Angreal
An MCP server providing AI assistants with discovery capabilities for angreal projects.
Angreal MCP Server
An MCP (Model Context Protocol) server that provides AI assistants with discovery capabilities for angreal projects.
Overview
This server exposes angreal's command tree structure to MCP-compatible clients, enabling AI assistants to:
- Discover available commands and tasks in your angreal project
- Understand project structure and capabilities
- Suggest appropriate commands based on context
Prerequisites
- angreal must be installed and available in your PATH
- Rust toolchain (for building from source)
Installation
# Install from crates.io
cargo install angreal_mcp
# Or install from git (latest development version)
cargo install --git https://github.com/colliery-io/angreal-mcp
Usage
With Claude Code
Create or update your Claude Code MCP configuration file at ~/.config/claude-code/mcp_servers.json:
{
"mcpServers": {
"angreal": {
"command": "angreal_mcp",
"args": [],
"env": {}
}
}
}
With Claude Desktop
Add to your Claude Desktop configuration:
{
"mcpServers": {
"angreal": {
"command": "angreal_mcp",
"args": [],
"cwd": "/path/to/your/angreal/project"
}
}
}
With Cline (VS Code)
Add to your Cline configuration:
{
"mcp": {
"servers": [
{
"name": "angreal",
"command": ["angreal_mcp"],
"cwd": "${workspaceFolder}"
}
]
}
}
Command Line Testing
You can test the MCP server directly via command line:
# List available tools
echo '{"jsonrpc": "2.0", "id": 1, "method": "tools/list"}' | angreal_mcp
# Get angreal command tree
echo '{"jsonrpc": "2.0", "id": 1, "method": "tools/call", "params": {"name": "angreal_tree", "arguments": {"format": "json"}}}' | angreal_mcp
Available Tools
angreal_check
Check if the current directory is an angreal project and get project status including available commands.
angreal_tree
Get a structured view of all available angreal commands and tasks in the project.
Parameters:
format(optional): Output format -"json"(default) or"human"
angreal_run
Execute an angreal command or task with optional arguments.
Parameters:
command(required): The angreal command/task to executeargs(optional): Additional arguments and flags
Agent Usage Guide
When working in angreal projects, use these tools for intelligent command discovery and execution:
Tool Usage Workflow
- Start with discovery: Use
angreal_checkto verify project status and capabilities - Explore commands: Use
angreal_treeto see available commands with rich metadata - Execute intelligently: Use
angreal_runwith context-aware parameter selection
Best Practices
- Always check
when_to_useandwhen_not_to_useguidance fromangreal_tree - Use parameter recommendations (
recommended_when) for better results - Verify prerequisites before executing commands
- Let command output and exit codes guide success validation
Common Patterns
- Setup workflows: Run setup/install commands before build/test commands
- Development cycles: Use test commands after code changes, build commands before deployment
- Parameter selection: Use verbose flags when debugging, production flags for releases
Troubleshooting
- If MCP server becomes unavailable, restart Claude Code to reinitialize
- Check that angreal binary is installed and accessible in PATH
- Verify you're in an angreal project directory (contains .angreal/ folder)
The angreal MCP server provides enhanced command metadata including usage context, parameter guidance, and intelligent categorization to enable better automation decisions.
Development
Running Tests
This project uses angreal as its test runner:
# Run tests using angreal
angreal test
Project Structure
angreal_mcp/
├── src/
│ ├── main.rs # Main server loop
│ ├── mcp.rs # MCP protocol implementation
│ └── angreal.rs # Angreal integration
├── examples/ # Configuration examples
└── tests/ # Integration tests
Building
# Debug build
cargo build
# Release build
cargo build --release
License
This project is dual-licensed under MIT OR Apache-2.0.
Related Servers
MCP Server Starter
A TypeScript starter project for building Model Context Protocol (MCP) servers with Bun.
MCP Gateway
A gateway to translate MCP tool calls into HTTP API requests, configurable via YAML.
MCP Server Demonstration
A demonstration on setting up and using MCP servers within Cursor, with Docker examples.
FastMCP
A TypeScript framework for building MCP servers with client session handling.
Remote MCP Server (Authless)
A remote MCP server deployable on Cloudflare Workers, without authentication.
Kafka MCP
A natural language interface to manage Apache Kafka operations.
OpenRouter MCP Client for Cursor
An MCP client for Cursor that uses OpenRouter.ai to access multiple AI models. Requires an OpenRouter API key.
Figma Context MCP
Extract code, assets, and component structures from Figma designs to use within AI workflows.
Swagger MCP
Scrapes Swagger UI to dynamically generate MCP tools at runtime using LLMs.
Creatify
MCP Server that exposes Creatify AI API capabilities for AI video generation, including avatar videos, URL-to-video conversion, text-to-speech, and AI-powered editing tools.