ClawHub

Side Hustle

Security checks across malware telemetry and agentic risk

Overview

The skill appears intended to search for side-hustle opportunities, but its triggers and documentation are broad and polluted with unrelated learned content while delegating behavior to an external browser-search function not included in the package.

Install only if you are comfortable with a skill that may open a browser and search the web when broad money-related phrases are used. Review or obtain the referenced capabilities_clawhub.py implementation before using it in a sensitive environment, and prefer a cleaned version with narrow triggers and unrelated learned content removed.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • 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
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (5)

Description-Behavior Mismatch

Medium
Confidence
94% confidence
Finding
The skill file contains large amounts of embedded learned content unrelated to finding side-hustle opportunities, including entertainment/media references and auto-merged artifacts. This creates capability drift: users and downstream agents may invoke the skill expecting job/opportunity search behavior, but the skill context has been polluted with irrelevant material that can confuse execution, retrieval, or trigger matching.

Intent-Code Divergence

Medium
Confidence
87% confidence
Finding
The top-level description says the skill searches for side-hustle opportunities, but the rest of the file documents largely unrelated content. This mismatch can mislead users and orchestrators into authorizing browser actions under false assumptions, increasing the chance of unintended navigation or unsafe automation.

Vague Triggers

Medium
Confidence
91% confidence
Finding
The trigger phrases include broad terms such as '赚钱' and '额外收入', which are common in ordinary conversation and can cause accidental invocation. Because this skill opens a browser and performs searches, overly broad activation increases the risk of unintended external actions and privacy-impacting behavior without clear user intent.

Vague Triggers

Medium
Confidence
82% confidence
Finding
The file presents 'side hustle' as the explicit trigger, but nearby content expands activation semantics beyond that exact phrase. Ambiguous invocation rules can cause inconsistent or accidental triggering by the agent framework, especially when combined with side effects like browser automation.

Missing User Warnings

Medium
Confidence
96% confidence
Finding
The skill description does not clearly warn users that invoking it will open a browser and perform web searches. Hidden side effects reduce informed consent and can expose user context, browsing activity, or cause unexpected automation in environments where external navigation should be explicit.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal