ClawHub

UnifAI

Security checks across malware telemetry and agentic risk

Overview

UnifAI is a disclosed CLI integration, but it needs Review because it can sign and submit real blockchain transactions when wallet private keys are configured.

Install only if you intentionally want a broad UnifAI service-invocation CLI with DeFi transaction support. Keep wallet private keys unset unless actively signing, prefer dedicated low-balance wallets, avoid storing private keys in shared config or shell history, and manually inspect every payload before using --sign or tx sign.

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 provides concrete `--sign` examples for transfers and swaps that can move funds on-chain, but it does not place an explicit, prominent warning next to those examples that these operations are irreversible and may spend real assets. In an agent context, executable examples are especially risky because they can normalize direct use of signing flows and increase the chance of accidental fund movement.

Missing User Warnings

Medium
Confidence
90% confidence
Finding
The documentation instructs users to provide raw private keys via environment variables but does not clearly warn that these secrets are highly sensitive and may be exposed through shell history, process inspection, logs, crash reports, or inherited subprocess environments. In an agent/CLI setting, encouraging raw key export without strong handling guidance materially increases credential-compromise risk.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal