Runs the ZepAi MCP server to connect with the ZepAi module via UDP-like MVP command flow.
Configuration
View docs{
"mcpServers": {
"annamariakv-mcp-zepai": {
"command": "uv",
"args": [
"run",
"--directory",
"<path_to_the_mcp_directory>",
"python",
"-m",
"mcp_zepai"
],
"env": {
"ZEP_API_KEY": "YOUR_API_KEY"
}
}
}
}The ZepAi MCP Server lets you run a dedicated MCP endpoint that connects to the ZepAi module, enabling you to interact with model context data through a consistent, programmable interface. It streamlines how you start, configure, and secure the ZepAi MCP pipeline for your applications.
You run the MCP server locally and connect your MCP client to it to send and receive model context actions. Start the server with your chosen MCP directory and Python module so the server can load the ZepAi MCP implementation. Set your API key as an environment variable before starting. When the server is running, your client can request context data, submit actions, and receive responses through the MCP protocol using the standard client-server flow.
Prerequisites you need before starting the MCP server:
- A working shell environment (bash, zsh, etc.).
- The uv runtime available on your system. You will use uv to launch the MCP server.
- Python installed (to run the MCP Python module within the uv command).
- A path where you place the MCP directory that contains the ZepAi MCP module.
Follow these steps to install and prepare the server:
1) Create or choose a directory for the MCP server files.
2) Place or reference the ZepAi MCP module in that directory so the server can load it.
3) Set your ZEP API key as an environment variable before starting the server.
4) Start the server using the exact command below.
5) Verify the server is running and your MCP client can connect.
{
"mcpServers": {
"zepai": {
"command": "uv",
"args": [
"run",
"--directory",
"<path_to_the_mcp_directory>",
"python",
"-m",
"mcp_zepai"
],
"cwd": "<path_to_the_mcp_directory>",
"env": {
"ZEP_API_KEY": "<your_api_key_here>"
}
}
}
}