Skills
skills/gooseworks-ai/goose-skills/remix-chatgpt-ad-from-sample/Gen Agent Trust Hub

remix-chatgpt-ad-from-sample

Pass

Audited by Gen Agent Trust Hub on Jul 3, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructs the agent to execute system commands including ffmpeg for video processing and npx playwright for browser-based recording of the chat interface.
  • [COMMAND_EXECUTION]: The workflow involves running a local Node.js script (node clips/record-master.js) to drive the rendering process using the create-chatgpt-mockup component.
  • [DATA_EXFILTRATION]: The skill performs file uploads to project-specific paths using get_upload_url and PUT operations. This is the intended mechanism for publishing finished ad assets to the Gooseworks platform and is scoped to the project context.
  • [PROMPT_INJECTION]: The skill processes external data from ad templates and brand research files to generate new conversation scripts. This creates a surface for indirect prompt injection if the source templates contain malicious instructions. The skill includes compliance checks to prevent source-brand leakage but lacks explicit delimiter-based sanitization for the LLM processing the data.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 3, 2026, 12:54 PM
Security Audit — agent-trust-hub — remix-chatgpt-ad-from-sample