NYC

reviewing-changes

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFE
Full Analysis
  • SAFE (SAFE): The skill consists entirely of markdown instructions and metadata. No executable scripts, shell commands, or external network requests are present.
  • Indirect Prompt Injection (SAFE): The skill naturally processes untrusted pull request data as part of its code review function. 1. Ingestion points: PR titles and code changes (SKILL.md Steps 1 and 2). 2. Boundary markers: Employs internal thinking tags to structure analysis. 3. Capability inventory: Reads local reference files and utilizes standard JIRA/GitHub tools. 4. Sanitization: Relies on the base agent's existing safety protocols for handling untrusted data.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 06:25 PM