11labs
Pass
Audited by Gen Agent Trust Hub on Oct 26, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process external content from URLs and local files to generate audio scripts. This creates a potential surface for indirect prompt injection if the source material contains hidden instructions.
- Ingestion points: The skill instructions explicitly direct the agent to fetch content from a URL (using WebFetch) or read from a local file path provided by the user (SKILL.md).
- Boundary markers: The instructions do not define clear delimiters or "ignore instructions" wrappers for the external content being processed.
- Capability inventory: The skill's scripts and instructions are limited to text restructuring, formatting, and audio tagging; no evidence of arbitrary code execution, file writing, or unauthorized data exfiltration was found.
- Sanitization: There are no specific instructions for the agent to sanitize or filter out potential instructions found within the ingested source text before conversion.
Audit Metadata