agent-browser

Pass

Audited by Gen Agent Trust Hub on Mar 3, 2026

Risk Level: SAFECOMMAND_EXECUTIONDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: Uses the agent-browser CLI to automate web navigation and interaction commands.
  • [DATA_EXFILTRATION]: Provides the ability to navigate to arbitrary URLs and retrieve page content, titles, or screenshots, which could be used to gather information from sensitive or untrusted sites.
  • [PROMPT_INJECTION]: Subject to indirect prompt injection attacks where malicious instructions on a website could influence the agent's behavior.
  • Ingestion points: Web content is ingested via snapshot, get text, and console commands.
  • Boundary markers: Lacks explicit delimiters or instructions to ignore embedded commands in the processed web data.
  • Capability inventory: Includes navigation (open), interaction (click, fill), session state management (state save/load), and file output (screenshot).
  • Sanitization: Does not show any sanitization or validation of the data retrieved from websites.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 3, 2026, 04:22 PM