ClawHub
Back to skill

Security audit

plugin skill for clawchain agent to use pancake exchange

Security checks across malware telemetry and agentic risk

Overview

This skill is a coherent PancakeSwap trading helper, but it gives the agent persistent control over a crypto wallet and broad trading authority that users should review carefully.

Install only for a dedicated low-balance wallet, preferably testnet first. Do not reuse a main wallet or store large funds in wallet.json; restrict file permissions, require explicit approval for every transaction, set maximum spend and slippage limits, and verify any ClawChain credential use before running the CLI commands.

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
93% confidence
Finding
The skill instructs the agent to generate and persist a raw blockchain private key in a local JSON file, and only later mentions file permissions and compromise risk. Storing hot-wallet keys unencrypted on disk materially increases the chance of theft by malware, other local users, backups, logs, or accidental disclosure, especially because the same file is intended for ongoing autonomous trading.

Natural-Language Policy Violations

Low
Confidence
82% confidence
Finding
The skill explicitly allows trading of any BEP-20 token and encourages dynamic token resolution without meaningful allowlisting, policy checks, or user-confirmed constraints. In a live trading context this can expose the agent and user funds to scam tokens, toxic approvals, manipulated liquidity pools, and unintended trades outside the user's expected scope.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal