ClawHub
Back to skill

Security audit

sugerclawdy skill

Security checks across malware telemetry and agentic risk

Overview

This skill appears to register an agent on SugarClawdy, but it asks the AI to create and locally save Ethereum wallet secrets without clear safeguards.

Review carefully before installing. Do not fund or reuse any wallet generated through this skill unless you independently secure the private key and mnemonic. Prefer using a wallet address you already control, avoid storing seed phrases in agent-accessible files, and understand that the workflow runs an external npm package and sends your wallet address and agent name to SugarClawdy.

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

High
Confidence
97% confidence
Finding
The skill instructs the agent to generate an Ethereum wallet and exposes highly sensitive secrets including the private key and mnemonic, then says to save them locally without any warning, encryption guidance, or user consent flow. This creates a real risk of credential theft, unintended persistence, and downstream loss of control over the wallet or any assets later associated with it.

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The skill directs the agent to transmit wallet-derived identity data to a third-party service and to use the wallet address itself as a bearer token, but provides no disclosure, consent step, or trust boundary warning. Even if the wallet address is public by nature, using it as an authentication token encourages insecure auth design and may expose the agent or user to tracking, unauthorized registration actions, or reliance on a weak identity mechanism.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.