webperf-core-web-vitals

Pass

Audited by Gen Agent Trust Hub on Sep 18, 2026

Risk Level: SAFEOBFUSCATIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [OBFUSCATION]: The JavaScript files in the scripts/ directory are minified. This obscures the logic of the performance auditing tools and makes it difficult for users to inspect the code for potential side effects or malicious behavior before executing it within the browser context.
  • [INDIRECT_PROMPT_INJECTION]: The skill extracts information from the Document Object Model (DOM) and the Performance API of the pages being audited. An attacker-controlled website could manipulate element attributes or performance entries to deliver malicious instructions to the agent.
  • Ingestion points: Scripts such as INP.js, LCP.js, and LCP-Trail.js extract string data from element.id, element.className, and performance entry URLs.
  • Boundary markers: The skill does not implement explicit boundary markers or warnings to the agent to ignore instructions embedded in the audit results.
  • Capability inventory: The skill utilizes the mcp__chrome-devtools__evaluate_script tool, which allows the execution of arbitrary JavaScript in the user's browser.
  • Sanitization: The scripts perform basic string manipulation on class names and URLs but do not fully sanitize or escape the content to prevent it from being interpreted as instructions by the downstream LLM.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 18, 2026, 09:00 AM
Security Audit — agent-trust-hub — webperf-core-web-vitals