ClawHub
Back to skill

Security audit

Crypto Analytics

Security checks across malware telemetry and agentic risk

Overview

This skill does what it claims: it queries public blockchain analytics APIs and caches results locally, with some manageable privacy and API-key handling caveats.

Before installing, understand that addresses and transaction hashes you query will be sent to Etherscan, Blockchair, or public Solana RPC providers, and short-lived local cache files may contain those identifiers. Prefer setting ETHERSCAN_API_KEY as a narrow environment variable rather than putting broad secrets in a shared workspace .env file.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (4)

Context-Inappropriate Capability

Medium
Confidence
95% confidence
Finding
The code walks up to the workspace root and loads a .env file, then implicitly consumes ETHERSCAN_API_KEY. For a read-only blockchain analytics skill, reading workspace-scoped secrets is broader than necessary and creates unnecessary credential exposure if the skill is reused in a larger environment containing unrelated secrets.

Missing User Warnings

Medium
Confidence
84% confidence
Finding
The documentation tells users to create a .env file in the workspace root for API keys without explicitly warning against committing that file, sharing the workspace, or exposing secrets to other tools in the same environment. In an agent-skill context, workspace-root secrets can be more accessible to adjacent scripts, logs, or accidental uploads, increasing the chance of credential leakage.

Missing User Warnings

Medium
Confidence
85% confidence
Finding
User-supplied wallet addresses, transaction hashes, and related identifiers are sent to third-party providers for lookup, which can leak sensitive usage patterns and link a user's interests or holdings to the service operator. In a blockchain analytics skill, these identifiers may be pseudonymous rather than secret, but querying them through external APIs still creates privacy and metadata exposure that users may not expect.

Missing User Warnings

Medium
Confidence
89% confidence
Finding
The client sends wallet addresses, transaction hashes, chain identifiers, and optionally an API key to third-party services without any user-facing disclosure, consent boundary, or minimization. In a security-sensitive agent context, blockchain identifiers can still be sensitive metadata, and silent transmission to external providers increases privacy and key-handling risk.

VirusTotal

63/63 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.