SCANMYVIBE MCP SERVER
Scan websites, audit GitHub repos, and query recent breaches from Claude Code, Cursor, or GitHub Copilot — all via a single HTTP endpoint. Zero install, zero API key.
INSTALL — CLAUDE CODE
claude mcp add --transport http scanmyvibe https://scanmyvibe.co/api/mcpThat's it. Restart Claude Code and try “scan stripe.com for vulnerabilities” or “is github.com/berriai/litellm safe to install?”
INSTALL — CURSOR
{
"mcpServers": {
"scanmyvibe": {
"url": "https://scanmyvibe.co/api/mcp"
}
}
}INSTALL — ANY MCP CLIENT
ScanMyVibe speaks the Model Context Protocol over HTTP (JSON-RPC 2.0). Point any MCP-compatible client at:
https://scanmyvibe.co/api/mcpStateless transport · No authentication · 60 requests/minute per IP
AVAILABLE TOOLS
EXAMPLES
Scan stripe.com for security issues
Calls scan_url("https://stripe.com"), gets 16-check report with CVSS scores, and explains each finding with a ready-to-paste fix.
Is github.com/tj-actions/changed-files safe to install?
Calls scan_repo("tj-actions", "changed-files"), finds the March 2025 supply-chain compromise in our curated incident DB, and warns you not to use unpinned versions.
What are the 5 most recent data breaches?
Calls list_breaches(limit: 5), summarizes each by company, affected count, and severity.
WHY MCP?
Model Context Protocol is the open standard for connecting AI assistants to external tools and data. Instead of copy-pasting URLs into a web scanner, your assistant runs the scan, reads the report, and suggests fixes — all in one flow.
ScanMyVibe's MCP server is stateless — every request is a single HTTP POST, no sessions, no sockets, no local binary. It works from any machine, any MCP client, on day one.