ClawHub
Back to skill

Security audit

Build

Security checks across malware telemetry and agentic risk

Overview

This is a developer scaffolding skill for building Aomi Rust plugins, with disclosed workspace writes and build commands but no evidence of hidden execution, exfiltration, or fund movement by the skill itself.

Install this only if you want an agent to create and build Aomi Rust plugin code in your workspace. Review generated code before compiling or loading it, use trusted/local specs where possible, and require explicit confirmation, limits, sandbox/testnet defaults, and post-action verification for any generated tool that places orders, signs messages, submits transactions, or changes account state.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Tool MisuseTool Parameter Abuse, Chaining Abuse, Unsafe Defaults
  • 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 (2)

Missing User Warnings

Medium
Confidence
87% confidence
Finding
The examples include signed trading and wallet-signing workflows that can lead to real financial actions, but they do not consistently require an explicit user-confirmation or risk-warning pattern before order placement or signing. In an agent-skill context, examples strongly influence generated implementations, so this omission can normalize unsafe automation of sensitive actions.

Tool Parameter Abuse

High
Category
Tool Misuse
Content
| `binance_get_klines` | "give me OHLC for technical analysis" | `GET /klines` |
| `binance_get_24hr_stats` | "rolling 24h stats" | `GET /ticker/24hr` |
| `binance_place_order` | "submit a buy/sell order" | `POST /order` (signed) |
| `binance_cancel_order` | "cancel my order" | `DELETE /order` (signed) |
| `binance_get_account` | "what's my balance?" | `GET /account` (signed) |
| `binance_get_trades` | "my fill history" | `GET /myTrades` (signed) |
Confidence
82% confidence
Finding
The example encourages exposing direct signed tools for placing and canceling orders, but it does not show strong parameter constraints, confirmation gating, or misuse-prevention for destructive trading actions. In an agent setting, broad order/cancel capabilities can be abused through prompt injection, user misunderstanding, or malformed tool arguments, causing unwanted trades or loss of funds.

VirusTotal

59/59 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.