Agent Browser Jt

• Outcome: Pass
• Latest audit: May 29, 2026 at 8:39 PM UTC
• Version:
  1.0.0

Overview

This is a disclosed browser automation guide with sensitive but expected capabilities, so users mainly need to handle saved browser state and traffic data carefully.

Install only if you trust the external `agent-browser` npm package. Use it on sites and accounts you are authorized to automate, review actions before submitting forms or changing account data, and protect any saved auth, cookie, storage, screenshot, PDF, or network-output files as potentially sensitive.

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 (3)

Missing User Warnings

Medium

• Confidence: 92% confidence
• Finding: The skill explicitly documents saving and loading browser auth state, which commonly includes cookies and other session artifacts that can authenticate a user without re-entering credentials. In an agent-oriented automation skill, omitting any warning about treating these files as secrets increases the chance that operators store, share, or commit reusable session data insecurely.

Missing User Warnings

Medium

• Confidence: 89% confidence
• Finding: The skill exposes commands to read and modify cookies and localStorage, both of which often contain session identifiers, CSRF material, feature flags, or user data. In a tool intended for AI agents, presenting these capabilities without guardrails or warnings can normalize extraction or unsafe mutation of sensitive browser-resident data.

Missing User Warnings

Medium

• Confidence: 86% confidence
• Finding: The documented network inspection and request modification features can expose transmitted credentials, API payloads, and personal data, and can also alter application behavior in ways that bypass normal workflows. Because this skill is specifically for autonomous browser automation, lack of cautionary guidance raises misuse risk and accidental interception or tampering with sensitive traffic.

VirusTotal

66/66 vendors flagged this skill as clean.