Security audit

Skills

Security checks across malware telemetry and agentic risk

Overview

This is a disclosed crypto automation helper for Rerange orders, with financial risk but no hidden credential collection, install-time execution, or automatic signing found.

Install only if you intend to use Rerange in a wallet-aware environment. Keep signing in your wallet or trusted runtime, never provide private keys or seed phrases, prefer canonical Rerange deployments and trusted RPC providers, and review every approval, Permit2 signature, vault delegation, close, call, multicall, or resolver automation before allowing it to run.

SkillSpector

By NVIDIA

Vulnerability Patterns

MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
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 (1)

Tp4

High

Category: MCP Tool Poisoning
Confidence: 93% confidence
Finding: The skill advertises a narrow purpose around building and managing Rerange orders, but it also exposes generic introspection primitives: raw ABI retrieval, arbitrary calldata encoding, and generic read-only calls to caller-supplied addresses and RPC URLs. Those capabilities broaden the trust boundary and can be abused to probe unintended contracts, exfiltrate metadata to attacker-controlled RPC endpoints, or support follow-on transaction crafting beyond the stated user expectation.

VirusTotal

63/63 vendors flagged this skill as clean.

View on VirusTotal