ClawHub

earth2037

Security checks across malware telemetry and agentic risk

Overview

This is a coherent Earth2037 game helper, but it can use your game credentials/token to perform real in-game actions and post chat messages.

Install only if you trust the Earth2037 server and publisher. Use a unique game password, verify the configured apiBase before logging in, keep EARTH2037_TOKEN private, review chat/combat/build/recruit/airdrop commands before running them, and delete local cache files if you do not want game-state data retained.

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 (3)

Missing User Warnings

Medium
Confidence
90% confidence
Finding
The skill instructs the agent to collect usernames and passwords for registration, login, and key recovery flows without any visible privacy notice, storage policy, or handling limits. Because the same skill also writes local cache files and uses networked auth endpoints, users may disclose credentials without understanding retention, logging, or exposure risks.

Missing User Warnings

Medium
Confidence
91% confidence
Finding
The `airdropres` path issues the state-changing `AIRDROPRES` command immediately, and the CLI `main()` will even fall back to a cached village ID when no explicit target is provided. That makes it easy for an agent or user to trigger an in-game resource drop unintentionally, consuming a limited action without a confirmation prompt or clear warning about side effects.

Missing User Warnings

Medium
Confidence
86% confidence
Finding
The send-world and send-ally paths transmit chat messages to the live game service immediately after CLI parsing, with no confirmation prompt, dry-run mode, or explicit user acknowledgment. In an agent skill context, this is risky because a mistaken invocation, prompt injection, or ambiguous user request could cause unintended external actions on behalf of the user using their stored token.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal