Getting Started with Cursor, VS Code & Windsurf
Speak AI's MCP server works with any editor that supports HTTP MCP. This guide covers Cursor, VS Code with GitHub Copilot, Windsurf, and Zed.
Step 1: Get your API key
Go to app.speakai.co/developers and copy your API key.
Cursor
Open Cursor Settings → MCP → Add MCP Server, paste https://api.speakai.co/v1/mcp as the URL and add the authorization header. Or add directly to your MCP config:
{
"mcpServers": {
"speakai": {
"url": "https://api.speakai.co/v1/mcp",
"headers": { "Authorization": "Bearer your-api-key-here" }
}
}
}
VS Code (GitHub Copilot)
Add to your VS Code settings.json:
{
"github.copilot.mcpServers": {
"speakai": {
"url": "https://api.speakai.co/v1/mcp",
"headers": { "Authorization": "Bearer your-api-key-here" }
}
}
}
Open the Command Palette, run MCP: List Servers, and verify Speak AI appears.
Windsurf
Add to ~/.codeium/windsurf/mcp_config.json:
{
"mcpServers": {
"speakai": {
"serverUrl": "https://api.speakai.co/v1/mcp",
"headers": { "Authorization": "Bearer your-api-key-here" }
}
}
}
Restart Windsurf after saving. The Speak AI tools will appear in the Cascade panel.
Zed
Zed uses the stdio transport (local npm package), not the HTTP URL. If your workflow requires HTTP transport, Cursor or VS Code are better choices.
{
"context_servers": {
"speakai": {
"command": { "path": "npx", "args": ["-y", "@speakai/mcp-server"] },
"settings": {}
}
}
}
Still stuck? Chat with us below or email [email protected].