ClawHub

Tasker

Security checks across malware telemetry and agentic risk

Overview

This is a broad instruction-only workflow skill that can guide real work, but it discloses that scope and requires confirmation before execution.

Install this only if you want a broad workflow skill to structure task execution. Before confirming any plan, review the proposed side effects, validation method, and files or systems involved; do not rely on silence as meaningful acceptance for important work.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • 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
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (3)

Intent-Code Divergence

Medium
Confidence
92% confidence
Finding
The auto-close text allows the skill to treat silence as acceptance, which weakens the skill's own confirmation and closure guarantees. In an agent workflow, this can cause tasks to be marked complete without explicit user approval, increasing the risk of unreviewed changes, missed failures, or premature termination of recovery/escalation flows.

Vague Triggers

High
Confidence
95% confidence
Finding
The invocation scope is extremely broad, covering development, ops, analysis, writing, planning, review, troubleshooting, and dissatisfaction handling, with lightweight matching language. This can cause the skill to be selected for many ordinary interactions, effectively overriding more specialized skills and increasing the chance that risky execution-oriented behavior is activated in the wrong context.

Vague Triggers

High
Confidence
96% confidence
Finding
The auto-discovery hints include highly common phrases such as analyze, review, summarize, check, and user dissatisfaction cues, which are likely to appear in normal conversations. Ambiguous triggers increase unintended routing and may steer benign requests into an execution workflow with confirmation, gating, and tool-use assumptions that do not fit the user's intent.

VirusTotal

63/63 vendors flagged this skill as clean.

View on VirusTotal