AGENT CONTEXT SCANNER

Scan SKILL.md, AGENTS.md, CLAUDE.md, Cursor rules, Copilot instructions, and MCP configs before your team trusts them. Full scans require Pro, Team, or Enterprise access.

Agent Context Risk Report

Checking access...

We auto-discover SKILL.md, AGENTS.md, CLAUDE.md, Cursor rules, Copilot instructions, and MCP configs.

What we inspect

Prompt injection
Command execution
Network exfiltration
Secret exposure
MCP permissions
Obfuscated payloads
Unsafe installs
Hidden persistence

Agent Context Risk Report will appear here

Scan a repo or upload a ZIP to inventory trusted agent files