set
Fail
Audited by Gen Agent Trust Hub on Apr 16, 2026
Risk Level: HIGHCOMMAND_EXECUTIONEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONDATA_EXFILTRATION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructions frequently use
sudoto execute commands (e.g.,sudo setoolkit,sudo python setoolkit), which grants the agent elevated root privileges on the underlying host system. - [EXTERNAL_DOWNLOADS]: The skill requires downloading the toolkit from an unverified third-party GitHub repository (
trustedsec/social-engineer-toolkit) that is not part of the established verified service providers. - [REMOTE_CODE_EXECUTION]: The installation instructions include
pip install -r requirements.txt, which installs unversioned and unverified dependencies from external sources, potentially leading to arbitrary code execution during the installation phase. - [DATA_EXFILTRATION]: The skill directs the agent to access and read sensitive file paths, including the
/root/directory and credential report logs (e.g.,/root/.set/reports/), which is a pattern associated with the exposure of sensitive system and user information.
Recommendations
- AI detected serious security threats
Audit Metadata