The recommended way to get started is the CLI. The MCP
server is an optional, advanced integration for AI IDEs that support the
MCP protocol.
CLI usage does not require an API key in open beta. Some MCP client setups may
still ask for an explicit
--api-key parameter depending on the host tool.Cursor
Cursor
Go to:
Settings -> Cursor Settings -> MCP -> Add new global MCP serverPaste the following configuration:Claude Code
Claude Code
Run the following command:
VS Code
VS Code
Add this to your VS Code settings (JSON):
Windsurf
Windsurf
Add this to your
~/.codeium/windsurf/mcp_config.json file:Claude Desktop
Claude Desktop
Add this to your
claude_desktop_config.json file:Cline
Cline
Go to:
Settings -> MCP Servers -> Add new MCP serverZed
Zed
Add this to your Zed settings.json:
JetBrains IDEs
JetBrains IDEs
Go to:
Settings -> Tools -> AI Assistant -> Model Context Protocol (MCP)Amazon Q Developer CLI
Amazon Q Developer CLI
Add this to your
~/.aws/amazonq/mcp.json file:Warp
Warp
Go to:
Settings -> AI -> Manage MCP serversRoo Code
Roo Code
Go to:
Settings -> MCP Servers -> Add new MCP serverAugment Code
Augment Code
Go to:
Settings -> MCP Servers -> Add ServerTrae
Trae
Go to:
Settings -> MCP Servers -> Add ServerGemini CLI
Gemini CLI
Add this to your
~/.gemini/settings.json file:Advanced Usage
Using Bun
Using Bun
Using Deno
Using Deno