Overview
What is MCP Server for Zep Cloud?
MCP Server for Zep Cloud provides a bridge between Claude Desktop and the Zep Cloud API, enabling memory management for AI assistants.
How to use MCP Server for Zep Cloud?
Install Python 3.8+, clone the repository, create a virtual environment, install dependencies from config/requirements.txt, copy .env.example to .env and add your Zep Cloud API key. Run the server via scripts/run_server.sh (macOS/Linux) or scripts/run_server.bat (Windows). Configure Claude Desktop by adding the absolute paths to your Python executable and core/zep_cloud_server.py in claude_desktop_config.json.
Key features of MCP Server for Zep Cloud
- User management (create, retrieve, update, delete, list users)
- Collections management (create, get, update, delete, list collections)
- Memory operations (add, get, list, search memories)
- Fallback mode when Zep Cloud API is inaccessible
- Modern implementation using the FastMCP approach
Use cases of MCP Server for Zep Cloud
- Giving Claude Desktop persistent memory for users and sessions
- Storing and retrieving user-specific conversation context
- Searching across stored memories by content
- Managing memory collections for different domains or applications
FAQ from MCP Server for Zep Cloud
What is fallback mode?
If the server cannot connect to the Zep Cloud API, it automatically starts in fallback mode. In this mode, all API operations are simulated and return success, but no data is sent to or received from the actual API.
How do I configure Claude Desktop?
Open your Claude Desktop configuration file at ~/Library/Application Support/Claude/claude_desktop_config.json (macOS) or %APPDATA%\Claude\claude_desktop_config.json (Windows) and add the absolute paths to your Python executable and the core/zep_cloud_server.py script.
What tools are available after configuration?
Claude Desktop will have access to tools for user management (create, get, update, delete, list users), collections management (create, get, update, delete, list collections), memory operations (add, get, list, search memories), and a check_connection tool.
What are the requirements?
Python 3.8+ and a Zep Cloud API key are required.
How should I protect my API key?
Never commit your API key to version control. Use the provided .env.example as a template, create your own .env file with your actual API key, and regularly rotate keys if compromised.
