ClawHub
Back to skill

Security audit

GoldRush MCP Skill

Security checks across malware telemetry and agentic risk

Overview

This skill is a coherent GoldRush blockchain-data integration, but users should treat wallet lookups and the GoldRush API key as sensitive.

Install only if you intend to use GoldRush for wallet and chain-data lookups. Use a dedicated least-privilege GoldRush API key, avoid querying addresses without authorization, keep wallet-history scans narrow, consider pinning or reviewing the npm package before use, and remove the uxc credential/link when no longer needed.

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)

Missing User Warnings

Medium
Confidence
90% confidence
Finding
The skill instructs users to inject GOLDRUSH_API_KEY into a child process environment when launching a package fetched via npx. Environment-variable injection is common, but it increases secret exposure risk through process inspection, inherited subprocesses, shell history mistakes, logs, or execution of an unpinned third-party package version (@latest).

Vague Triggers

Medium
Confidence
91% confidence
Finding
The default prompt broadly instructs the agent to use the GoldRush MCP for a wide range of wallet, portfolio, NFT, transfer, approval, price, and chain-data actions without specifying when invocation is appropriate or what user authorization checks must occur first. In a security-sensitive context involving blockchain and wallet intelligence, this can cause overbroad tool use, unnecessary exposure of sensitive financial data, or agent actions based on ambiguous user requests.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.