storyboard-to-seedance-prompt

Pass

Audited by Gen Agent Trust Hub on Apr 1, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [SAFE]: The skill performs expected file I/O operations by reading input files and writing prompt files, which is necessary for its core functional purpose.\n- [SAFE]: Interaction with the associated jimeng-skill tool for video generation is legitimate and follows the documented workflow.\n- [PROMPT_INJECTION]: The skill has an indirect prompt injection surface as it processes external text to drive downstream tool actions. This is a functional design and no malicious patterns were identified.\n
  • Ingestion points: Reads storyboard content from input text files.\n
  • Boundary markers: Absent; input text is processed without delimiters.\n
  • Capability inventory: Local file system writes and invocation of video generation tools.\n
  • Sanitization: No input validation or sanitization is performed on the storyboard content before prompt generation.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 1, 2026, 12:51 PM