Agent Browser Tianjin

• Outcome: Review
• Latest audit: May 29, 2026 at 9:25 PM UTC
• Version:
  0.2.0

Overview

This is a legitimate browser automation skill, but it gives agents broad control over authenticated web sessions and saved session files without enough safety guidance.

Install only if you need agent-driven browser automation and trust the external agent-browser package. Use isolated browser sessions, avoid high-value logged-in accounts, treat saved state files and captured media as secrets, delete them after use, and require explicit confirmation before submissions, uploads, purchases, posts, deletions, or account changes.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access
• Supply Chain: Unpinned Dependencies, External Script Fetching, Obfuscated Code
• Excessive Agency: Unrestricted Tool Access, Autonomous Decision Making, Scope Creep

Findings (1)

Missing User Warnings

Medium

• Confidence: 92% confidence
• Finding: The skill explicitly documents saving and loading browser session state, which commonly includes cookies and authentication tokens, but provides no warning that these artifacts are sensitive secrets. In an agent context, this can lead to credential/session leakage through insecure storage, accidental reuse across tasks, or exfiltration of persistent authenticated state.

VirusTotal

59/59 vendors flagged this skill as clean.