video-outpainting

Pass

Audited by Gen Agent Trust Hub on Jul 13, 2026

Risk Level: SAFE
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill references the installation of the @runcomfy/cli package from the NPM registry. This package is the official command-line tool for the RunComfy service.
  • [CREDENTIALS_UNSAFE]: The skill identifies the storage path for API tokens at ~/.config/runcomfy/token.json. This is documented as the standard configuration location for the vendor's CLI tool and is managed with restricted file permissions.
  • [PROMPT_INJECTION]: The skill notes the risk of indirect prompt injection from processing external video URLs. It provides remediation guidance, such as only processing user-verified URLs and monitoring for unexpected model behavior.
  • [COMMAND_EXECUTION]: The skill utilizes the runcomfy command-line interface to interact with model endpoints. The implementation includes documentation on how the CLI manages input boundaries to prevent shell injection during execution.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 13, 2026, 04:10 AM
Security Audit — agent-trust-hub — video-outpainting