Gencodedoc

🚀 GenCodeDoc

An intelligent versioning system and documentation generator with full support for the MCP protocol.

GenCodeDoc is a powerful tool that revolutionizes the way you track the evolution of a code project. It combines an intelligent snapshot system with deduplication, a Markdown documentation generator, and a comprehensive interface (CLI, REST API, MCP) for automation and integration with AI assistants.

For detailed technical documentation (architecture, full API, etc.), please refer to our .

🚀 Quick Start

Once installed, here's how to get started with an existing project:

1. Initialize GenCodeDoc in your project:

   # Navigate to your project
   cd /path/to/your/project
   
   # Initialize with a preset (e.g., python)
   poetry run gencodedoc init --preset python
   

   Available presets: python, nodejs, go, web.

2. Create your first snapshot:

   poetry run gencodedoc snapshot create --message "Initial project version" --tag v1.0
   

3. Generate documentation:

   poetry run gencodedoc doc generate --output "PROJECT_DOCS.md"
   

   A PROJECT_DOCS.md file will be created with complete documentation.

4. (Optional) Enable autosave:

   poetry run gencodedoc config set autosave.enabled true
   

✨ Features

• 📸 Intelligent Snapshots: Create "photos" of your project with ~70% space savings through hash-based content deduplication.
• 🔄 Intelligent Autosave: Protect your work with 3 automatic saving modes (time-based, change-based, or hybrid).
• 📝 Automated Documentation: Generate comprehensive Markdown documentation for your project, including file trees and code snippets.
• 🔍 Advanced Diff: Compare any versions of your project with unified diff (Git-style), JSON (for scripts), or semantic (AST, experimental).
• 🔌 Triple Interface: Use GenCodeDoc in the way that suits you best:
  • CLI: A complete and intuitive command-line interface.
  • REST API: Integrate GenCodeDoc into your own applications via HTTP.
  • MCP: Control the tool with an AI (Gemini, Claude) through 17 tools exposed via the Model-Context-Protocol.

📦 Installation

Prerequisites

• Python 3.10+
• Poetry for dependency management.

Installation Steps

1. Clone the repository:

   git clone https://github.com/VOTRE_NOM/gencodedoc.git
   cd gencodedoc
   

2. Install dependencies with Poetry:

   poetry install
   

3. Verify the installation:

   poetry run gencodedoc --help
   

🔌 MCP Integration (For AI Assistants)

GenCodeDoc is designed to be controlled by AIs. It exposes its 17 tools via 3 transport modes.

• stdio: For CLI clients like Gemini CLI.
• SSE (Server-Sent Events): For web or desktop applications like Claude Desktop.
• REST: For any custom integration via an HTTP API.

Configuration for Gemini CLI (stdio)

1. Find the path to your Poetry virtual environment:

   # From the gencodedoc directory
   poetry env info --path 
   # Copy the returned path
   

2. Add the server to your MCP client configuration: (Example for gemini-cli)

   {
     "mcpServers": {
       "gencodedoc": {
         "command": "/path/to/your/poetry-venv/bin/python",
         "args": ["-m", "gencodedoc.mcp.server_stdio"],
         "env": {
           "PROJECT_PATH": "/path/to/your/target-project"
         }
       }
     }
   }
   

   Replace /path/to/your/poetry-venv with the path from step 1 and PROJECT_PATH with the project you want to manage.

Start the SSE / REST Servers

# For the SSE server (Claude Desktop)
poetry run python -m gencodedoc.mcp.server_sse

# For the REST server
poetry run python -m gencodedoc.mcp.server

Both servers run on http://127.0.0.1:8000.

🧰 Tool Reference

GenCodeDoc offers a complete CLI and a rich MCP/REST API.

Main CLI Commands

• gencodedoc init: Initialize a project.
• gencodedoc snapshot create|list|show|diff|restore|delete: Manage snapshots.
• gencodedoc doc generate|preview|stats: Manage documentation.
• gencodedoc config show|edit|set|preset|ignore: Manage configuration.
• gencodedoc status: Show the current project status.

MCP Tools (17 tools)

A summary of the tools available for AIs:

• Snapshot Management (6 tools): create_snapshot, list_snapshots, get_snapshot_details, restore_snapshot, delete_snapshot, diff_versions.
• Documentation (3 tools): generate_documentation, preview_structure, get_project_stats.
• Project Management (2 tools): init_project, get_project_status.
• Configuration (3 tools): get_config, set_config_value, apply_preset, manage_ignore_rules.
• Autosave (3 tools): start_autosave, stop_autosave, get_autosave_status.

Once configured, you can simply ask your AI:

"Create a snapshot with the tag v2.1 and the message 'Fixed connection bug'" "Show me the differences between version v2.0 and the current version" "Generate complete project documentation"

🏗️ Architecture and Operation

GenCodeDoc is built around a logical core that manages configuration, versioning, and document generation. This core is exposed via different interfaces (CLI, MCP).

The storage is optimized for both performance and space efficiency:

• SQLite is used to store all metadata (snapshots, files, etc.).
• The content of files is deduplicated via a SHA256 hash. The same file, even present in 100 snapshots, is stored only once.
• The contents are then compressed with zstandard to further reduce the disk footprint.

🤝 Contribution

Contributions are welcome!

1. Fork the project.
2. Create a branch for your feature (git checkout -b feature/my-great-feature).
3. Commit your changes (git commit -m 'Add my great feature').
4. Push your branch (git push origin feature/my-great-feature).
5. Open a Pull Request.

📄 License

This project is licensed under the MIT License. See the LICENSE file for more details.