ClawHub
Back to skill

Security audit

Pump MCP Server

Security checks across malware telemetry and agentic risk

Overview

This skill is purpose-aligned but needs Review because it handles Solana private keys and signing while providing only prose, not auditable code or install details.

Review before installing or relying on this. Do not use real funds or production Solana keys unless the publisher provides auditable source code, dependency manifests, and install/run instructions. Treat restored or generated secret keys as highly sensitive, and use only throwaway development wallets until the implementation can be inspected.

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 (1)

Missing User Warnings

Medium
Confidence
82% confidence
Finding
The skill exposes wallet generation, key restoration, and message signing capabilities, but the documentation does not give explicit user-facing warnings about the sensitivity of private key material or the risks of restoring and holding secret keys in session memory. In an AI-agent context, this increases the chance that users or downstream agents may handle secrets unsafely, misunderstand the trust boundary, or use the tool for real funds without adequate caution.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.