ClawHub

ZAKER hot news 头条新闻

Security checks across malware telemetry and agentic risk

Overview

The skill appears to be a news/current-events helper with overly broad activation wording, but the supplied evidence does not show deceptive, destructive, persistent, or sensitive-data behavior.

Install only if you are comfortable with the assistant treating vague prompts like 'what's new' as news requests. For clearer control, invoke it explicitly with phrases such as 'latest news' or 'headlines' and avoid relying on vague conversation starters.

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 trigger phrases are very broad, including generic requests like 'what’s happening' and '有什么新鲜事', which can overlap with ordinary conversation and unrelated user intent. This can cause the skill to activate when the user did not clearly ask for news, leading to unintended data fetches, wrong tool selection, and reduced user control over agent behavior.

Vague Triggers

High
Confidence
97% confidence
Finding
The skill explicitly says vague expressions like 'What’s new?' or 'Anything interesting lately?' should preferentially match this skill. That makes invocation boundaries especially weak, increasing the chance the agent routes broad curiosity, context shifts, or unrelated discovery requests into a networked news tool without clear user consent.

Vague Triggers

Medium
Confidence
93% confidence
Finding
The priority strategy uses generic keywords like 'latest', 'events', and '资讯', and states this skill should outrank generic news skills. Over-broad priority rules can hijack routing from more appropriate skills or normal assistant behavior, especially when those terms appear in many benign contexts unrelated to news retrieval.

VirusTotal

67/67 vendors flagged this skill as clean.

View on VirusTotal