seedance

Pass

Audited by Gen Agent Trust Hub on Aug 3, 2026

Risk Level: SAFE
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill references installation guidelines and supplemental resources from the official repository of the inference.sh platform. These references are used for standard configuration of the supported CLI tool.
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests user-defined prompts and media URLs to generate video content. While these are untrusted data entry points, the resulting content is processed by an external inference service, presenting no immediate risk to the local environment.
  • Ingestion points: Prompt strings and media URLs (image, video, audio) passed to the belt CLI.
  • Boundary markers: None explicitly defined.
  • Capability inventory: Invokes the belt CLI through Bash to interact with a remote API.
  • Sanitization: The skill relies on the remote platform's safety filters for content generation.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 3, 2026, 03:53 PM
Security Audit — agent-trust-hub — seedance