Wiki

SkillSearch

The wiki skill is a persistent markdown project wiki stored under repository omx_wiki with keyword search and lifecycle capture. It lets an agent keep project notes as markdown pages in categories such as architecture, decisions, and debugging, and find them again by keyword or tag. Pages live in the repository, so the knowledge stays with the project.

Available today. Use it from your connected AI after setup.

Have the omx CLI available and a repository where the wiki should live.

Then ask your AI: use the Wiki skill

What your AI can do with it

  • Add, list, read, update, and delete wiki pages through the omx CLI
  • Query pages by keyword, tag, or category with wiki_query
  • Organize pages into categories like architecture, decision, and debugging
  • Link pages together with [[wiki-links]] cross-references
  • Capture session findings as session-log pages at session end
  • Lint the wiki with wiki_lint and refresh it with wiki_refresh

Getting started

  1. Have the omx CLI available and a repository where the wiki should live.
  2. Run omx wiki wiki_add with a title, content, tags, and category to create the first page.
  3. Run omx wiki wiki_query with a query, tags, or category to search existing pages.
  4. Set wiki.autoCapture in .omx-config.json if session findings should be saved automatically.
  5. Run omx wiki wiki_lint to check the wiki for problems.

What this skill tells your AI

The instructions your AI receives, as published by yeachan-heo/oh-my-codex in skills/wiki/SKILL.md and read by ahel’s review.

Persistent, self-maintained markdown knowledge base for project and session knowledge.

Operations

Ingest

omx wiki wiki_ingest --input '{"title":"Auth Architecture","content":"...","tags":["auth","architecture"],"category":"architecture"}' --json

Query

omx wiki wiki_query --input '{"query":"authentication","tags":["auth"],"category":"architecture"}' --json

Lint

omx wiki wiki_lint --json

Quick Add

omx wiki wiki_add --input '{"title":"Page Title","content":"...","tags":["tag1"],"category":"decision"}' --json

List / Read / Delete

omx wiki wiki_list --json
omx wiki wiki_read --input '{"page":"auth-architecture"}' --json
omx wiki wiki_delete --input '{"page":"outdated-page"}' --json
omx wiki wiki_refresh --json

Categories

architecture, decision, pattern, debugging, environment, session-log, reference, convention

Storage

  • Pages: omx_wiki/*.md
  • Index: omx_wiki/index.md
  • Log: omx_wiki/log.md

Cross-References

Use [[page-name]] wiki-link syntax to create cross-references between pages.

Auto-Capture

At session end, discoveries can be captured as session-log-* pages. Configure via wiki.autoCapture in .omx-config.json.

Hard Constraints

  • No vector embeddings — query uses keyword + tag matching only
  • Wiki files are repository project knowledge under omx_wiki/; legacy .omx/wiki/ is read-only compatibility input when no canonical wiki exists

Signals

GitHub stars
33k
Forks
3k
Last commit
Oct 2026

Questions

Where are the wiki pages stored?
Pages are markdown files under omx_wiki/*.md, with an index at omx_wiki/index.md and a log at omx_wiki/log.md. A legacy .omx/wiki/ folder is read-only compatibility input when no canonical wiki exists.
How does search work?
Queries use keyword and tag matching only. There are no vector embeddings, so results come from matching words and tags rather than semantic similarity.
What categories can pages use?
architecture, decision, pattern, debugging, environment, session-log, reference, and convention.
Advanced
Item type
skill
Key
wiki-2
Source
github.com/yeachan-heo/oh-my-codex