ClawHub

graphify

Security checks across malware telemetry and agentic risk

Overview

The skill’s behavior is mostly disclosed, but its install metadata points to a Python package name that current upstream docs warn is not the official package. ([pypi.org](https://pypi.org/project/graphifyy/))

Review before installing. Do not use this version’s package install path unless the publisher corrects it or you independently verify the `graphify` package provenance; current upstream docs point to `graphifyy` as the official package. If you proceed after verification, add a `.graphifyignore`, exclude secrets and private files, use scoped API keys, and enable assistant hooks or git hooks only when you want persistent behavior. ([pypi.org](https://pypi.org/project/graphifyy/))

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (2)

Missing User Warnings

Medium
Confidence
92% confidence
Finding
The skill states that semantic extraction uses an external LLM API key, but it does not clearly and explicitly warn that project documents, PDFs, images, and potentially other indexed content may be sent to a third-party service for processing. In a coding-assistant context, users may run this against proprietary repositories or sensitive internal documentation, so insufficient disclosure can lead to unintended data exfiltration beyond the local machine.

Missing User Warnings

Medium
Confidence
93% confidence
Finding
The documentation encourages fetching and indexing arbitrary external URLs but does not warn users that doing so may transmit network metadata to third parties, pull in sensitive or untrusted content, or create privacy/compliance issues when used in corporate environments. In the context of an AI assistant skill that promotes adding web, social, and video content into a knowledge graph, the omission is more dangerous because users may treat the workflow as routine and overlook data-handling implications.

VirusTotal

67/67 vendors flagged this skill as clean.

View on VirusTotal