ClawHub

GMNCODE Usage

Security checks across malware telemetry and agentic risk

Overview

The skill appears to be a disclosed API helper that uses credentials and token caching for its service, but users should treat the local token cache as sensitive.

Install only if you trust the service and publisher. Use a dedicated account or limited-scope credentials where possible, avoid pasting passwords into chat, inspect where the token cache is written, ensure it is not world-readable, and revoke or rotate the token if the environment is shared or compromised.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (2)

Lp3

Medium
Category
MCP Least Privilege
Confidence
92% confidence
Finding
The skill explicitly instructs use of environment variables, local .env files, local scripts, shell commands, file access, and outbound HTTP requests, yet no permissions are declared. This creates a real least-privilege and transparency problem: an agent may read secrets, write token caches, and make authenticated network calls without an explicit permission boundary visible to users or the runtime.

Credential Access

High
Category
Privilege Escalation
Content
脚本已经处理好了：
- 通过 `/api/v1/auth/login` 登录
- access token 本地缓存
- 遇到 `401` / `INVALID_TOKEN` 时自动重新登录并重试一次
- dashboard 所需的 referer / headers
Confidence
88% confidence
Finding
access token

VirusTotal

57/57 vendors flagged this skill as clean.

View on VirusTotal