free-groq-voice
Pass
Audited by Gen Agent Trust Hub on Mar 10, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The script
transcribe.shexecutes system commands includingcurlandjqto perform audio transcription and response parsing. This is standard behavior for the skill's stated purpose. - [EXTERNAL_DOWNLOADS]: The skill transmits audio data to
api.groq.com. This is a well-known service for AI inference, and the network activity is consistent with the skill's primary function. - [PROMPT_INJECTION]: The skill features an indirect prompt injection surface as it processes external, untrusted audio data.
- Ingestion points: Audio files provided to the
transcribe.shscript for transcription. - Boundary markers: The skill does not employ delimiters or specific instructions to isolate the transcribed text from the rest of the agent's context.
- Capability inventory: The skill utilizes
curlfor network requests and can read local files specified by the agent. - Sanitization: No sanitization or verification is performed on the transcribed output before it is returned to the agent.
Audit Metadata