skills/ninehills/skills/agent-browser/Gen Agent Trust Hub

agent-browser

Pass

Audited by Gen Agent Trust Hub on Mar 28, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: Installs the 'agent-browser' NPM package and additional browser binaries during setup.
  • [COMMAND_EXECUTION]: Relies on shell commands to control browser sessions and page interactions.
  • [DATA_EXFILTRATION]: Provides mechanisms to save and load sensitive browser state (cookies and local storage) to 'auth.json' and capture screenshots of page data.
  • [PROMPT_INJECTION]: Vulnerable to indirect prompt injection as it ingests and processes untrusted content from web pages.
  • Ingestion points: Page snapshots and text extraction.
  • Boundary markers: None identified.
  • Capability inventory: Includes form submission, state management, and file operations.
  • Sanitization: No evidence of content filtering or instruction scrubbing is implemented for page data.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 28, 2026, 12:00 AM