ClawHub

Knowledge-Base

Security checks across malware telemetry and agentic risk

Overview

This is an instruction-only Markdown knowledge-base management skill with disclosed local note edits, and no evidence of hidden execution, exfiltration, or credential use.

Install this only if you want an agent to manage the configured local Markdown knowledge base. Keep secrets and private customer or business data out of notes, review diffs before bulk edits, archive moves, deletions, or index rewrites, inspect the local wiki-lint.js script before running it, and approve git commits or pushes only after checking exactly what will be staged.

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
91% confidence
Finding
The skill description is broad enough to match many ordinary knowledge-work requests, which can cause unintended activation outside the user’s actual intent. In an agent setting, overbroad triggering increases the chance of unnecessary file operations, note reorganization, or application of this protocol to unrelated content, which can lead to integrity and privacy issues in the knowledge base.

Natural-Language Policy Violations

Medium
Confidence
80% confidence
Finding
Allowing a reference configuration to dictate the document language without incorporating user preference or a documented compliance basis can cause the agent to transform or store content in an unexpected language. This is risky because it may degrade accuracy, obscure meaning, and mishandle user-provided sensitive or regulated content when language choice affects reviewability and correctness.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal