Skip to main content

MCP Server Configuration

MCP servers are configured in .radium/mcp-servers.toml in your workspace root.

Configuration File Location​

The configuration file is located at:

  • Workspace root: .radium/mcp-servers.toml
  • Default location: ~/.radium/mcp-servers.toml (if no workspace)

Transport Types​

Stdio Transport​

For local MCP servers that run as processes:

[[servers]]
name = "local-server"
transport = "stdio"
command = "mcp-server"
args = ["--config", "config.json"]

Fields:

  • name: Unique identifier for the server
  • transport: Must be "stdio"
  • command: Command to execute the MCP server
  • args: Optional array of command arguments

SSE Transport​

For remote servers using Server-Sent Events:

[[servers]]
name = "remote-server"
transport = "sse"
url = "https://api.example.com/mcp/sse"

Fields:

  • name: Unique identifier for the server
  • transport: Must be "sse"
  • url: SSE endpoint URL

HTTP Transport​

For remote servers using HTTP streaming:

[[servers]]
name = "http-server"
transport = "http"
url = "https://api.example.com/mcp"

Fields:

  • name: Unique identifier for the server
  • transport: Must be "http"
  • url: HTTP endpoint URL

Multiple Servers​

You can configure multiple MCP servers:

[[servers]]
name = "database-server"
transport = "stdio"
command = "mcp-database"
args = ["--db", "postgresql://localhost/mydb"]

[[servers]]
name = "api-server"
transport = "http"
url = "https://api.example.com/mcp"

Authentication​

For servers requiring authentication, see Authentication Guide.

Verifying Configuration​

Test your configuration:

# List configured servers
rad mcp list

# Test connection
rad mcp test

# Test specific server
rad mcp test --server database-server

Examples​

Example 1: Local Filesystem Server​

[[servers]]
name = "filesystem"
transport = "stdio"
command = "npx"
args = ["-y", "@modelcontextprotocol/server-filesystem", "/path/to/directory"]

Example 2: PostgreSQL Database Server​

[[servers]]
name = "postgres"
transport = "stdio"
command = "npx"
args = ["-y", "@modelcontextprotocol/server-postgres", "postgresql://user:pass@localhost/dbname"]

Example 3: Remote API with OAuth​

[[servers]]
name = "api-server"
transport = "http"
url = "https://api.example.com/mcp"
auth = {
    auth_type = "oauth",
    params = {
        token_url = "https://api.example.com/oauth/token",
        client_id = "your-client-id",
        client_secret = "your-client-secret"
    }
}

Example 4: Multiple Servers​

# Database server
[[servers]]
name = "postgres-main"
transport = "stdio"
command = "mcp-postgres"
args = ["postgresql://localhost/maindb"]

# File server
[[servers]]
name = "filesystem"
transport = "stdio"
command = "mcp-filesystem"
args = ["--root", "/home/user/documents"]

# Remote API
[[servers]]
name = "github-api"
transport = "http"
url = "https://api.github.com/mcp"
auth = {
    auth_type = "oauth",
    params = {
        token_url = "https://github.com/login/oauth/access_token",
        client_id = "github-client-id",
        client_secret = "github-client-secret"
    }
}

More Examples​

See the examples directory for complete configuration examples: