cc4d-01-describing-what-you-want

Pass

Audited by Gen Agent Trust Hub on Feb 16, 2026

Risk Level: LOWPROMPT_INJECTION
Full Analysis
  • [Prompt Injection] (LOW): The skill instructs the agent to update the CLAUDE.md file based on user interactions and preferences. Because this file often serves as a persistent context or instruction set for the agent, a malicious user could potentially inject instructions via conversation that would then be saved and influence the agent's behavior in future sessions.
  • [Information] (INFO): Mentions the external tool agentation.dev. This is a recommendation for the user to provide better feedback and does not involve automated data transfer or execution by the skill itself.
Audit Metadata
Risk Level
LOW
Analyzed
Feb 16, 2026, 12:22 AM