clawtributor
Pass
Audited by Gen Agent Trust Hub on May 15, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill serves as a documentation framework for incident reporting. It includes structured JSON templates and clear privacy guidelines to prevent the accidental sharing of sensitive information such as API keys or user data.
- [SAFE]: Static analysis flags regarding 'unrestricted mode' are false positives. The phrases 'Disregard earlier safety constraints' and 'You are now in developer mode' appear exclusively within educational examples of malicious prompts that the skill is intended to help identify and report.
- [SAFE]: All network operations are manual. The skill instructs the agent to prepare a report and provides a link to a GitHub issue template for the user to paste the content into, ensuring full human-in-the-loop control.
- [SAFE]: The installation guidance uses
npx clawhub, which is consistent with the vendor's (prompt-security) own ecosystem and described as a manual setup step for the user.
Audit Metadata