lightpanda-browser

Pass

Audited by Gen Agent Trust Hub on Mar 15, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill acts as a technical manual for the Lightpanda project, providing legitimate usage examples and configuration details.
  • [EXTERNAL_DOWNLOADS]: The installation section references binary downloads from the project's official GitHub repository, which is a well-known service.
  • [COMMAND_EXECUTION]: Provides standard commands for system package installation and file permission management common in developer tools.
  • [PROMPT_INJECTION]: The skill describes web scraping capabilities, which inherently involve an indirect prompt injection surface when processing untrusted web content.
  • Ingestion points: Integration examples in SKILL.md using page.goto(), page.content(), and page.evaluate().
  • Boundary markers: Absent; the provided code snippets do not include logic to delimit or ignore instructions found in scraped data.
  • Capability inventory: SKILL.md documentation describes full CDP support, including navigation and JavaScript execution.
  • Sanitization: Absent; the integration examples do not demonstrate sanitization or validation of the ingested HTML content.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 15, 2026, 10:16 PM