landing-experience-checker
Pass
Audited by Gen Agent Trust Hub on Jul 6, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill retrieves landing page content from user-supplied URLs using a scraping tool.
- [COMMAND_EXECUTION]: It invokes a localized Python script (
firecrawl.py) via the command line to process and render the target landing pages. - [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze untrusted external data (scraped web content and CSV exports). It mitigates this risk by including a mandatory instruction for the agent to treat these sources as untrusted input and to ignore any embedded commands or instructions.
Audit Metadata