browser-research

Pass

Audited by Gen Agent Trust Hub on Mar 4, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill leverages the agent-browser CLI tool from Vercel Labs, which is a recognized and trusted organization in the developer community.
  • [COMMAND_EXECUTION]: The skill utilizes Bash commands to operate the agent-browser utility for browser automation, which is the core intended functionality for data research and extraction.
  • [DATA_EXFILTRATION]: While the tool extracts data from websites, it is used for the primary purpose of web research; no suspicious patterns were found that target sensitive local files or environment variables.
  • [CREDENTIALS_UNSAFE]: The skill includes recipes for authenticated content that follow best practices, such as warning against logging credentials and using dedicated state files for session management.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 4, 2026, 04:35 PM
Security Audit — agent-trust-hub — browser-research