Integration
CLI, Skill, or Raw API: How to Choose the Right md2wechat Entry Point
CLI, skill, and raw API are not three separate products. They are three different entry points. This article explains which one fits which stage.
Integration
Why md2wechat Works Better as a Discovery-First Agent Workflow
md2wechat becomes much steadier when agents inspect the current capability surface first, then choose mode, modules, and resources before execution.
Integration
Why Agents Should Check the Layout Catalog Before Writing
Many advanced layout failures start before execution. They begin when agents write from memory instead of checking the current module catalog first.
Integration
md2wechat 2.0.6: You Can Now Install It with npm
A practical release note for md2wechat 2.0.6, focused on npm installation, shared GitHub Release binaries, checksum verification, and why this matters for Claude Code, Codex, OpenCode, and other agent workflows.
Integration
md2wechat 2.0.7: Native Volcengine and Doubao Image Support Is Here
A practical release note for md2wechat 2.0.7, covering Volcengine / volc support, Doubao Seedream models, provider-aware image defaults, provider discovery, and clearer ModelNotOpen troubleshooting.
Integration
md2wechat 2.0.5: Building a Real Confirmation Layer Before Publishing
A practical release note for md2wechat 2.0.5, focused on inspect, preview, cleaner JSON output, better error guidance, and why this release is more about trust than about adding more features.