elevenlabs
Pass
Audited by Gen Agent Trust Hub on Jul 2, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The
/generate-voiceovertoolkit command processes content from an external file namedVOICEOVER-SCRIPT.md. This ingestion of untrusted data creates a surface for indirect prompt injection, where malicious instructions embedded in the script could potentially influence the agent's behavior. - Ingestion points: Data is ingested from
VOICEOVER-SCRIPT.mdduring the execution of the/generate-voiceovercommand. - Boundary markers: The instructions do not define explicit boundary markers or delimiters to separate the script content from the agent's internal instructions.
- Capability inventory: The skill has capabilities including writing to the file system (saving .mp3, .json manifest, and .json project files) and performing network requests to the ElevenLabs API.
- Sanitization: There is no evidence of content sanitization or validation performed on the script data before it is processed by the agent.
Audit Metadata