skills/inference-sh-9/skills/seedance/Gen Agent Trust Hub

seedance

Pass

Audited by Gen Agent Trust Hub on Jul 20, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes the belt command-line interface to interact with the generation API. Use of this tool is restricted via the allowed-tools frontmatter field, which limits the agent's shell access to the belt command only, following the principle of least privilege.
  • [EXTERNAL_DOWNLOADS]: The documentation references installation resources and supplementary skills hosted in the vendor's official GitHub repository (inference-sh). These references are used for standard environment configuration and do not involve untrusted third-party sources.
  • [SAFE]: No suspicious patterns such as credential harvesting, obfuscation, or unauthorized persistence were identified. The instructions and examples are consistent with the skill's stated purpose of providing a video generation interface.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 20, 2026, 04:02 PM
Security Audit — agent-trust-hub — seedance