find-react-native-animation-opportunities

Pass

Audited by Gen Agent Trust Hub on Jul 24, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill is strictly read-only and explicitly prohibits the agent from editing application source code or creating implementation plans, which minimizes the risk of unauthorized file modifications.
  • [PROMPT_INJECTION]: The skill includes proactive defense instructions against indirect prompt injection by requiring the agent to treat repository text as data and to flag prompt-like instructions found within the source code instead of following them.
  • [DATA_EXFILTRATION]: While the skill inspects project files such as package.json and Expo configurations, it does so to identify library versions and tokens. It does not access sensitive system paths or credentials, nor does it perform network operations to send data externally.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 24, 2026, 01:37 PM