docmd was engineered to occupy the space between simple Markdown parsers and heavy-weight framework applications (like Docusaurus). It provides the speed and SEO of a static site with the interactive feel of a modern Single Page Application (SPA).
Feature Matrix
| Feature | docmd | Docusaurus | MkDocs | Mintlify |
|---|---|---|---|---|
| Ecosystem | Node.js | React.js | Python | SaaS |
| Navigation | Instant SPA | React SPA | Full Reloads | Hosted SPA |
| Base Payload | < 20kb | > 200kb | Minimal | Medium |
| Versioning | Native | Complex FS | Plugin-based | Native |
| i18n Support | Coming Soon | Native | Plugin-based | Native |
| Search | Built-in (Offline) | Algolia (Cloud) | Built-in | Cloud-based |
| PWA | Built-in (Plugin) | Plugin | None | Hosted |
| AI Optimization | Built-in (llms.txt) | Manual | None | Proprietary |
| Setup | Instant (-z) | ~15 mins | ~10 mins | ~5 mins |
The docmd Advantage
1. AI-First Architecture
Unlike traditional generators, docmd recognizes that AI agents are now primary consumers of technical documentation. Our built-in LLM Plugin automatically generates llms.txt and llms-full.txt files, providing structured context for LLM-driven development tools.
2. Zero-Config PWA
Transform your documentation into a high-performance, installable mobile and desktop application with a single plugin. docmd handles the service worker logic, manifest generation, and offline caching automatically.
3. Balanced Performance
By generating pure, semantic HTML and subsequent navigations via a micro-SPA router, docmd ensures peak SEO performance without sacrificing the fluidity of a modern web application.
Choosing the Right Tool
- Use Docusaurus if: You require high-complexity React components within your Markdown (MDX) or have urgent multi-language needs today.
- Use MkDocs if: Your environment is strictly Python-based and you prefer the legacy static page-reloading model.
- Use docmd if: You value speed, developer experience (DX), a modern SPA feel, and want your documentation to be easily digestible by both humans and AI agents.