Agent API DocsCodex
Install md2wechat for Codex tasks, verify discovery, and use direct prompts for conversion and draft workflows.
Codex
Installation and setup paths are distribution evidence, not proof of end-to-end compatibility with any named host. Treat each runtime as a validation candidate until its install, discovery, conversion, preview, and—when applicable—draft flow passes an end-to-end smoke test.
Recommended path
For the Codex validation candidate, begin with the same installation sequence documented for Claude Code:
- install the CLI
- install the skill
- verify discovery
- ask the agent to run the publishing task
Step 1. Install the CLI
If you already have Node and npm, start here:
npm install -g @geekjourneyx/md2wechatIf you prefer Homebrew on macOS:
brew install geekjourneyx/tap/md2wechatOptional when Go is already available:
go install github.com/geekjourneyx/md2wechat-skill/cmd/md2wechat@v2.4.0Fallback:
curl -fsSL https://github.com/geekjourneyx/md2wechat-skill/releases/download/v2.4.0/install.sh | bash
export PATH="$HOME/.local/bin:$PATH"Step 2. Install the skill
npx skills add https://github.com/geekjourneyx/md2wechat-skill --skill md2wechatStep 3. Verify discovery
md2wechat version --json
md2wechat config init
md2wechat capabilities --json
md2wechat themes list --json
md2wechat prompts list --kind image --jsonPrompts you can send to Codex
Install md2wechat CLI and the md2wechat skill, verify version and capabilities, then stop only after every command succeeds.Use md2wechat to preview article.md as WeChat HTML. Run capabilities first and tell me which mode you are using.Use md2wechat in AI mode with theme autumn-warm, then explain whether the result is final HTML or an AI request for a second model step.If you prefer the raw API path
This site also documents the HTTP APIs directly:
Use the raw API path only when the skill or CLI is not the right integration layer for your runtime.