Securick in your editor (MCP)
Connect Cursor, Claude Code or Windsurf to Securick. Your editor's AI can then scan your app, read the issues and pull a fix prompt for each one, right where you're coding.
1. Create an API key
On your account page, under API keys, create a key. It starts with lp_live_ and is shown once. Treat it like a password: keep it out of code you commit (use an environment variable, as below), and revoke it from the same page if it leaks.
2a. Cursor
Add this to .cursor/mcp.json in your project (or ~/.cursor/mcp.json for every project), and set SECURICK_API_KEY in your environment.
{
"mcpServers": {
"securick": {
"url": "https://your-securick-site/mcp",
"headers": { "Authorization": "Bearer ${env:SECURICK_API_KEY}" }
}
}
}
2b. Claude Code
Run this once in your terminal:
claude mcp add --transport http securick https://your-securick-site/mcp --header "Authorization: Bearer $SECURICK_API_KEY"
Or in a .mcp.json file: {"mcpServers": {"securick": {"type": "http", "url": "…/mcp", "headers": {"Authorization": "Bearer ${SECURICK_API_KEY}"}}}}. Check it with claude mcp list.
2c. Windsurf
Add this to ~/.codeium/windsurf/mcp_config.json (Windsurf uses serverUrl):
{
"mcpServers": {
"securick": {
"serverUrl": "https://your-securick-site/mcp",
"headers": { "Authorization": "Bearer lp_live_…" }
}
}
}
3. What your editor can do
scan_url: start a public scan of a URL. If it's one of your projects, the scan is added there.get_scan: status, launch score and public score. Full findings for paid reports and your own Builder or Agency projects.list_issues: a project's issues, most severe first, filtered by status or severity.get_fix_prompt: the fix prompt and masked proof for an issue, ready for your editor's AI.mark_issue: ignore an issue or accept its risk (with a reason), or reopen it.
Try asking: "Scan my-app.lovable.app with Securick and fix the critical issues." Limits: 60 calls per key per hour; scans share your hourly scan allowance.