seedance-2-0

Pass

Audited by Gen Agent Trust Hub on Oct 26, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill contains multiple links to external documentation and APIs for video generation providers including fal.ai, Replicate, and Runway. These are well-known services and the links are provided for informational purposes related to the skill's primary function.
  • [INDIRECT_PROMPT_INJECTION]: The skill defines templates for interpolating user-provided prompts into tool calls.
  • Ingestion points: PROMPT variable within Python code snippets in SKILL.md.
  • Boundary markers: No specific boundary markers or safety warnings are included in the code examples.
  • Capability inventory: The skill requests Bash, Read, and Write permissions; it uses video_selector and seedance_video tools to perform video generation.
  • Sanitization: No input sanitization or validation logic is present in the skill instructions.
  • [COMMAND_EXECUTION]: The skill requests access to the Bash tool. The documentation describes how the agent should execute internal tools via a Python registry. This usage is consistent with the skill's stated purpose of orchestrating video generation tasks.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 26, 2026, 06:31 PM