browser-use

Pass

Audited by Gen Agent Trust Hub on Mar 29, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill requires the agent to execute shell commands in the terminal to set up virtual environments, install dependencies, and run Python/TypeScript scripts for browser automation.
  • [EXTERNAL_DOWNLOADS]: Instructions include downloading the 'browser-use-sdk' and 'browser-use' packages from official Python package registries, as well as using 'uvx' to install the Chromium browser binary.
  • [REMOTE_CODE_EXECUTION]: The skill's primary function is to dynamically generate Python code based on user natural language and execute that code in the local environment.
  • [DATA_EXFILTRATION]: While not performing malicious exfiltration, the skill handles sensitive credentials including 'BROWSER_USE_API_KEY', 'OPENAI_API_KEY', and others, advising the use of environment variables or .env files for management.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 29, 2026, 08:52 AM