iam-review
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: No security concerns were identified. The references to 'ignore previous instructions' within the skill file are part of defensive injection-hardening guidelines instructing the agent to flag prompt injection attempts in reviewed metadata as security findings rather than complying with them. The skill is entirely read-only, restricts its tools to basic file reading (Read, Grep, Glob), and explicitly forbids data exfiltration or credential disclosure.
Audit Metadata