ClawHub
Back to skill

Security audit

AppDrag

Security checks across malware telemetry and agentic risk

Overview

The skill is AppDrag-related, but it gives agents broad AppDrag backend function execution while describing itself mainly as search/read access.

Review before installing. Only use this with AppDrag projects where you are comfortable letting the agent run backend API functions through your connected OOMOL account. Confirm the exact AppDrag function, HTTP method, and payload before any call that could change data, and prefer a verified or pinned CLI installation method instead of direct pipe-to-shell setup.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Findings (3)

Description-Behavior Mismatch

High
Confidence
97% confidence
Finding
The skill metadata and description frame this as a read/search skill, but the only documented action is `execute_function`, which can invoke arbitrary AppDrag backend functions and may perform writes or destructive operations depending on the target function and HTTP method. This mismatch can mislead an agent into invoking the skill without appropriate confirmation or safety checks, increasing the chance of unintended state changes or abuse.

Vague Triggers

Medium
Confidence
85% confidence
Finding
The trigger text says to use this skill for ANY AppDrag request, which is overly broad and can cause the agent to route unrelated or higher-risk AppDrag tasks through a powerful function-execution interface by default. In context, this is more dangerous because the skill is not limited to passive reads despite the description suggesting that it is.

External Script Fetching

High
Category
Supply Chain
Content
- **`oo: command not found`** — install the oo CLI (other platforms: <https://cli.oomol.com/install-guide.md>):

  ```bash
  curl -fsSL https://cli.oomol.com/install.sh | bash    # macOS / Linux
  ```

  ```powershell
Confidence
98% confidence
Finding
The skill instructs users to install the CLI via `curl ... | bash`, which executes a remote script directly from the network without prior verification. If the install endpoint, transport path, or hosting environment is compromised, this can lead to arbitrary code execution on the user's machine.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.