Claude Integration Overview

TroveSec exposes your AWS security data to Claude Desktop via the Model Context Protocol (MCP). Claude calls TroveSec tools to fetch real findings from your account — it never guesses or hallucinates data.

How it works

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

What Claude can do with TroveSec

• List and filter your security findings by severity, service, or keyword
• Get a full risk summary and trend vs your previous scan
• Identify SOC2 compliance gaps grouped by trust service criteria
• Generate a prioritised remediation plan with CLI fix commands
• Trigger new scans and check scan status
• Look up all findings for a specific resource (bucket, role, instance)

What it is not

TroveSec does not give Claude the ability to modify your AWS environment. All tools are read-only (except trigger_scan, which starts a scan — it does not touch your AWS resources directly).