creativeclaw-gpt-image-2
Pass
Audited by Gen Agent Trust Hub on Sep 11, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill integrates with an external MCP tool, 'creative-claw', hosted on the vendor's infrastructure at
https://app.creativeclaw.co/mcp/chatgpt. All network operations and asset imports are routed through this specific toolset, which is consistent with the skill's stated purpose. - [SAFE]: Media handling procedures for image URLs, masks, and uploads follow established agent platform patterns (e.g., using
get_upload_urlfor HTTP PUT andimport_chatgpt_mediafor session attachments). No attempts to access sensitive local file paths (~/.ssh, ~/.aws, etc.) or hardcoded credentials were found. - [SAFE]: Instructions regarding prompt structure and image editing represent standard engineering techniques (e.g., identity preservation, typography control) and do not contain prompt injection markers or attempts to bypass safety filters.
- [SAFE]: There is no evidence of code obfuscation (Base64, zero-width characters), persistence mechanisms (cron, startup scripts), or privilege escalation (sudo).
Audit Metadata