VS Code
Set up SiteCMD with VS Code to fix issues with full scan context.
VS Code has built-in MCP support, so Copilot’s agent mode can read your SiteCMD scan results once you point it at the bundled server.
SiteCMD’s in-app Integrations connect button currently writes config for Cursor, Claude Code, and Codex. For VS Code, add the server by hand as below.
Configuration
Create .vscode/mcp.json in your workspace (or open your user config with MCP: Open User Configuration from the Command Palette) and add the bundled server. The top-level key is servers, and stdio servers set "type": "stdio":
{
"servers": {
"sitecmd": {
"type": "stdio",
"command": "node",
"args": [
"--disable-warning=ExperimentalWarning",
"/Applications/SiteCMD.app/Contents/Resources/sitecmd-mcp/sitecmd-mcp.mjs"
]
}
}
}
That path is the macOS location; on Windows and Linux the script lives in the resources/sitecmd-mcp/ folder next to the installed SiteCMD binary. MCP tools are available in Copilot’s Agent mode (see GitHub Copilot). The Cline extension uses its own setup, covered in Cline.
Note: VS Code uses the servers key, not the mcpServers key that Cursor and Claude Desktop use. Pasting an mcpServers block here won’t work.
What you get
Once connected, your AI assistant can call these tools:
get_projects- list all projects tracked in SiteCMDget_scan_score- fetch the latest score and category breakdownget_issues- list failing issues ranked by severity and impactget_fix_prompts- generate fix prompts for selected issuesget_scan_history- retrieve score history over timeget_dismissed_issues- review dismissed or not-applicable issuescompare_scans- compare two scans for fixed, new, and still-open issuesrequest_scan- get instructions for running a scan and comparing results