ClawHub

Doc Co

Security checks across malware telemetry and agentic risk

Overview

This is a disclosed document co-authoring workflow with privacy-sensitive connector use that is aligned with its purpose and should be scoped carefully by the user.

Install if you want a guided workflow for drafting and testing documents. When using it, point connectors only at the specific documents, threads, or channels needed, and avoid importing secrets, regulated data, personnel details, or unrelated confidential discussions. Review the draft before reader testing because that step may share the document content with another model context.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (2)

Vague Triggers

Medium
Confidence
92% confidence
Finding
The trigger conditions are broad enough to activate on many ordinary writing requests, which can cause the skill to take over interactions unexpectedly and steer users into a heavyweight workflow they did not explicitly request. In a security context, overbroad activation increases the chance that the skill will solicit or process unnecessary sensitive context and use integrations in situations where a narrower trigger would have avoided access.

Missing User Warnings

Medium
Confidence
97% confidence
Finding
The skill encourages pulling data from team channels, shared documents, and external connectors without first presenting a clear privacy notice, scope limitation, or sensitivity check. That can lead to unnecessary collection of confidential business, personal, or regulated information from integrated systems, especially because the workflow explicitly asks for broad organizational and stakeholder context.

VirusTotal

63/63 vendors flagged this skill as clean.

View on VirusTotal