profile-website-bot-detection
Pass
Audited by Gen Agent Trust Hub on Sep 16, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses the
kernelCLI to manage cloud-based browser sessions and employsjqfor processing JSON results. These are standard operations for this vendor's tools. - [EXTERNAL_DOWNLOADS]: The setup instructions include running
npm install, which fetches dependencies from the official npm registry, a well-known and trusted service. - [INDIRECT_PROMPT_INJECTION]: The skill has an attack surface for indirect prompt injection because it is designed to visit and analyze arbitrary external websites.
- Ingestion points: The script
scripts/test-kernel-bot-detection.tsuses Playwright to navigate to external URLs and scrapes data such as page titles, text, and cookies. - Boundary markers: Absent; data extracted from target websites is stored in a report and presented to the agent without explicit delimiters or instructions to ignore potential commands embedded in the scraped text.
- Capability inventory: The skill can write files to the local disk (
fs.writeFileSync), capture screenshots (page.screenshot), and manage cloud browser sessions via thekernelCLI. - Sanitization: Absent; the analysis script performs pattern matching and regex searches directly on raw content extracted from the target website's DOM and network traffic.
Audit Metadata