Neo — Web App API Discovery

• Outcome: Review
• Latest audit: May 29, 2026 at 9:36 PM UTC
• Version:
  2.1.0

Overview

Neo is a powerful browser automation skill that can use real Chrome sessions and cookies to call website APIs, post content, and manage login state, so it needs review before installation.

Install only if you trust the @4ier/neo CLI with the browser profile you use. Prefer a separate Chrome profile with only the needed accounts, avoid sensitive sites, and require explicit approval before posting, API writes, cookie export/import/clear, request replay, or running page JavaScript.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Trigger Abuse: Overly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
• 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

Findings (5)

Vague Triggers

Medium

• Confidence: 87% confidence
• Finding: The skill’s invocation language is extremely broad and covers routine browsing, scraping, API access, and social-media actions, which can cause the agent to select this powerful skill for many ordinary requests. In context, that increases the chance of unnecessary use of browser automation and authenticated web actions when a less-privileged tool would suffice.

Missing User Warnings

High

• Confidence: 97% confidence
• Finding: The cookie-management section documents listing, exporting, importing, and clearing browser cookies without any warning that these artifacts may contain live authenticated session state. This is dangerous because an agent or user could exfiltrate, persist, or replay login sessions, enabling account takeover or unauthorized access across websites.

Missing User Warnings

High

• Confidence: 98% confidence
• Finding: The skill explicitly states that API calls can use browser-derived authentication automatically, but it does not warn that requests may execute as the signed-in user and transmit authenticated data. In this skill’s context, that makes arbitrary site API interaction much more dangerous because the agent can perform state-changing operations against real accounts with little friction or visibility.

Ssd 3

High

• Confidence: 99% confidence
• Finding: These instructions operationalize export and reuse of browser cookies in plain language, directly enabling persistence and transfer of authenticated session material. Because cookies often function as bearer credentials, this can lead to credential theft, replay, cross-device session cloning, and unauthorized account access.

Ssd 3

Medium

• Confidence: 91% confidence
• Finding: The profile-management guidance includes discovery of Chrome profiles and associated emails, which exposes private account metadata unrelated to many requested tasks. In a browser-automation skill, this increases privacy risk and can help an agent choose or target the most sensitive signed-in profile without clear user approval.

VirusTotal

65/65 vendors flagged this skill as clean.