wiki
Persistent markdown knowledge base that compounds across sessions — add, query, lint, and ingest project knowledge stored under .omx/wiki/.
$wiki maintains a self-organising markdown knowledge base under .omx/wiki/. Each page is a plain markdown file with tags and a category. The wiki compounds over time: discoveries made in one session become searchable context in future sessions, reducing the cost of re-exploring the same parts of a codebase.
When to use
- You want to record an architectural decision, debugging finding, or recurring pattern for future sessions
- You want to query what the project wiki already knows about a topic before starting an investigation
- You want to ingest external content — docs, specs, research — into the project knowledge base
- You want to lint the wiki to find broken links, missing tags, or stale pages
- You say "wiki add", "wiki query", "wiki lint", "add to wiki", or "what does the wiki say about"
How to invoke
Natural language triggers: "wiki add", "wiki query", "wiki lint", "add to wiki", "search wiki".
Explicit slash: $wiki
codex
> wiki add "Auth uses JWT with 15-minute access tokens and 7-day refresh tokens stored in httpOnly cookies"codex
> wiki query "authentication flow"codex
> $wiki lintWhat happens
Wiki exposes five core operations. wiki_add creates a new page with a title, content, tags, and a category chosen from the standard set: architecture, decision, pattern, debugging, environment, session-log, reference, or convention. wiki_query performs keyword and tag matching across all pages and returns ranked results — there are no vector embeddings, so queries work best with specific terms. wiki_ingest accepts bulk content such as external docs or spec files and splits it into categorised pages. wiki_lint scans the wiki for broken [[page-name]] cross-references, missing required tags, and pages that have not been updated recently. wiki_read, wiki_list, and wiki_delete provide direct page management. All pages live under .omx/wiki/ as plain markdown files with a shared index at .omx/wiki/index.md, keeping the knowledge base fully portable and inspectable without tooling.
Outputs
.omx/wiki/{slug}.md— individual knowledge pages.omx/wiki/index.md— auto-maintained index of all pages with tags and categories.omx/wiki/log.md— append-only log of add, update, and delete operations- Lint report listing broken cross-references, missing tags, and stale pages
Related skills
$deepsearch— exhaustive codebase search whose findings are worth capturing in the wiki$trace— session execution trace that can be ingested as a session-log wiki page