wiki — Claude Code skill: what it does, how to install
Keeps a markdown knowledge base for the project in .omc/wiki/: pages, an index and an operation log that survive between sessions.
npx skills add Yeachan-Heo/oh-my-claudecode --skill wiki -a claude-codeWhat wiki does
The skill turns .omc/wiki/ into a project knowledge base. Pages are markdown with YAML frontmatter, index.md works as a catalog, and log.md records what was done. Sessions end, the notes stay.
A handful of operations covers the loop: wiki_ingest files material across pages, wiki_add writes a single page, wiki_query searches by keyword and tag, wiki_lint reports orphan pages, stale content, broken cross-references and oversized pages. Reading and deleting work the same way. Categories are fixed: architecture, decision, pattern, debugging, environment, session-log. Links between pages use [[page-name]].
There are no vector embeddings here, so wiki_query matches words and tags rather than meaning — a real limit once the wiki grows. By default .omc/wiki/ stays out of git. wiki.autoCapture in .omc-config.json decides whether findings get captured at the end of a session.
What it helps with
- record an architecture decision so the agent stops asking about it
- check what is already known about authentication without rereading the repo
- clear orphan pages and broken links before handing a project over
- keep a bug breakdown in
session-loginstead of debugging it twice - add a short note with one call rather than a full ingest
Example prompts for the agent
- wiki add: we went with JWT plus refresh tokens, category `decision`, tags `auth`
- wiki query: what do we already have on auth and sessions?
- wiki lint — check the wiki for broken links and stale pages
How to install wiki
1Skills CLI (Vercel)
npx skills add Yeachan-Heo/oh-my-claudecode --skill wiki -a claude-code2Manually: put the files in place
~/.claude/skills/wiki/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/wiki/
1Skills CLI (Vercel)
npx skills add Yeachan-Heo/oh-my-claudecode --skill wiki -a codex2Manually: put the files in place
.agents/skills/wiki/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add Yeachan-Heo/oh-my-claudecode --skill wiki -a cursor2Manually: put the files in place
.cursor/skills/wiki/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install Yeachan-Heo/oh-my-claudecode wiki2Manually: put the files in place
.github/skills/wiki/SKILL.md1Skills CLI (Vercel)
npx skills add Yeachan-Heo/oh-my-claudecode --skill wiki -a gemini-cli2Platform CLI
gemini skills install https://github.com/Yeachan-Heo/oh-my-claudecode --path skills/wikiManually: put the files in place
.devin/skills/wiki/SKILL.mdThe legacy .windsurf/skills/ path still works
Skills CLI (Vercel)
npx skills add Yeachan-Heo/oh-my-claudecode --skill wiki -a opencodeClaude.ai
Zip the folder and upload it in Settings → Capabilities → Skills
From the same set26
The whole set →Similar
Questions about wiki
How do I install wiki in Claude Code?
Run npx skills add Yeachan-Heo/oh-my-claudecode --skill wiki -a claude-code. The skill lands at ~/.claude/skills/wiki/SKILL.md.
What does wiki do?
It stores project notes as markdown pages under .omc/wiki/ and gives the agent operations to write, search and check them.
Where are the pages stored, and do they end up in git?
Pages sit in .omc/wiki/*.md, with index.md as the catalog and log.md as the operation history. The folder is git-ignored by default and stays local to the project.
Source text
SKILL.md · MIT
---
name: wiki
description: LLM Wiki — persistent markdown knowledge base that compounds across sessions (Karpathy model)
triggers: ["wiki", "wiki this", "wiki add", "wiki lint", "wiki query"]
---
# Wiki
Persistent, self-maintained markdown knowledge base for project and session knowledge. Inspired by Karpathy's LLM Wiki concept.
## Operations
### Ingest
Process knowledge into wiki pages. A single ingest can touch multiple pages.
```
wiki_ingest({ title: "Auth Architecture", content: "...", tags: ["auth", "architecture"], category: "architecture" })
```
### Query
Search across all wiki pages by keywords and tags. Returns matching pages with snippets — YOU (the LLM) synthesize answers with citations from the results.
```
wiki_query({ query: "authentication", tags: ["auth"], category: "architecture" })
```
### Lint
Run health checks on the wiki. Detects orphan pages, stale content, broken cross-references, oversized pages, and structural contradictions.
```
wiki_lint()
```
### Quick Add
Add a single page quickly (simpler than ingest).
```
wiki_add({ title: "Page Title", content: "...", tags: ["tag1"], category: "decision" })
```
### List / Read / Delete
```
wiki_list() # Show all pages (reads index.md)
wiki_read({ page: "auth-architecture" }) # Read specific page
wiki_delete({ page: "outdated-page" }) # Delete a page
```
### Log
View wiki operation history by reading `.omc/wiki/log.md`.
## Categories
Pages are organized by category: `architecture`, `decision`, `pattern`, `debugging`, `environment`, `session-log`
## Storage
- Pages: `.omc/wiki/*.md` (markdown with YAML frontmatter)
- Index: `.omc/wiki/index.md` (auto-maintained catalog)
- Log: `.omc/wiki/log.md` (append-only operation chronicle)
## Cross-References
Use `[[page-name]]` wiki-link syntax to create cross-references between pages.
## Auto-Capture
At session end, significant discoveries are automatically captured as session-log pages. Configure via `wiki.autoCapture` in `.omc-config.json` (default: enabled).
## Hard Constraints
- NO vector embeddings — query uses keyword + tag matching only
- Wiki pages are git-ignored by default (`.omc/wiki/` is project-local)
## Model Routing
- `haiku` — quick lookups, lightweight inspection, narrow docs work
- `sonnet` — standard implementation, debugging, and review
- `opus` — architecture, deep analysis, consensus planning, and high-risk review
- `fable` — Claude Fable 5 (above Opus); pass it explicitly on the Task call or pin it per agent with `agents.<name>.model`
- The session model chosen with `/model` applies to the main loop only. Delegated agents run on their pinned tier unless the Task call passes `model` explicitly or the agent is overridden via `agents.<name>.model`. To run delegated work on Fable, use one of those two surfaces; selecting Fable in `/model` alone does not change delegation.
© Yeachan-Heo · MIT · Yeachan-Heo/oh-my-claudecode