ClawHub

COCOS CLI Game Development

Security checks across malware telemetry and agentic risk

Overview

This Cocos game-development skill is mostly coherent, but it asks the agent to automatically start and trust a long-running local MCP server without a clear user approval or shutdown step.

Install only if you trust the local Cocos CLI and MCP server. Before use, confirm the project directory, approve any MCP registration yourself, ensure the server is bound locally and stopped after the task, and review `.learnings/ERRORS.md` before sharing or committing the project.

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 (2)

Missing User Warnings

Medium
Confidence
90% confidence
Finding
The skill instructs the agent to start a background MCP server, automatically connect/register it, and enumerate its tools without waiting for user approval. That creates an implicit trust boundary crossing and can expose local project context, available capabilities, or network-reachable interfaces before the user has consented, which is especially risky because MCP tools may enable powerful file or system actions.

Missing User Warnings

Medium
Confidence
88% confidence
Finding
The skill requires logging exact commands, error output, and context to `.learnings/ERRORS.md` without notifying the user that potentially sensitive information will be persisted to disk. Build errors and command output often contain file paths, environment details, tokens, stack traces, or proprietary project data, so automatic retention increases the risk of unintended disclosure and long-term leakage.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal