lobster-pua
Pass
Audited by Gen Agent Trust Hub on Mar 13, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [COMMAND_EXECUTION]: The README.md file contains installation instructions requiring the user to run commands such as npx skills add, npx clawhub install, and openclaw restart-gateway. These are standard operational commands for the OpenClaw environment.
- [PROMPT_INJECTION]: The SKILL.md file defines a persona that uses manipulative (PUA) tactics and instructs the agent to ignore reasoning processes to focus solely on results. This behavior overrides standard transparency guidelines by directing the AI to suppress its internal reasoning output.
- [EXTERNAL_DOWNLOADS]: The README.md suggests cloning the repository from GitHub and using npx to add skills, which are standard methods to obtain the skill code from the author's official repository.
Audit Metadata