speak
Pass
Audited by Gen Agent Trust Hub on Sep 19, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill directs the agent to interact with a local text-to-speech utility by generating Bash commands. It specifically instructs the agent to chain multiple tool calls using shell operators such as '&&' and to use 'for' loops for serial execution.
- [INDIRECT_PROMPT_INJECTION]: The skill ingest user-supplied text and interpolates it into shell command arguments without explicit sanitization.
- Ingestion points: User input provided via the argument-hint or during interactive voice sessions (SKILL.md).
- Boundary markers: No delimiters or escaping instructions are provided to protect the shell execution environment from malicious content embedded in the text.
- Capability inventory: The skill leverages the 'bash' tool to execute 'say' commands (SKILL.md).
- Sanitization: There is no guidance provided to the agent regarding the sanitization or validation of external input before it is incorporated into executable command strings.
Audit Metadata