conduit
Conduit - A Model Context Protocol (MCP) server for Phabricator and Phorge.
Documentation
Conduit - The MCP Server for Phabricator and Phorge
Conduit is a Model Context Protocol (MCP) server that provides seamless integration with Phabricator and Phorge APIs, enabling advanced automation and interaction capabilities for developers and tools.
Conduit
Modern HTTP Client: Built with `httpx` for HTTP/2 support and better performance
MCP Integration: Ready-to-use MCP tools for task management
Type Safety: Full type hints and runtime validation for better development experience
Secure: Token-based authentication with environment variable configuration
Enhanced Features:
- Advanced error handling with detailed error codes and suggestions
- Token optimization for efficient API responses
- Smart pagination and intelligent data limiting
- Runtime validation client for type safety
- Configurable client with caching and retry mechanisms
Usage
Via `uvx`
You need to install `uv` first. If it is not installed, run the following command:
curl -LsSf https://astral.sh/uv/install.sh | shAfter installation, restart your shell or terminal to apply the environment variable changes.
Then run:
uvx --from git+https://github.com/mcpnow-io/conduit conduit-mcpFrom Source
To install from source for development or contribution:
# Clone the repository
git clone https://github.com/mcpnow-io/conduit.git
cd conduit
# Install in development mode with all dependencies
pip install -e .[dev]This will install the package in editable mode with all development dependencies.
Docker
We are still working on Docker support. We estimate it will be available soon.
As HTTP/SSE Server
Conduit can run as an HTTP/SSE server for multi-user scenarios. This mode allows multiple clients to connect simultaneously, each using their own authentication tokens.
conduit-mcp --host 127.0.0.1 --port 8000When running as an HTTP server, authentication tokens are provided via HTTP headers instead of environment variables.
X-PHABRICATOR-TOKEN: your-32-character-token-hereConfiguration
Before running the server, you need to set up the following environment variables:
Environment Variables
export PHABRICATOR_TOKEN=your-api-token-here
export PHABRICATOR_URL="https://your-phabricator-instance.com/api/"
export PHABRICATOR_PROXY="socks5://127.0.0.1:1080" # Optional, if your network is behind a firewall
export PHABRICATOR_DISABLE_CERT_VERIFY=1 # Optional, if your network is under HTTPS filter (WARNING: Disabling certificate verification can expose you to security risks. Only set this if you trust your network environment.)Do note that in HTTPS/SSE mode, `PHABRICATOR_TOKEN` is NOT needed.
Getting Your API Token
1. Log into your Phabricator instance
2. Go to Settings > API Tokens
3. Generate a new token
4. Copy the 32-character token and use it as `PHABRICATOR_TOKEN`
Contributing
There are many ways in which you can participate in this project, for example:
- Submit bugs and feature requests, and help us verify as they are checked in
- Review source code changes
- Review the wiki and make pull requests for anything from typos to additional and new content
If you are interested in fixing issues and contributing directly to the code base, please see the document How to Contribute:
License
Copyright (c) 2025 mpcnow.io. all rights reserved.
Licensed under the MIT license.
Frequently asked questions
What is conduit?
conduit is Conduit - A Model Context Protocol (MCP) server for Phabricator and Phorge.
How do I install conduit?
Open the GitHub repository and follow its README. Most MCP servers are added to your client's MCP config, then called by your agent.
Is conduit open source?
Yes — it is hosted on GitHub at https://github.com/mcpnow-io/conduit and has 4 stars.
Related MCP tools
An LLM agent that conducts deep research (local and web) on any given topic and generates a long report with citations. Built for the Model Context Protocol to
Expose your FastAPI endpoints as Model Context Protocol (MCP) tools, with Auth! Python-based implementation. Trusted by 11000+ developers.
AI-powered reverse engineering assistant that bridges IDA Pro with language models through MCP. Python-based implementation. Trusted by 4100+ developers.
An MCP server that autonomously evaluates web applications. Python-based implementation. Trusted by 1100+ developers. Trusted by 1100+ developers.
An official Qdrant Model Context Protocol (MCP) server implementation Python-based implementation. Trusted by 1000+ developers.
A middleware to provide an openAI compatible endpoint that can call MCP tools Python-based implementation. Trusted by 800+ developers.
Run your own MCP server? See who uses it and what to fix.
Measure it with TrackMCP