mobile-viewport-audit

Pass

Audited by Gen Agent Trust Hub on Jul 6, 2026

Risk Level: SAFE
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface as it ingests untrusted data from external websites' DOM (Document Object Model) via the mcp__playwright__browser_evaluate function in SKILL.md.
  • Ingestion points: Data enters the context from any URL the user provides for auditing.
  • Boundary markers: None used when passing the extracted DOM data back to the agent.
  • Capability inventory: Uses Playwright MCP tools for browser navigation, resizing, and JavaScript evaluation.
  • Sanitization: The skill performs basic sanitization by truncating extracted text content using .slice(0, 30) to limit the amount of untrusted text processed by the model.
  • [EXTERNAL_DOWNLOADS]: The skill navigates to external URLs provided by the user and references the author's domain (helpmetest.com) for additional services. These are standard features for a web-based auditing tool and do not involve downloading or executing untrusted binary or script files on the host system.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 6, 2026, 09:54 AM
Security Audit — agent-trust-hub — mobile-viewport-audit