> ## Documentation Index
> Fetch the complete documentation index at: https://docs.plyne.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Connect VS Code to Plyne

> Add Plyne to Copilot's remote MCP configuration and authorize a workspace.

Use VS Code's Copilot MCP setup to connect to the remote Plyne server. Select the **Copilot** harness in a chat session, run **Chat: Open Customizations**, choose **MCP Servers**, and add a remote HTTP server.

Use `https://api.plyne.ai/mcp` as the server URL. Choose the configuration destination offered by your installed VS Code version, then complete the browser authentication.

## Manual Copilot configuration

For current Copilot configurations, merge this entry into `~/.copilot/mcp-config.json`, or `$COPILOT_HOME/mcp-config.json` when that environment variable is set:

```json mcp-config.json theme={null}
{
  "mcpServers": {
    "plyne": {
      "type": "http",
      "url": "https://api.plyne.ai/mcp"
    }
  }
}
```

For shared project configuration, current Copilot versions also support a workspace-root `.mcp.json` using the same `mcpServers` format.

> [!NOTE]
>
> Older VS Code setups use the editor-managed `.vscode/mcp.json` format with a top-level `servers` object. That is a separate configuration format. Use the setup flow and destination documented for your installed harness rather than mixing both schemas.

## Verify the connection

Sign in to Plyne when prompted, choose the workspace, and approve read access. Ask Copilot: **“List my Signal Agents.”** Adding configuration alone does not confirm that the current chat session can use the server; verify a tool call.

See Microsoft's [MCP configuration reference](https://code.visualstudio.com/docs/agents/reference/mcp-configuration) for current locations and setup details.


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.