ClawHub
Back to skill

Security audit

WalletPilot

Security checks across malware telemetry and agentic risk

Overview

WalletPilot is a clearly described crypto wallet automation skill, but it asks users to run missing/unreviewed setup code and entrust an agent with persistent wallet authority for transactions and signatures.

Review this carefully before installing. Do not fund any wallet until the complete implementation, package manifest, lockfile, config files, and guardrail code are available and inspected. If testing, use a brand-new low-balance wallet only, never a main wallet, and require manual confirmation for every transaction and signature.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (2)

Vague Triggers

Medium
Confidence
89% confidence
Finding
The documented activation phrases are very broad, imperative command patterns such as connect, swap, send, and sign, without any clear requirement for explicit user confirmation, trusted invocation context, or exclusion rules. In a wallet automation skill, this creates a high-risk prompt-injection and accidental-trigger surface where untrusted page content, chat text, or embedded instructions could cause the agent to perform sensitive financial actions.

Missing User Warnings

Medium
Confidence
94% confidence
Finding
The skill exposes a sign <message> capability but does not warn that off-chain signatures can authorize token approvals, exchange orders, login attestations, permit flows, or other delegated actions without an on-chain transaction. In the context of a universal wallet automation skill, this omission is especially dangerous because users may wrongly treat signing as harmless compared with sending funds, while an attacker could use signatures to obtain substantial control or value.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.