ClawHub

Data

Security checks across malware telemetry and agentic risk

Overview

This instruction-only data skill is broadly purpose-aligned, but it includes production migration and destructive write patterns without enough approval, rollback, or scoping guidance.

Install only if you are comfortable with an agent helping on data workflows. Use read-only or least-privilege credentials, require previews before transformations, and require explicit confirmation plus backups/rollback plans before any production cutover, deprecation, scheduled run, or delete/insert operation.

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
90% confidence
Finding
The skill uses broad natural-language trigger examples like "What does this data tell us?" and "Clean this and load it there," which are common phrases that may appear in many unrelated conversations. This can cause unintended invocation of the Data skill, leading the agent to enter a data-processing workflow when the user did not explicitly request it, potentially affecting task routing, data handling, or downstream actions.

Missing User Warnings

Medium
Confidence
86% confidence
Finding
The data migration workflow explicitly includes a production cutover and deprecating the old source, which are potentially destructive operational steps, but it provides no safety guidance such as backups, rollback plans, approval gates, or user confirmation. In a general-purpose data skill, these steps could be followed by an agent or user without appreciating the risk of downtime, data loss, or irreversible transition errors.

Missing User Warnings

Medium
Confidence
91% confidence
Finding
The idempotent writes example demonstrates a delete-then-insert pattern without emphasizing transaction safety, scoping validation, or warning that an incorrect predicate could erase valid data. In the context of a data automation skill, this snippet could be copied directly into production workflows, increasing the chance of accidental destructive modification or partial writes.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal