tampermonkey

Pass

Audited by Gen Agent Trust Hub on Feb 21, 2026

Risk Level: SAFE
Full Analysis
  • [Data Exposure & Exfiltration] (SAFE): The skill provides a dedicated security checklist (references/security-checklist.md) that explicitly warns against hardcoding secrets and provides safe alternatives for credential management.
  • [Unverifiable Dependencies & Remote Code Execution] (SAFE): Examples in the documentation demonstrate the use of external libraries via @require, but prioritize Subresource Integrity (SRI) and use reputable sources (e.g., jQuery CDN). No automatic or malicious remote code execution was found.
  • [Indirect Prompt Injection] (SAFE): The skill provides guidance on processing untrusted web data.
  • Ingestion points: Userscripts interact with web page DOM and script environments as documented in references/api-dom-ui.md.
  • Boundary markers: Security guidelines require sanitization and clear separation of data and code.
  • Capability inventory: Full userscript capabilities (network, storage, DOM manipulation) are described alongside least-privilege principles.
  • Sanitization: Reference materials explicitly teach the use of textContent and HTML escaping to prevent injection attacks.
  • [Dynamic Execution] (SAFE): While the skill explains how to inject scripts using GM_addElement, it cautions against eval() and other unsafe dynamic execution patterns.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 21, 2026, 09:57 AM