ai-video-generation

Pass

Audited by Gen Agent Trust Hub on May 15, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes the runcomfy CLI for its primary operations. The allowed-tools frontmatter restricts the execution environment to this specific tool, minimizing the risk of arbitrary command execution.
  • [EXTERNAL_DOWNLOADS]: Instructions include installing the @runcomfy/cli package via standard package managers (NPM/NPX). This tool is maintained by the skill's vendor and is essential for the functionality described.
  • [CREDENTIALS_UNSAFE]: The documentation provides clear guidance on secure token management, advising the use of environment variables or local configuration files with restricted permissions (0600). It explicitly warns against leaking tokens in prompts or logs.
  • [PROMPT_INJECTION]: The skill includes a dedicated security section that identifies the risk of indirect prompt injection via user-provided media assets (images, audio, video). It provides specific mitigation strategies for the agent to follow when processing these inputs.
Audit Metadata
Risk Level
SAFE
Analyzed
May 15, 2026, 12:18 PM