product-detail-maker
Pass
Audited by Gen Agent Trust Hub on Jul 28, 2026
Risk Level: SAFECOMMAND_EXECUTIONDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill includes a local Node.js script (
scripts/validate-product-detail-maker-skill.mjs) and instructs the agent to execute it for validation purposes. - [COMMAND_EXECUTION]: It references and directs the execution of scripts from an external skill directory (
skills/image-maker/scripts/archive-generated-images.mjs) for asset management tasks. - [DATA_EXFILTRATION]: The skill implements a research workflow using the Chrome DevTools Protocol (CDP) to connect to local debugging endpoints and external URLs to extract layout and copy information.
- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection due to its research phase which ingests untrusted content from user-provided URLs. Ingestion points: Rendered page data captured via browser automation (browser-link-research.ko.md). Boundary markers: None identified. Capability inventory: Figma MCP integration, local file writing, and script execution. Sanitization: None identified.
Audit Metadata