hotseat
Pass
Audited by Gen Agent Trust Hub on May 17, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill instructions are benign and focused on improving user plans through a research-driven interrogation process. No malicious code, persistence mechanisms, or privilege escalation patterns were found.
- [PROMPT_INJECTION]: The skill does not attempt to bypass safety guidelines or override system instructions. Its structure is designed to guide agent behavior within the scope of a design review.
- [DATA_EXFILTRATION]: While the skill encourages reading the local codebase and performing web research, it lacks instructions to send sensitive data to external endpoints or collect credentials.
- [COMMAND_EXECUTION]: The skill does not contain executable code, shell scripts, or instructions that would lead to unauthorized command execution. It relies on standard environment tools for research purposes.
Audit Metadata