cheerio
Pass
Audited by Gen Agent Trust Hub on Sep 9, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted HTML content from external websites, creating a surface for potential indirect prompt injection attacks. Since scraping is the skill's primary function, this risk is inherent to its operation.\n
- Ingestion points:
scrape_site.js(Step 3) ingests data from external URLs via thefetchAPI.\n - Boundary markers: None present in the provided instructions to delimit external content from agent instructions.\n
- Capability inventory: Network access (
fetch), HTML parsing (cheerio), and console logging.\n - Sanitization: The skill performs basic text trimming but lacks specific sanitization for potential injection payloads within the HTML.\n- [EXTERNAL_DOWNLOADS]: The skill installs the
cheeriolibrary from the official NPM registry, which is a standard and trusted package for Node.js development.\n- [COMMAND_EXECUTION]: The instructions include standardnpm installcommands for managing dependencies, which is expected behavior for the described use case.
Audit Metadata