web-perf

Pass

Audited by Gen Agent Trust Hub on Sep 18, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill provides instructions for setting up the chrome-devtools-mcp package using npx. This targets the official Chrome DevTools project resources, which is a well-known and trusted service.
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to audit external websites, creating a surface for indirect prompt injection where a target site could contain instructions intended to influence the agent.
  • Ingestion points: The skill ingests untrusted data from external websites through tools such as navigate_page, list_network_requests, and take_snapshot as described in Phases 1 through 4 of the workflow in SKILL.md.
  • Boundary markers: The instructions do not define explicit delimiters or warnings to isolate external data retrieved from target websites from the agent's core instructions.
  • Capability inventory: The skill utilizes browser interaction capabilities, including page navigation, performance trace recording, network inspection, and accessibility tree snapshots.
  • Sanitization: There are no specific sanitization or filtering steps defined for processing data retrieved from the target page's DOM or network response headers.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 18, 2026, 08:51 AM