video-translate
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFE
Full Analysis
- [CREDENTIALS_UNSAFE]: The skill correctly manages authentication by using the HEYGEN_API_KEY environment variable, avoiding hardcoded secrets and following standard security practices.
- [DATA_EXFILTRATION]: Network communication is restricted to the official HeyGen API domain (api.heygen.com) to perform the documented translation tasks. No sensitive local files are accessed or transmitted.
- [INDIRECT_PROMPT_INJECTION]: The skill processes external data such as video URLs, titles, and custom instructions.
- Ingestion points: Fields like 'video_url', 'title', 'instruction', and 'vocabulary' in SKILL.md.
- Boundary markers: None present in the documentation snippets.
- Capability inventory: Network access via the 'mcp__heygen__*' toolset to interface with the HeyGen API.
- Sanitization: Not explicitly implemented in the provided documentation snippets.
- [COMMAND_EXECUTION]: Includes standard bash curl examples for API interaction, which are benign and intended for developer guidance.
Audit Metadata