ClawHub

Knowledge Distill

Security checks across malware telemetry and agentic risk

Overview

This skill is a local knowledge-archiving helper that requires preview and confirmation before writing notes, with no evidence of hidden execution or data exfiltration.

Review the preview and destination carefully before confirming a save, avoid archiving secrets or private data, and set KNOWLEDGE_BASE_PATH to a private folder if the notes may contain sensitive information.

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 (2)

Vague Triggers

Medium
Confidence
94% confidence
Finding
The trigger phrase "存起来" is broad enough to appear in normal conversation, which can cause the skill to activate when the user did not intend to archive content. In this skill, unintended activation matters because the workflow can lead to persistent storage of conversation-derived content into local knowledge files, creating privacy, integrity, and data pollution risks even though a confirmation step reduces the likelihood of final write-through.

Vague Triggers

Medium
Confidence
91% confidence
Finding
Using "总结一下" as an activation condition is ambiguous because it is a common conversational request that often means "summarize in chat" rather than "persist to disk." In this skill's context, that ambiguity can turn an ordinary summarization request into a storage action pipeline, increasing the chance of accidental retention of sensitive or irrelevant material.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal