seedance-comic-drama-video
Pass
Audited by Gen Agent Trust Hub on Sep 25, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill retrieves generated media content from the service's API endpoints (ai-hive.iclip.cn) and saves it to the local file system.
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-provided prompts and service responses. 1. Ingestion points: User-supplied prompt argument and API task status responses in scripts/videogen.py. 2. Boundary markers: None explicitly defined in the CLI script logic. 3. Capability inventory: The skill uses the requests library for network communication and performs file system read/write operations for media processing. 4. Sanitization: API request bodies are constructed using standard JSON serialization.
- [SAFE]: The skill performs expected data transfers of user-selected media files to the service's infrastructure (ai-hive.iclip.cn) for video processing. It also securely manages an API key in ~/.ai-hive/config.json by automatically applying restricted file permissions (mode 0600), adhering to security best practices for local secret management.
Audit Metadata