ClawHub
Back to skill

Security audit

Collab Edit

Security checks across malware telemetry and agentic risk

Overview

This skill is coherent for real-time collaboration, but it documents exposing an unauthenticated browser IDE and local services to the public internet with insufficient warning.

Review carefully before installing. Use this only for deliberate collaboration sessions, enable authentication on any editor or whiteboard, avoid tunneling sensitive repos or credentials, do not expose SSH/databases through this skill, prefer pip/uv installation, and stop both the tunnel and collaboration server immediately when finished.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (5)

Context-Inappropriate Capability

Medium
Confidence
77% confidence
Finding
The CLI reference broadens the skill from browser-based collaborative editing to generic TCP tunneling, including examples like SSH and databases. That capability can be repurposed to expose sensitive internal services and increases the chance an agent or user will use the skill for unintended remote access beyond the stated collaboration use case.

Context-Inappropriate Capability

Medium
Confidence
96% confidence
Finding
The skill explicitly recommends launching code-server bound to 0.0.0.0 with '--auth none', then exposing it through a public tunnel. This creates an internet-reachable development environment with no access control, which can allow unauthorized users to view, edit, or execute code and access local files through the IDE.

Missing User Warnings

Medium
Confidence
90% confidence
Finding
The skill instructs users to expose a local collaboration service to the internet and share the public link, but it does not give a clear upfront warning that anyone with the URL may gain access to live content and possibly editing capability. In the context of collaborative tools, this can expose sensitive code, documents, comments, and potentially connected local resources depending on the tool.

Missing User Warnings

High
Confidence
98% confidence
Finding
Running code-server with authentication disabled and then exposing it via a tunnel creates a publicly accessible IDE without any explicit warning about the severity of that risk. In practice, this can permit arbitrary third parties who obtain the link to browse project files, modify source code, and potentially run commands in the host environment.

External Script Fetching

High
Category
Supply Chain
Content
## CLI Reference

The `aitun` command (installed via `pip install aitun`, or alternatively `curl -fsSL https://aitun.cc/install.sh | bash` / `irm https://aitun.cc/install.ps1 | iex` on Windows) accepts these flags:

| Flag | Description |
|---|---|
Confidence
88% confidence
Finding
curl -fsSL https://aitun.cc/install.sh | bash

VirusTotal

VirusTotal findings are pending for this skill version.

View on VirusTotal

Static analysis

No suspicious patterns detected.