specification

Pass

Audited by Gen Agent Trust Hub on Jul 14, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill generates a local, self-contained HTML review artifact using resources/spec-review-template.html. This process is handled locally by the agent and does not involve remote code execution or external downloads.
  • [SAFE]: The SKILL.md instructions include mandatory sanitization steps for the agent, requiring that all instances of < be escaped as \u003c when populating the JSON data slot. This is a specific mitigation against potential script injection or XSS within the generated review page.
  • [SAFE]: The review template uses standard browser features (localStorage and navigator.clipboard) to facilitate local feedback persistence and user-initiated data transfer back to the agent session. It does not perform any unauthorized network operations or data exfiltration.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 14, 2026, 03:12 PM
Security Audit — agent-trust-hub — specification