wiki-lint
Audits llm-wiki/wiki for orphans, stale claims, missing links, and contradictions.
Skill by SkinnnyJay
Chalk Labs
Need help deploying wiki-lint?
Chalk Labs sets up agents like this inside real teams: connected to your tools, tested on your data, with a human in the loop where it matters.
Install
Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.
npx agentmarket install skill/skinnnyjay-wiki-lint
Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.
Get the weekly diff
The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.
About
Use cases: ml
Reviews
Write a reviewNo reviews yet. Installed it? Reviews come from verified installers only. How reviews work
Permissions
What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.
- Tools
- No tools found
- Secrets
- None asked for
- Install scripts
- None found
Network hosts it talks to 1
codeload.github.com· found in install command
Files it touches
~/.claude/skills/wiki-lint
Tools it pre-approves 4
From allowed-tools in SKILL.md: the agent may use these without asking you each time.
ReadGrepGlobBash
What's inside
- Skill name
wiki-lint- Description
- Audits llm-wiki/wiki for orphans, stale claims, missing links, and contradictions. Use when user asks to health-check the wiki.
- License
- MIT
- Allowed tools
- Read, Grep, Glob, Bash
- Instructions
- 79 lines in SKILL.md · view on GitHub
Before installing
- Pre-approves any shell command
Files (1)
- SKILL.md
SKILL.md (first lines)
# Wiki lint — Auditor Audit **`wiki/`** for **orphans**, **broken `[[wikilinks]]`**, **contradictions**, and **gaps**. Prefer the deterministic CLI/MCP gates when available; use skill judgment for semantic contradictions the machine cannot see. For **`raw/`** that is still structurally broken **before** wiki merge, use **wiki-raw-prepare** / **`llm-wiki raw validate`** first; **wiki-lint** focuses on **`wiki/`** coherence. ## Pre-flight 1. Read **`wiki/index.md`** and scan **`wiki/**/*.md`** (or scope to paths the user named). 2. If **`outputs/`** exists, include drafts vs **`wiki/`** duplicate or conflicting claims in scope. 3. Run machine lint first (then deepen with this skill): ```bash llm-wiki lint --write-report # after wiki merge / before shipping: llm-wiki compile --no-site # surgical (pages citing one source): llm-wiki compile --raw raw/path.md --no-site ``` MCP: **`wiki_lint`**, **`wiki_compile`** (same gates). Reports: **`outputs/lint-report.json`**, **`outputs/claims.json`**. ## Steps ### Step 1 — Index vs filesystem - List **orphans**: pages not linked from **`wiki/index.md`** where you expect them listed (project convention from **`llm-wiki/CLAUDE.md`**). - List **broken wikilinks**: targets that do not exist. - Compare with **`llm-wiki lint`** / **`outputs/lint-report.json`** (`orphan`, `broken_wikilink`). ### Step 2 — Contradictions and gaps - Note **contradictions** between pages about the same entity (different claims). - Check **`llm-wiki kg conflicts`** and claim IR in **`outputs/claims.json`**. - Suggest **next sources** or **open questions** to resolve gaps. ### Step 3 — Outputs vs wiki - If **`outputs/`** exists, flag **duplicate claims** or **contradictions** between **`outputs/`** and **`wiki/`** (treat outputs as drafts until promoted with evidence). ### Step 4 — Refresh site viewer If auto-fixable edits were applied and **`viewer.enabled`** is not false in config: ```bash llm-wiki build-site --if-stale # or fold into: llm-wiki compile ``` This rebuilds **`wiki/.og/`** only when wiki content is newer than the last build. ## Done looks like - Deliverable is a **structured report**: orphans, broken links, contradictions, gaps — each with **concrete next actions** (create page, remove link, merge, research). - Machine gates (**`lint`** / **`compile`** / optional **`knowledge-test`**) are green or listed with fixes. - User knows whether the wiki is **ship-ready** for their use case or needs follow-up **wiki-research** / **wiki-maintainer** work. ## Artifacts | Reads | Writes | |-------|--------| | `wiki/index.md`, `wiki/**/*.md`, optional `outputs/**` | Prefer **`outputs/lint-report.json`** / **`claims.json`** via CLI; optional `outputs/lint-report.md` | Downstream: **wiki-maintainer** (fixes), **wiki-pipeline** (gate). See **`skills/references/pipeline-artifacts.md`**. Shared config: **`skills/references/mcp-and-kg.md`** (`compile.*`, aliases, ontology). ## Related skills - **wiki-raw-prepare** — structural cleanup in **`raw/`** before merge. - **wiki-maintainer** — apply cross-link and index fixes. - **wiki-query** — Q&A from wiki after lint passes. - **wiki-session-memory** — session memory files under **`raw/memory/`** (optional context). ## Smoke check - **CLI:** `llm-wiki lint --write-report`; `llm-wiki compile --no-site`; `llm-wiki knowledge-test --file examples/knowledge-tests.json` (plugin examples; scaffolded vaults pass). - **Prompt:** Invoke this skill; confirm the report structure (orphans, links, gaps) is produced. Optional: `skills/references/context-persona.md` when running checks via tools; `persona.name` in config (default **Gennie**).
From SkinnnyJay/wiki-llm, MIT.
More from SkinnnyJay/wiki-llm
- 0.529score
wiki-ingest
SkinnnyJay
Merges new raw sources into llm-wiki/wiki/ with index and log updates.
SkillBFreshness: BClaude Code +9 - 0.529score
wiki-maintainer
SkinnnyJay
Maintains the LLM Wiki vault — index, log, cross-links, contradictions.
SkillBFreshness: BClaude Code +9 - 0.521score
wiki-extract
SkinnnyJay
Index and routing for source-specific vault extractors (wiki-extract-*).
SkillBFreshness: BClaude Code +9 - 0.521score
wiki-learn
SkinnnyJay
Cross-session memory for the vault — learn, recall, and review patterns, preferences, pitfalls, and next steps in llm-wiki/.agent-memory.md.
SkillBFreshness: BClaude Code +9 - 0.521score
wiki-pipeline
SkinnnyJay
End-to-end vault workflow — status, research/fetch, raw prepare, wiki ingest, lint, build-site, validate — with optional user gates between stages.
SkillBFreshness: BClaude Code +9 - 0.520score
wiki-onboard
SkinnnyJay
First-run router for a new or existing llm-wiki vault.
SkillBFreshness: BClaude Code +9 - 0.519score
wiki-extract-github
SkinnnyJay
Extract GitHub repo README, issues, PRs, releases, and changelogs into raw/.
SkillBFreshness: BClaude Code +9 - 0.519score
wiki-extract-ebook
SkinnnyJay
Extract text from EPUB, MOBI, PDF, and AZW3 ebook files into raw/.
SkillBFreshness: BClaude Code +9 - 0.518score
wiki-fetch
SkinnnyJay
Orchestrate fetching a URL or research query into the vault using the best available adapter.
SkillBFreshness: BClaude Code +9 - 0.516score
wiki-extract-paywall
SkinnnyJay
Bypass paywalls to get full article text into raw/.
SkillBFreshness: BClaude Code +9 - 0.516score
wiki-extract-podcast
SkinnnyJay
Extract podcast episodes from RSS and transcribe via Whisper or AssemblyAI into raw/.
SkillBFreshness: BClaude Code +9 - 0.516score
wiki-extract-wikipedia
SkinnnyJay
Extract Wikipedia articles and Wikidata infobox data into raw/ via REST API and SPARQL.
SkillBFreshness: BClaude Code +9
Signals
- Stars
- 0
- Forks
- 0
- Open issues
- 0
- Last commit (this folder)
- Jul 29, 2026 (3 months ago)
- Last release
- Jul 28, 2026 (3 months ago)
From GitHub (fetched Oct 24, 2026).
Details
- License
- MIT
- Repository
- github.com/SkinnnyJay/wiki-llm
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- GitHubgithub-graphql
More from SkinnnyJay
- 0.529score
wiki-ingest
SkinnnyJay
Merges new raw sources into llm-wiki/wiki/ with index and log updates.
SkillBFreshness: BClaude Code +9 - 0.529score
wiki-maintainer
SkinnnyJay
Maintains the LLM Wiki vault — index, log, cross-links, contradictions.
SkillBFreshness: BClaude Code +9 - 0.521score
wiki-pipeline
SkinnnyJay
End-to-end vault workflow — status, research/fetch, raw prepare, wiki ingest, lint, build-site, validate — with optional user gates between stages.
SkillBFreshness: BClaude Code +9 - 0.521score
wiki-learn
SkinnnyJay
Cross-session memory for the vault — learn, recall, and review patterns, preferences, pitfalls, and next steps in llm-wiki/.agent-memory.md.
SkillBFreshness: BClaude Code +9 - 0.521score
wiki-extract
SkinnnyJay
Index and routing for source-specific vault extractors (wiki-extract-*).
SkillBFreshness: BClaude Code +9 - 0.520score
wiki-onboard
SkinnnyJay
First-run router for a new or existing llm-wiki vault.
SkillBFreshness: BClaude Code +9
Done for you
Rather not set it up yourself?
Chalk Labs scopes, builds and runs agents for your team: the task audit, the eval harness, the integrations and the weekly ops report. Fixed scope, priced before we start.
See how we build AI agents →Read the case studies
Chalk Labs is a two-founder AI and growth studio. We build AI agents, CRMs and MVPs for Web3 and AI companies, and have shipped 50+ product and token launches.
- launches shipped
- 50+
- launches shipped
- media placements
- 500+
- media placements
- founders on every account
- 2
- founders on every account
Press we've placed: Forbes · CoinDesk · Decrypt · Cointelegraph · The Block
Rahil Jain, Co-founder, Chalk Labs. Ex-Goldman Sachs engineer. You talk to the person who builds it.
Get wiki-lint running in your business
Tell us what you want to automate. A founder replies within one working day with a plan, not a pitch deck.
Cite this page
wiki-lint skill. Agent Mart. https://agentmart.dev/skills/skinnnyjay-wiki-lint (last updated 2026-10-24)