method-crm
Production-ready Model Context Protocol (MCP) server for Method CRM API integration. Enables LLMs to interact with Method CRM through 20 comprehensive tools.
claude mcp add --transport stdio avisangle-method-crm-mcp python -m method_mcp.server \ --env METHOD_API_KEY="your_api_key_here" \ --env METHOD_HTTP_PORT="8000" \ --env METHOD_TRANSPORT="stdio" \ --env METHOD_API_BASE_URL="https://rest.method.me/api/v1/"
How to use
The Method CRM MCP Server exposes a comprehensive set of tools that let large language models interact with Method CRM data through a stable MCP interface. It provides 20+ domain tools for table operations, file management, user information, event automation, and API key management, all with type-safe input validation, clear error handling, and support for multiple response formats. You can run the server locally in stdio mode for integrated desktop MCP clients, or enable HTTP transport for remote clients, allowing scalable access from cloud-based agents. The server handles rate limiting with exponential backoff and retry guidance, simplifying integration with rate-constrained environments. Tools are organized into categories such as Table Operations, File Management, User Information, Event Automation, and API Key Management, each with read-only and destructive capabilities as described in the feature list. When using an MCP client or Inspector, you can configure the client to point at the method-crm MCP server and issue tool calls that perform CRM actions like querying customers, uploading documents, creating events, or managing API keys. The server also supports multiple response formats (JSON and Markdown) to suit different client capabilities.
How to install
Prerequisites:
- Python 3.10 or higher
- Method CRM account with API access
- API key (obtain from Method CRM dashboard)
Install dependencies:
# Clone the repository
git clone https://github.com/avisangle/method-crm-mcp.git
cd method-crm-mcp
# Install dependencies
pip install -e .
# Or install with dev dependencies
pip install -e ".[dev]"
Configuration and setup:
- Copy the example environment file:
cp .env.example .env
- Edit .env to configure authentication and transport:
# Required: Method CRM API Key
METHOD_API_KEY=your_api_key_here
# Optional: Customize API base URL
METHOD_API_BASE_URL=https://rest.method.me/api/v1/
# Optional: Transport configuration
METHOD_TRANSPORT=stdio # or 'http'
METHOD_HTTP_PORT=8000 # only used if TRANSPORT=http
Run the server:
# StdIO local run (default)
python -m method_mcp.server
# Or run via module directly
python src/method_mcp/server.py
For remote access via HTTP transport:
# Set transport in .env (already shown above)
METHOD_TRANSPORT=http
METHOD_HTTP_PORT=8000
# Start server
python -m method_mcp.server
Note: The server listens at http://localhost:8000 when HTTP transport is enabled.
Additional notes
Tips and common considerations:
- Ensure METHOD_API_KEY is kept secret and not committed to version control.
- If you switch to HTTP transport, ensure network security and appropriate firewall rules are configured to allow inbound connections on the specified port.
- The MCP server supports pagination, rate limit handling, and retry strategies; adjust API rate expectations in your client accordingly.
- The available tools are categorized (Tables, Files, Users, Events, API Keys). When building prompts for your LLM, prefer filtering and pagination parameters to minimize data transfer and improve responsiveness.
- If you encounter authentication issues, verify METHOD_API_BASE_URL and the API key scope in your Method CRM account.
- For debugging, you can run in dev mode and inspect logs for tool invocations and responses.
Related MCP Servers
lc2mcp
Convert LangChain tools to FastMCP tools
MCPHammer
MCP security testing framework for evaluating Model Context Protocol server vulnerabilities
skill-to
Convert AI Skills (Claude Skills format) to MCP server resources - Part of BioContextAI
fastmcp-builder
A comprehensive Claude Code skill for building production-ready MCP servers using FastMCP. Includes reference guides, runnable examples, and a complete implementation with OAuth, testing, and best practices.
openroad
Model Context Protocol (MCP) server for OpenROAD
lucius
A Model Context Protocol (MCP) server for Allure TestOps.