Coding Agents

This page is the shortest way to understand how md2wechat-skill fits different agent runtimes.

Shared rule across Claude Code, Codex, and OpenCode

For the Coding Agent path, always think in this order:

1. install the CLI
2. install the skill
3. run discovery
4. run the actual job

If you already have Node and npm, use:

npm install -g @geekjourneyx/md2wechat
npx skills add https://github.com/geekjourneyx/md2wechat-skill --skill md2wechat
md2wechat version --json
md2wechat capabilities --json

If you prefer Homebrew on macOS, replace the first line with:

brew install geekjourneyx/tap/md2wechat

If you used install.sh, refresh PATH first:

export PATH="$HOME/.local/bin:$PATH"

Discovery-first workflow

Do not guess the current runtime state. Ask the CLI first:

md2wechat capabilities --json
md2wechat providers list --json
md2wechat themes list --json
md2wechat prompts list --kind image --json

That tells the agent:

• which mode is the default
• which providers are configured
• which themes exist
• which prompt presets are available

What changes by runtime

First tasks to give any Coding Agent

Please install md2wechat CLI first, then install the md2wechat skill, verify version and capabilities, and stop only after every command succeeds.

Run md2wechat capabilities --json first, then convert article.md into WeChat HTML and show me the preview.

Use md2wechat in AI mode and tell me whether the current output is final HTML or an AI request for a second model step.

Important reminders

• npx skills add installs the skill only.
• convert defaults to API mode.
• AI mode is explicit and multi-step.
• Draft creation still needs WeChat credentials.
• OpenClaw is separate from the Coding Agent skill path.

Environment-specific pages

• Claude Code
• Codex
• OpenCode
• Claudian
• OpenClaw