qrspi-iterate-research-questions
Pass
Audited by Gen Agent Trust Hub on Jun 21, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: No security risks were identified. The skill is designed for documentation management and operates exclusively within a restricted project directory.
- [PROMPT_INJECTION]: The skill instructions do not attempt to bypass AI safety guidelines or override core agent behavior. It emphasizes objective research and the removal of bias from questions.
- [DATA_EXFILTRATION]: There are no network-based commands (e.g., curl, wget) or attempts to access sensitive system files (e.g., credentials, SSH keys). Its operations are limited to reading and writing research files in the specified project path.
- [REMOTE_CODE_EXECUTION]: The skill does not download external packages or execute remote scripts. It uses a local editing tool for text-based modifications.
- [COMMAND_EXECUTION]: No arbitrary shell command execution or privilege escalation attempts were found. The skill relies on standard file-system interactions for project-specific files.
Audit Metadata