ClawHub

Knowledge Mapper

Security checks across malware telemetry and agentic risk

Overview

This is a simple note-mapping skill that analyzes user-provided text and does not include code, persistence, network activity, or privileged actions.

Install if you want a Chinese-language assistant for mapping relationships among notes you provide. Review inputs before sharing private notes, since the skill is meant to process whatever text you paste into the assistant.

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

Medium
Confidence
86% confidence
Finding
The trigger phrases are generic note-taking and knowledge-mapping requests, so the skill could be invoked when a user intended a broader assistant capability rather than this specific skill. This creates routing ambiguity and unintended activation risk, though the skill itself only performs local text analysis and does not request dangerous actions or sensitive operations.

Vague Triggers

Medium
Confidence
88% confidence
Finding
The description is written as a broad, natural-language invitation ('send me fragmented notes') without any clear boundary for when the skill should activate. This can cause over-triggering or ambiguous routing, leading the agent to invoke the skill in situations the user did not explicitly intend, especially because note-taking and knowledge organization are common tasks.

Natural-Language Policy Violations

Medium
Confidence
80% confidence
Finding
The manifest description is entirely in Chinese and does not indicate whether other languages are supported, which can cause mismatches between user intent and skill selection in multilingual environments. While not directly a code-execution issue, it can reduce transparency and lead to accidental invocation or poor user understanding of the skill's behavior.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal