skills/openai/plugins/agent-browser/Gen Agent Trust Hub

agent-browser

Pass

Audited by Gen Agent Trust Hub on May 9, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • Browser Interaction and External Content Ingestion: The skill automates web navigation and interaction, which inherently involves processing content from external websites. This introduces a surface for indirect prompt injection, where instructions embedded in a web page could influence the agent's behavior. Standard safety protocols for handling untrusted data are recommended.
  • Dynamic JavaScript Execution: The eval command enables the execution of JavaScript within the browser's sandbox. This is a powerful feature used for complex data extraction and DOM manipulation, common in advanced web automation workflows.
  • Session State Management: The ability to save and load browser states (e.g., state save auth.json) allows the agent to maintain authenticated sessions across tasks. As these state files may contain sensitive session information like cookies, they should be stored securely and managed with appropriate file system permissions.
Audit Metadata
Risk Level
SAFE
Analyzed
May 9, 2026, 03:11 PM