MCP Integrations – Connect TroveSec to Claude, Cursor, Windsurf & More

TroveSec exposes your AWS security data via the Model Context Protocol (MCP) — an open standard for giving AI models access to external tools and data. Any AI client that supports MCP tool calling can connect to TroveSec.

How it works

You ask your AI client a security question
  → The client calls a TroveSec MCP tool
    → TroveSec fetches real data from your latest AWS scan
      → The AI answers with your actual resource names, findings, and fix commands

The AI never guesses or generates findings — it reads them directly from your scan results.

Supported clients

Claude Desktop Anthropic's desktop app. The most capable MCP client for multi-step security analysis.

Windsurf Codeium's AI IDE. Ask security questions without leaving your editor.

Cursor AI-native code editor with MCP support via project or global config.

Perplexity Perplexity Pro with MCP tools support.

Any other MCP-compatible client works the same way — use the endpoint and auth details below with whatever client you prefer.

Connection details

Endpoint	https://mcp.trovesec.io/mcp
Auth	Authorization: Bearer <your-api-key>
Protocol	MCP over HTTP (streamable)
Transport helper	mcp-remote — needed for clients that use stdio-based MCP config

Generate your API key under Integrations → API Keys in your dashboard. Keys start with cs_live_.

The 16 tools

TroveSec exposes 16 tools covering findings, risk prioritisation, posture tracking, SOC2 compliance, and risk acceptance. See the Tools Reference for the full list with example prompts.