seedance-video-gen
Pass
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied scene descriptions which can serve as a vector for malicious instructions if the prompt source is untrusted.\n
- Ingestion points: The
promptparameter inscripts/generate_video.pyaccepts arbitrary text input that is passed to the generation model.\n - Boundary markers: The implementation does not utilize delimiters or specific framing instructions to prevent the model from obeying instructions embedded within the prompt.\n
- Capability inventory: The skill possesses capabilities for network communication (Replicate API) and filesystem writes (saving .mp4 files).\n
- Sanitization: There is no evidence of prompt filtering or sanitization to identify and remove potential injection patterns.\n- [COMMAND_EXECUTION]: The skill relies on a Python script (
scripts/generate_video.py) to execute its core functionality, including API interaction and local file management.\n- [EXTERNAL_DOWNLOADS]: The skill fetches generated video assets from Replicate's infrastructure. This is a standard functional requirement for an AI media generation tool using a well-known service.
Audit Metadata