script-to-voiceover
Pass
Audited by Gen Agent Trust Hub on Sep 26, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied plain-text scripts to generate narration, which creates an attack surface where instructions embedded within the text could potentially influence agent behavior.\n
- Ingestion points: The skill explicitly requests 'The script
- plain text, in full' as a user input in SKILL.md.\n
- Boundary markers: The workflow does not specify the use of delimiters or 'ignore' instructions when handling the input script.\n
- Capability inventory: The skill uses
clueso-mcpto browse voice engines and perform audio generation (Workflow Step 7).\n - Sanitization: No mention of input validation or escaping is provided for the script content.\n- [EXTERNAL_DOWNLOADS]: The skill provides configuration instructions to connect a remote service.\n
- Evidence: The documentation provides setup steps for various assistants using
https://connect.clueso.io/mcpas the transport address.
Audit Metadata