ClawHub
Back to skill

Security audit

Agent identity and reputation registration

Security checks across malware telemetry and agentic risk

Overview

This skill is purpose-aligned, but it gives an agent high-impact crypto wallet and blockchain transaction authority without strong confirmation boundaries.

Install only if you trust the @quantulabs/8004-mcp package and are comfortable letting it manage wallet material. Use testnet and dry-run modes first, prefer a dedicated low-balance wallet, avoid passing unnecessary environment variables, do not put real passwords or private keys into chat or copied examples, and require explicit approval before mainnet switches, wallet imports, transfers, registrations, feedback submissions, URI updates, or any other write operation.

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
92% confidence
Finding
The README encourages wallet creation, feedback submission, agent registration, and switching to mainnet, but the examples do not prominently warn that these actions can trigger real blockchain transactions, spend funds, or alter on-chain state. In an agent/LLM-driven MCP context, users may treat examples as safe defaults and unintentionally authorize costly or irreversible operations, especially after switching to mainnet.

Missing User Warnings

Medium
Confidence
91% confidence
Finding
The skill explicitly states that IPFS is pre-configured with a shared Pinata account and that no setup is required, but it does not warn users that uploaded registration or JSON content may be visible to, attributable through, rate-limited by, or otherwise affected by a shared third-party account. In an agent skill that may handle wallet-linked identities, metadata, and registration artifacts, this can lead to unintended data disclosure, cross-tenant exposure, and weak accountability boundaries.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

Detected: suspicious.exposed_secret_literal

File appears to expose a hardcoded API secret or token.

Critical
Code
suspicious.exposed_secret_literal
Location
README.md:282

File appears to expose a hardcoded API secret or token.

Critical
Code
suspicious.exposed_secret_literal
Location
skill.md:226