shell
Fail
Audited by Gen Agent Trust Hub on Sep 3, 2026
Risk Level: HIGHCOMMAND_EXECUTIONPRIVILEGE_ESCALATIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill enables the execution of arbitrary shell commands provided by the user. It explicitly instructs the agent to run user-supplied text immediately using a terminal tool without any validation, filtering, or sanitization.
- [PRIVILEGE_ESCALATION]: By allowing direct access to the shell, the skill permits any action authorized for the agent's process, which may include accessing credentials, environment variables, or private files.
- [INDIRECT_PROMPT_INJECTION]: The skill defines a high-privilege trigger (/shell) that can be exploited if the agent processes untrusted data. Ingestion points: Any text (files, web content, etc.) that the agent reads containing the /shell trigger. Boundary markers: None. The instructions bypass typical safety reviews. Capability inventory: Full terminal shell access. Sanitization: Explicitly disabled; the agent is told not to rewrite or review the command.
Recommendations
- AI detected serious security threats
Audit Metadata