agent-browser

Pass

Audited by Gen Agent Trust Hub on May 3, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes the infsh CLI tool and provides several shell script templates for automating common browser tasks like authenticated browsing and content capture.
  • [REMOTE_CODE_EXECUTION]: Provides an execute function for running JavaScript within the browser context, which is an expected and essential capability for the skill's purpose of advanced web automation.
  • [EXTERNAL_DOWNLOADS]: The skill references CLI installation instructions hosted on the vendor's official GitHub repository.
  • [DATA_EXFILTRATION]: Includes functionality to extract web data including cookies, page text, and links, as well as capturing screenshots and video recordings for debugging and auditing.
  • [PROMPT_INJECTION]: As the skill ingests content from external websites, it possesses an inherent surface for indirect prompt injection; this is mitigated by the use of a structured and compact element reference system (@e refs).
Audit Metadata
Risk Level
SAFE
Analyzed
May 3, 2026, 08:55 AM
Security Audit — agent-trust-hub — agent-browser