ClawHub

Grill Me

Security checks across malware telemetry and agentic risk

Overview

The only negative signal is broad auto-activation language, with no evidence of hidden code, credential use, persistence, exfiltration, or destructive behavior.

Reasonable to install if you want this workflow helper, but be aware it may activate on generic phrases like 'go' or '执行'. Prefer explicit invocation when using it, and review the skill text for any automatic chaining or confirmation steps before relying on it in important workflows.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (3)

Vague Triggers

High
Confidence
95% confidence
Finding
The skill auto-activates on very broad phrases like '开始吧', 'go', and '执行', which are common in normal conversation and can easily be triggered unintentionally. This can cause the agent to enter a confirmation workflow the user did not explicitly request, creating prompt-routing confusion and increasing the chance of misexecution or denial of intended behavior.

Vague Triggers

Medium
Confidence
89% confidence
Finding
Conditions such as 'multiple modules/team collaboration', 'clear deadline or quality requirements', and 'technical approach unclear' are subjective and lack enforceable boundaries. Ambiguous activation logic makes the skill prone to over-triggering across many ordinary tasks, which can hijack workflow selection and reduce predictability of agent behavior.

Vague Triggers

High
Confidence
94% confidence
Finding
The trigger-command section repeats the same broad auto-trigger behavior for generic words like '开始', 'go', and '执行', reinforcing unsafe activation design. Repetition in both the main trigger section and the command section increases the likelihood that the skill will activate in routine dialogue and chain into downstream skills without sufficiently clear user intent.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal