DocsGo is a high-performance Markdown documentation server built with Go, featuring real-time rendering, full-text search, and access control. A lightweight alternative to static site generators like Hugo and Jekyll.
In modern software development, managing and sharing technical documentation is crucial:
- ❌ Static Generators (Hugo/Jekyll): Require recompilation on every change, no real-time preview
- ❌ Cloud Docs (Notion/Yuque): Data hosted by third parties, limited privacy and customization
- ❌ Traditional Wiki (Confluence): Bloated and complex, high learning curve
✅ DocsGo: Single binary deployment, real-time rendering, data ownership, ready to use
- Real-time Rendering - Auto-refresh on file changes, no recompilation needed
- Full-text Search - SQLite FTS-based search with automatic indexing and Chinese support
- Password Protection - Site-level and document-level dual encryption
- Embedded Deployment - Static assets bundled into binary for single-file deployment
- Responsive UI - Perfectly adapted for desktop and mobile devices
- Syntax Highlighting - Support for 100+ programming languages
- Technical Documentation Hub - API docs, architecture docs for dev teams
- Product Knowledge Base - User manuals, FAQs, changelogs
- Personal Knowledge Management - Learning notes, tech blogs
- Enterprise Internal Docs - Internal knowledge base, training materials
# Linux
wget https://github.com/jinsi-ai/docs-go/releases/latest/download/docs-go-linux-amd64
chmod +x docs-go-linux-amd64
./docs-go-linux-amd64
# Windows
curl -LO https://github.com/jinsi-ai/docs-go/releases/latest/download/docs-go-windows-amd64.exe
./docs-go-windows-amd64.exe
# macOS
wget https://github.com/jinsi-ai/docs-go/releases/latest/download/docs-go-darwin-amd64
chmod +x docs-go-darwin-amd64
./docs-go-darwin-amd64# Clone repository
git clone https://github.com/jinsi-ai/docs-go.git
cd docs-go
# Build (requires Go 1.22+)
go build -o docs-go
# Run
./docs-go
# Visit http://localhost:8080Place Markdown files in the docs/ directory for automatic rendering.
Create data/.env:
PORT=8080
DOCS_DIR=./docs
SITE_TITLE=Documentation Center
PASSWORD_SITE= # Site password (optional)
EMBED_WEB=true # Embed static assets (default: true)Or use command line arguments (higher priority):
./docs-go -port 8080 -docs ./docs -password-site mypassAdd YAML configuration at the beginning of documents:
---
title: "Page Title"
order: 1
password: "access password" # Optional
---| Parameter | Description |
|---|---|
title |
Page title |
order |
Sort weight (smaller = higher) |
password |
Document access password |
| Feature | DocsGo | Hugo |
|---|---|---|
| Architecture | Real-time server | Static compilation |
| Dev Experience | Live preview, no compile | Rebuild after changes |
| Deployment | Single binary, zero deps | Static files deployment |
| Search | Built-in full-text | Requires additional setup |
| Access Control | Built-in password | Requires external system |
| Best For | Dynamic docs, knowledge base | Static blogs, websites |
docs-go/
├── docs/ # Documentation directory
├── data/ # Data and configuration
├── web/ # Frontend assets
├── app/ # Application layer
└── pkg/ # Core packages
- Backend: Go + Gin Web Framework
- Frontend: Tailwind CSS + Vanilla JavaScript
- Markdown: goldmark (GFM support, syntax highlighting)
- Search: SQLite FTS5 full-text indexing
- File Watching: fsnotify cross-platform monitoring
# Run
go run main.go
# Format code
go fmt ./...
# Vet
go vet ./...
# Cross-compile
./build.sh # or build.bat (Windows)MIT © JinSi AI 309328809@qq.com
Go documentation generator, Markdown server, document management system, knowledge base tool, technical documentation platform, live preview, static site generator alternative, open source documentation tool, self-hosted wiki, enterprise documentation, team collaboration tool
近思切问，AI务实
NearThink AI, Pragmatic Tech
JinSi AI (近思AI) derives its name from the Confucian classic The Analects: "博学而笃志，切问而近思，仁在其中矣" ("Extensive study with steadfast purpose, earnest inquiry with reflection on things near at hand—benevolence lies therein").
In the wave of AI, some chase trends, some sell anxiety, and some speak in abstract concepts.
JinSi AI chooses a different path.
"近思" (NearThink) means starting from problems at hand, thinking about what truly matters. We don't talk about the distant future—we solve present problems. We don't chase hot topics—we dig into fundamentals.
Pragmatic, grounded, step by step—from near to far.
- Pragmatic | 务实 — Solve real problems, not chase hype
- Deep Thinking | 深度 — Go beyond surface, dig into fundamentals
- Progressive | 渐进 — From shallow to deep, accompany your growth
- Genuine | 真诚 — Authentic content, no empty talk
This project is built with the JinSi AI philosophy: pragmatic tools for real problems. We believe good documentation tools should be simple, fast, and just work—so you can focus on what matters: creating great content.
近思AI，解决真问题
JinSi AI—Solving Real Problems