creativeclaw-minimax-h3-max
Pass
Audited by Gen Agent Trust Hub on Sep 12, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill provides structured guidance for professional video generation workflows. All referenced tools and external resources, including the MCP tool definition in the agent configuration, belong to the verified vendor infrastructure of 'creativeclawco'.
- [INDIRECT_PROMPT_INJECTION]: The skill establishes an attack surface by ingesting external media URLs (images, videos, audio) and user-supplied prompts. It mitigates accidental execution by instructing the use of explicit boundary markers (e.g., 'Image 1', 'Video 1') and a structured prompt format to separate references from instructions. This surface is inherent to the skill's primary purpose and is managed by standard platform guardrails.
- [EXTERNAL_DOWNLOADS]: The agent configuration in
agents/openai.yamlspecifies a dependency on an external Model Context Protocol (MCP) service athttps://app.creativeclaw.co/mcp/chatgpt. This resource is owned by the vendor and is required for the intended integration with the Creative Claw media platform.
Audit Metadata