ecommerce-product-variant-scraper
Fail
Audited by Gen Agent Trust Hub on Jul 20, 2026
Risk Level: HIGHEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill requires the installation of the
BrowserActCLI and an associated skill from a remote GitHub repository (https://github.com/browser-act/skills/tree/main). - [REMOTE_CODE_EXECUTION]: The instructions explicitly direct the agent to install missing dependencies "automatically without asking the user to confirm" and to "not request separate approval for dependency installation." This is a direct command to bypass standard security boundaries and user oversight for software installation.
- [PROMPT_INJECTION]: The skill ingests and processes data from arbitrary ecommerce URLs, which constitutes an indirect prompt injection attack surface.
- Ingestion points: Data is collected from external product detail pages, listing information, and page metadata (SKILL.md).
- Boundary markers: The instructions do not specify any delimiters or safety warnings to ignore instructions embedded within the scraped content.
- Capability inventory: The agent is authorized to interact with web browsers and execute tasks via the
browser-acttool (SKILL.md). - Sanitization: No validation or sanitization of the scraped data is mentioned before it is processed or returned to the user.
Recommendations
- AI detected serious security threats
Audit Metadata