ClawHub
Back to skill

Security audit

HyperGrok Hyperliquid Orders

Security checks for vulnerabilities and agentic risk

Overview

This skill gives instructions for approved Hyperliquid trading actions and handles a trading key, but the behavior is disclosed and aligned with its purpose.

Install only where an approved trading workflow exists, keep the API wallet tightly scoped, prefer testnet until rehearsed, and treat the environment variable or key file as highly sensitive because it can authorize real orders if pointed at mainnet.

SkillSpector

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 (1)

Env Variable Harvesting

High
Category
Data Exfiltration
Content
from hyperliquid.utils.types import Cloid

def load_key():
    k = os.environ.get("HYPERLIQUID_PRIVATE_KEY")
    if not k:
        p = os.path.expanduser("~/.hyperliquid/api-wallet.key")
        if os.path.exists(p):
Confidence
94% confidence
Finding
The skill instructs the agent to load a private signing key from an environment variable and, if absent, from a local file in the operator’s home directory. Even though this is presented as normal setup for authenticated trading, it exposes a highly sensitive secret to the skill execution path and creates a credential-access pattern that could be abused by any compromised tool, prompt injection, or downstream code path that gains read access to process environment variables or local files.

Static analysis

No suspicious patterns detected.