Skip to content

MCP servers

Add local and remote MCP tools.

You can add external tools to codeflow using the Model Context Protocol, or MCP. codeflow supports both:

  • Local servers
  • And remote servers

Once added, MCP tools are automatically available to the LLM alongside built-in tools.


Configure

You can define MCP servers in your codeflow config under mcp.


Local

Add local MCP servers using "type": "local" within the MCP object. Multiple MCP servers can be added. The key string for each server can be any arbitrary name.

codeflow.json
{
"$schema": "https://codeflow.ai/config.json",
"mcp": {
"my-local-mcp-server": {
"type": "local",
"command": ["bun", "x", "my-mcp-command"],
"enabled": true,
"environment": {
"MY_ENV_VAR": "my_env_var_value"
}
},
"my-different-local-mcp-server": {
"type": "local",
"command": ["bun", "x", "my-other-mcp-command"],
"enabled": true
}
}
}

You can also disable a server by setting enabled to false. This is useful if you want to temporarily disable a server without removing it from your config.


Remote

Add remote MCP servers under mcp with "type": "remote".

codeflow.json
{
"$schema": "https://codeflow.ai/config.json",
"mcp": {
"my-remote-mcp": {
"type": "remote",
"url": "https://my-mcp-server.com",
"enabled": true,
"headers": {
"Authorization": "Bearer MY_API_KEY"
}
}
}
}

Local and remote servers can be used together within the same mcp config object.

codeflow.json
{
"$schema": "https://codeflow.ai/config.json",
"mcp": {
"my-local-mcp-server": {
"type": "local",
"command": ["bun", "x", "my-mcp-command"],
"enabled": true,
"environment": {
"MY_ENV_VAR": "my_env_var_value"
}
},
"my-remote-mcp": {
"type": "remote",
"url": "https://my-mcp-server.com",
"enabled": true,
"headers": {
"Authorization": "Bearer MY_API_KEY"
}
}
}
}

Per agent

If you have a large number of MCP servers you may want to only enable them per agent and disable them globally. To do this:

  1. Configure the MCP server.
  2. Disable it as a tool globally.
  3. In your agent config enable the MCP server as a tool.
codeflow.json
{
"$schema": "https://codeflow.ai/config.json",
"mcp": {
"my-mcp": {
"type": "local",
"command": ["bun", "x", "my-mcp-command"],
"enabled": true
}
},
"tools": {
"my-mcp*": false
},
"agent": {
"my-agent": {
"tools": {
"my-mcp*": true
}
}
}
}