subagent-driven-development
Fail
Audited by Gen Agent Trust Hub on Mar 21, 2026
Risk Level: HIGHPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [PROMPT_INJECTION]: The skill includes instructions that attempt to override the agent's safety protocols. Specifically, the 'Launch Pattern' section provides commands that include the '--permission-mode bypassPermissions' flag, which is a direct attempt to circumvent security controls.
- [COMMAND_EXECUTION]: The skill promotes the use of the 'exec' tool to run background commands with security filters disabled. This allows for the execution of arbitrary tasks without adhering to the standard permission model or safety guardrails of the host environment.
Recommendations
- AI detected serious security threats
Audit Metadata