ClawHub
Back to skill

Security audit

Musallat Bot

Security checks across malware telemetry and agentic risk

Overview

This is a small Gemini-powered rude chatbot, but it exposes an apparent API key and sends user prompts to an external AI service with limited privacy guidance.

Review before installing. Do not use or rely on the embedded API key; the publisher should revoke and remove it. Only run the skill with your own restricted GEMINI_API_KEY if you are comfortable sending prompts to Google/Gemini and receiving intentionally abrasive chatbot responses.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (4)

Context-Inappropriate Capability

High
Confidence
99% confidence
Finding
The skill documentation contains a raw Google API key, which is a real secret exposure unless conclusively proven revoked and nonfunctional. Even in documentation, embedded credentials can be harvested, abused for unauthorized API usage, billing impact, quota exhaustion, or used as a foothold for broader compromise depending on project restrictions.

Context-Inappropriate Capability

Medium
Confidence
88% confidence
Finding
The skill initializes an external Gemini client with an environment-sourced API key and immediately sets up outbound LLM usage without any manifest, access control, or user disclosure. In this file's context, the skill's purpose is just persona-style text generation, so pulling a secret and transmitting prompts to a third-party service expands trust boundaries and creates unnecessary data-handling risk.

Missing User Warnings

Medium
Confidence
94% confidence
Finding
User-supplied input is sent directly to an external generative AI service with no warning, consent, redaction, or filtering. This is dangerous because users may provide sensitive data assuming local handling, causing unintended disclosure to a third-party provider and possible retention in external systems.

Missing User Warnings

Low
Confidence
83% confidence
Finding
The code reads an API credential from environment variables without any user-facing disclosure or operational transparency. While environment variables are a normal secret-storage mechanism, undisclosed credential use in a user-invoked skill obscures external dependencies and can undermine informed consent and reviewability.

VirusTotal

51/51 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.