constructive-gotchas

Warn

Audited by Gen Agent Trust Hub on Aug 27, 2026

Risk Level: MEDIUMCOMMAND_EXECUTIONPRIVILEGE_ESCALATION
Full Analysis
  • [COMMAND_EXECUTION]: The skill directs the agent to execute high-risk shell commands autonomously for process and file management. Evidence includes instructions to use 'kill -9' to terminate processes bound to specific ports, 'rm' to delete project files, and 'nohup' to launch background server processes without user oversight.
  • [PRIVILEGE_ESCALATION]: The skill provides instructions to bypass application security controls via direct SQL manipulation. Specifically, it contains SQL snippets to manually set 'is_verified' and 'is_approved' flags to true in the database, circumventing the platform's standard verification and membership approval workflows.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Aug 27, 2026, 01:07 PM
Security Audit — agent-trust-hub — constructive-gotchas