "No wiki found."

Arguments

  • /wiki-maintain — run full maintenance (all steps below)
  • /wiki-maintain lint — only fix broken links, missing frontmatter, orphans
  • /wiki-maintain dedup — only find and merge near-duplicate pages
  • /wiki-maintain gaps — only analyze knowledge gaps and missing coverage

Full Maintenance Steps

1. Lint

Launch wiki-auditor agent to:

  • Find [[wiki-links]] that point to non-existent pages — list for user
  • Fix missing frontmatter fields (add defaults)
  • Find orphan pages (no incoming links) — suggest connections
  • Remove dead index entries
  • Fix stale updated: dates on pages that were modified but not date-bumped

2. Deduplicate

Find pages with >60% slug token overlap (Jaccard similarity on slug words split by -):

  • Report pairs for review
  • For confirmed duplicates: auto-merge into one page, update all backlinks, add redirect note to absorbed page

3. Confidence Upgrade

Pages with 3+ independent sources in frontmatter get upgraded:

  • low → medium (if 2+ sources)
  • medium → high (if 3+ corroborating sources)
  • Run fact-checker agent on high-confidence pages to verify claims against external sources
  • Write the reason in log.md

4. Stale Detection (Freshness Tiers)

Pages are evaluated against an intelligent freshness system — not a flat threshold. Each page has a TTL based on its content type:

TierTTLExamples
live15 minstock prices, live scores, server status, deployment state
breaking1-6 hoursbreaking news, incident updates, release announcements
current1-3 daysnews articles, current events, trending topics
fast1-4 weeksAI/LLM/MCP, API changes, model benchmarks
moderate1-3 monthssoftware versions, frameworks, libraries, tools
standard6 monthsgeneral knowledge, how-to guides (default)
academic1 yearresearch papers, studies, formal publications
evergreen5 yearshistory, biographies, foundational concepts, laws, theorems
permanentneverpersonal notes, ideas, memories, journal entries

Resolution order:

  1. Explicit freshness_tier: in page frontmatter (user override)
  2. Explicit ttl: in frontmatter (custom duration like ttl: 30m or ttl: 2d)
  3. Auto-classification from tags, type, and content keywords

For stale pages:

  • Add stale: true to frontmatter
  • Suggest running /wiki-write --refresh-stale or /wiki-read to refresh

5. Concept Auto-Generation

Detect patterns spanning 3+ pages:

  • Find groups of pages that share 3+ common [[wiki-links]] targets
  • For each cluster: suggest a synthesis article that connects the concepts
  • If user approves: generate the synthesis page via wiki-writer agent

6. Regenerate Index

Rebuild .wiki/index.md from all pages:

  • Group by type: field (concept, entity, source, idea, status, etc.)
  • Alphabetical within each group
  • Include confidence badge and one-line description

7. Report

Print summary: broken links fixed, duplicates found, confidence upgrades, stale flags, concepts suggested, index updated.

Keep looking

Skills are one crate of 627,801. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.