dispatch
Pass
Audited by Gen Agent Trust Hub on Jul 15, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill generates and executes shell commands to launch new agent instances using
tmux new-window. The command string incorporates variables like$TITLE,$ROOT, and$PROMPT. While$PROMPTappears to be a predefined bootstrap instruction, the$TITLEis derived from the current 'focus', which may be influenced by untrusted context, potentially leading to command injection if not strictly sanitized. - [PROMPT_INJECTION]: The skill implements a workflow susceptible to Indirect Prompt Injection (Category 8) by relaying operator-provided content to child agents.
- Ingestion points: User-provided instructions ('what to send') are ingested and forwarded as 'message-zero' to the recipient agent.
- Boundary markers: There are no requirements for delimiters or 'ignore' instructions to wrap the forwarded content.
- Capability inventory: The skill can execute shell commands via
tmuxand provision new agent environments with full capabilities. - Sanitization: No sanitization or validation of the transferred message body is mentioned in the instructions.
Audit Metadata