alibaba-product-research
Pass
Audited by Gen Agent Trust Hub on Apr 7, 2026
Risk Level: SAFE
Full Analysis
- [COMMAND_EXECUTION]: The skill uses Playwright MCP tools (
browser_navigate,browser_type,browser_click,browser_evaluate) to interact with the Alibaba.com website. These interactions are strictly aligned with the skill's stated purpose of e-commerce product research. - [DYNAMIC_EXECUTION]: Statically defined JavaScript code is executed within the browser context via
browser_evaluateto extract specific DOM elements like product titles, URLs, and price tags. This implementation follows standard scraping patterns and does not involve the execution of untrusted or dynamically generated external scripts. - [PROMPT_INJECTION]: The instructions follow a structured workflow for data analysis without attempting to bypass safety filters or override system-level agent instructions. No adversarial role-play or instruction-bypass markers were identified.
- [DATA_EXFILTRATION]: While the skill extracts data from a remote website, the targets are public product listings on a well-known service (Alibaba.com). No access to sensitive local files, credentials, or environment variables is requested, and data is only processed for the generated report.
Audit Metadata