ClawHub
Back to skill

Security audit

Rydberg Agent Node

Security checks across malware telemetry and agentic risk

Overview

This skill is a disclosed ProbeChain testnet node installer, but it will download or build software, store a local node password, and run a background node process.

Install only if you intend to run a ProbeChain Rydberg Agent testnet node on this machine. Review the GitHub source or release being used, use a unique node password, and be comfortable with a long-running local process, blockchain data/logs under ~/rydberg-agent, and local account unlocking through IPC.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Behavioral ASTexec() Call, eval() Call, Dynamic Import
  • 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)

Missing User Warnings

Medium
Confidence
89% confidence
Finding
The instructions write a node password to disk and create a persistent installation directory, but they do not prominently warn the user that credentials and long-lived local artifacts will be created. This increases the chance a user consents without understanding that a reusable secret is being stored locally and that the process persists beyond the session.

Missing User Warnings

Medium
Confidence
94% confidence
Finding
The skill performs network downloads from GitHub and may compile and execute retrieved source code, yet it does not clearly warn the user that outbound connections and code execution will occur. In a skill with system:exec and network:outbound permissions, this omission materially increases risk because users may not realize they are authorizing remote code retrieval and local build/execution.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.