amazon-store-research
Pass
Audited by Gen Agent Trust Hub on Sep 29, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill automates the collection and analysis of public information from Amazon store pages. It uses standard Playwright MCP tools to navigate and evaluate web content.
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from external Amazon store pages (product titles, categories, prices).
- Ingestion points: Data is extracted from the DOM of Amazon store pages using
mcp__playwright__browser_evaluateinSKILL.md(Steps 1.2 to 1.5). - Boundary markers: There are no explicit delimiters or instructions provided to sub-agents to ignore embedded prompts in the scraped data.
- Capability inventory: The skill utilizes Playwright for browsing and file system access to write markdown reports and JSON data files.
- Sanitization: The skill performs basic data cleaning such as regex matching for ASINs and string truncation for titles, but does not sanitize the text for potential prompt injection patterns.
- [COMMAND_EXECUTION]: The skill includes JavaScript snippets for data extraction. These scripts are static, perform benign DOM traversal, and are executed within the controlled environment of the Playwright browser context.
Audit Metadata