UTCP Documentation MCP Server
Assists AI coding agents in understanding and implementing UTCP by providing documentation search, manual validation, code generation, OpenAPI conversion, and an LLM-powered expert agent.
README
UTCP Documentation MCP Server
A comprehensive MCP server that helps AI coding agents understand and implement the Universal Tool Calling Protocol (UTCP)
๐ Features
- ๐ค LLM-Powered Expert Agent: Ask questions in natural language to an OpenAI-powered agent with deep UTCP knowledge
- ๐ Documentation Search: Semantic search across complete UTCP specification
- โ Manual Validation: Validate UTCP manuals against v1.0.1 spec with detailed errors
- ๐จ Code Generation: Generate UTCP manual templates for HTTP, CLI, MCP, SSE protocols
- ๐ OpenAPI Conversion: Convert OpenAPI 3.0 specs to UTCP manuals automatically
- ๐ Examples Library: Ready-to-use examples for weather APIs, GitHub, databases, CLI tools
- ๐ก Best Practices: Built-in guidance for naming, authentication, and implementation
๐ฆ Installation
NPM
npm install -g utcp-docs
From Source
git clone https://github.com/yourusername/utcp-docs-mcp-server.git
cd utcp-docs-mcp-server
npm install
npm run build
npm link
๐ง Configuration
For Claude Desktop
Add to your claude_desktop_config.json:
{
"mcpServers": {
"utcp-docs": {
"command": "utcp-docs",
"env": {
"DOCS_PATH": "/path/to/utcp-docs",
"OPENAI_API_KEY": "sk-..."
}
}
}
}
For Cursor IDE
Add to .cursor/mcp.json in your project:
{
"mcpServers": {
"utcp-docs": {
"command": "npx",
"args": ["utcp-docs"],
"env": {
"OPENAI_API_KEY": "sk-..."
}
}
}
}
Environment Variables
Create a .env file or set in your MCP config:
# OpenAI API Key (required for ask_utcp_expert tool)
OPENAI_API_KEY=sk-...
# Documentation path (default: current directory)
DOCS_PATH=.
# Server configuration
SERVER_NAME=utcp-docs
SERVER_VERSION=1.0.0
LOG_LEVEL=info
๐ ๏ธ Available Tools
๐ 1. Ask UTCP Expert (LLM Agent)
NEW! Ask an OpenAI-powered agent any question about UTCP:
{
"tool": "ask_utcp_expert",
"arguments": {
"question": "How do I authenticate with bearer tokens?"
}
}
The agent:
- Has deep knowledge of the entire UTCP specification
- Retrieves relevant documentation automatically
- Provides detailed explanations with code examples
- Includes source references
- Offers best practices and guidance
Example Questions:
- "What is body_template and when should I use it?"
- "Show me a complete HTTP tool example with OAuth2"
- "How do I chain CLI commands together?"
- "Why is my tool name failing validation?"
Requires: OpenAI API key (set OPENAI_API_KEY environment variable)
2. Search UTCP Documentation
Search the complete UTCP specification:
{
"tool": "search_utcp_docs",
"arguments": {
"query": "how to implement HTTP authentication",
"section": "protocols", // optional: introduction, protocols, guides, api
"limit": 5 // optional: max results
}
}
Example Response:
# Search Results for "HTTP authentication"
## Result 1: HTTP Protocol Authentication
**Section:** protocols
**Relevance:** 45
Authentication in HTTP protocols can be configured using the auth field...
3. Validate UTCP Manual
Validate your UTCP manual against the specification:
{
"tool": "validate_utcp_manual",
"arguments": {
"manual": {
"manual_version": "1.0.0",
"utcp_version": "1.0.1",
"tools": [...]
}
}
}
Response:
โ
Valid UTCP Manual
The manual passes all validation checks and conforms to UTCP v1.0.1 specification.
Or with errors:
โ Invalid UTCP Manual
## Errors
- /tools/0/name: Tool name "GetWeather" must use snake_case
- /tools/0/tool_call_template: HTTP template requires a URL
## Warnings
- /tools/0/description: Tool description should be at least 10 characters
4. Generate UTCP Manual
Generate a UTCP manual template:
{
"tool": "generate_utcp_manual",
"arguments": {
"tool_name": "get_weather",
"description": "Get current weather for a location",
"protocol": "http",
"endpoint": "https://api.openweathermap.org/data/2.5/weather",
"method": "GET",
"parameters": {
"location": {
"type": "string",
"description": "City name",
"required": true
}
},
"include_auth": true
}
}
Supported Protocols:
http- RESTful HTTP APIscli- Command-line toolsmcp- Model Context Protocol serverssse- Server-Sent Eventsstreamable_http- Streaming HTTP responses
5. Convert OpenAPI to UTCP
Convert OpenAPI 3.0 specifications to UTCP manuals:
{
"tool": "convert_openapi_to_utcp",
"arguments": {
"openapi_spec": {
"openapi": "3.0.0",
"info": { "title": "My API", "version": "1.0.0" },
"paths": {
"/users": {
"get": {
"operationId": "getUsers",
"summary": "List users"
}
}
}
}
}
}
6. Get UTCP Examples
Get ready-to-use example UTCP manuals:
{
"tool": "get_utcp_examples",
"arguments": {
"use_case": "weather" // weather, github, database, cli-tool
}
}
Available Examples:
weather- OpenWeatherMap API integrationgithub- GitHub REST API (repos, issues)database- Database query tools via MCPcli-tool- Git command-line wrapper
7. Get Best Practices
Get UTCP implementation best practices:
{
"tool": "get_best_practices",
"arguments": {
"topic": "naming" // naming, authentication, error-handling, testing, general
}
}
๐ Resources
The server provides these resources via MCP:
| Resource URI | Description |
|---|---|
utcp://docs/full |
Complete UTCP documentation (9700+ lines) |
utcp://docs/introduction |
UTCP introduction and overview |
utcp://docs/protocols |
All protocol documentation |
utcp://examples/weather |
Weather API example manual |
utcp://examples/github |
GitHub API example manual |
utcp://schema/manual |
UTCP JSON Schema for validation |
๐งช Usage Examples
Example 1: Creating a New UTCP Manual
// 1. Generate a template
const result = await callTool("generate_utcp_manual", {
tool_name: "send_email",
description: "Send an email via SendGrid API",
protocol: "http",
endpoint: "https://api.sendgrid.com/v3/mail/send",
method: "POST",
parameters: {
to: { type: "string", description: "Recipient email", required: true },
subject: { type: "string", description: "Email subject", required: true },
body: { type: "string", description: "Email body", required: true }
},
include_auth: true
});
// 2. Validate the generated manual
const validation = await callTool("validate_utcp_manual", {
manual: JSON.parse(result)
});
// 3. Get best practices for authentication
const practices = await callTool("get_best_practices", {
topic: "authentication"
});
Example 2: Converting Existing OpenAPI Spec
// 1. Read your OpenAPI spec
const openApiSpec = JSON.parse(fs.readFileSync("api-spec.json"));
// 2. Convert to UTCP
const utcpManual = await callTool("convert_openapi_to_utcp", {
openapi_spec: openApiSpec
});
// 3. Validate the result
const validation = await callTool("validate_utcp_manual", {
manual: JSON.parse(utcpManual)
});
Example 3: Searching Documentation
// Search for specific implementation details
const results = await callTool("search_utcp_docs", {
query: "how to handle streaming responses",
section: "protocols",
limit: 3
});
// Get an example to reference
const example = await callTool("get_utcp_examples", {
use_case: "github"
});
๐๏ธ Development
Setup
# Clone repository
git clone https://github.com/yourusername/utcp-docs-mcp-server.git
cd utcp-docs-mcp-server
# Install dependencies
npm install
# Build
npm run build
# Run in development mode
npm run dev
Testing
# Run all tests
npm test
# Run tests in watch mode
npm run test:watch
# Generate coverage report
npm run test:coverage
Project Structure
utcp-docs/
โโโ src/
โ โโโ index.ts # Entry point
โ โโโ server.ts # MCP server implementation
โ โโโ types/
โ โ โโโ utcp.ts # TypeScript types
โ โโโ services/
โ โ โโโ documentation.ts # Documentation search
โ โ โโโ validator.ts # UTCP validation
โ โ โโโ generator.ts # Manual generation
โ โ โโโ converter.ts # OpenAPI conversion
โ โโโ schemas/
โ โโโ utcp-manual.schema.json
โโโ docs/
โ โโโ examples/ # Example UTCP manuals
โโโ tests/
โ โโโ services/ # Unit tests
โโโ llms.txt # Complete UTCP documentation
โโโ package.json
๐ Requirements
- Node.js: 20.x or higher
- TypeScript: 5.3 or higher
- MCP SDK: 1.0.4 or higher
๐ค Contributing
Contributions are welcome! Please follow these steps:
- Fork the repository
- Create a feature branch (
git checkout -b feature/amazing-feature) - Commit your changes (
git commit -m 'Add amazing feature') - Push to the branch (
git push origin feature/amazing-feature) - Open a Pull Request
Development Guidelines
- Write tests for new features
- Follow TypeScript best practices
- Update documentation for API changes
- Ensure all tests pass before submitting PR
๐ Resources
UTCP Resources
MCP Resources
๐ Troubleshooting
Server Not Starting
Issue: Error: Cannot find module 'ajv-formats'
Solution:
npm install ajv-formats
npm run build
Documentation Not Loading
Issue: Error loading llms.txt
Solution: Ensure DOCS_PATH environment variable points to the correct directory:
export DOCS_PATH=/path/to/utcp-docs
Validation Errors
Issue: Tool names failing validation
Solution: Use snake_case naming:
- โ
get_user_data - โ
GetUserData - โ
getUserData
๐ License
MIT License - see LICENSE file for details
๐ Acknowledgments
- UTCP Team for creating the Universal Tool Calling Protocol
- Anthropic for the Model Context Protocol
- All contributors to this project
๐ Support
- Issues: GitHub Issues
- Discussions: GitHub Discussions
- Email: support@example.com
Built with โค๏ธ for the AI development community
Making UTCP implementation easier, one tool at a time.
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.