interminai
Pass
Audited by Gen Agent Trust Hub on Sep 24, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes the
Shelltool to run theinterminaiutility, which starts and manages arbitrary CLI applications in a pseudoterminal (PTY). This allows the agent to execute any command supported by the underlying system.\n- [INDIRECT_PROMPT_INJECTION]: The skill is designed to have the agent read and interpret terminal output, which creates a surface for indirect prompt injection if the terminal displays content from untrusted files, websites, or other agents.\n - Ingestion points: Terminal output is ingested via the
interminai output,interminai status, andinterminai waitcommands (referencing filesSKILL.mdandreference.md).\n - Boundary markers: The documentation does not specify the use of delimiters or instructions to ignore embedded commands within the terminal output.\n
- Capability inventory: The agent can send arbitrary keyboard input (
interminai input), execute new processes (interminai start), and terminate processes (interminai kill).\n - Sanitization: There is no evidence of sanitization or filtering applied to the terminal output before it is processed by the agent.
Audit Metadata