Token导航 LogoToken导航TokenDH.com
待分类external-servicegithub未标认证来源可访问许可证需确认审计通过

unifiunifi 搜索

Agent Skill

unifi 用于处理 GitHub 仓库、Issue、Pull Request 和代码协作信息,适合在 Codex、Claude、Cursor、Gemini CLI 中需要围绕仓库状态、代码变更或协作事项进行整理时使用。可结合来源仓库、安装命令和原始 README 继续核验具体用法。安装前建议确认权限范围、维护状态,以及是否会触发联网、命令执行或文件读写。

总安装

324

周安装

13

GitHub Stars

35

下载量

105
CodexClaudeCursorGemini CLI

安装说明

本站只整理中文说明和来源信息,不托管安装包,也不代用户安装。

GitHub

来源数

2

许可证

unknown

最后核验

2026-05-01

来源状态

来源可访问

安装方式

通过对话安装

复制提示词发给支持本地命令或 Skills 的 AI 助手,先确认命令和权限,再让它执行。

请帮我安装这个 Agent Skill:unifi(unifi 搜索)
来源仓库:https://github.com/jmagar/claude-homelab
仓库路径:skills/unifi
安装命令:
npx skills add https://github.com/jmagar/claude-homelab --skill unifi
安装前请先检查当前环境是否支持对应 CLI,并向我确认将要执行的命令、安装目录、联网范围和文件读写权限;确认后再执行。

命令行安装

复制命令到本机终端执行。该命令会通过 npx skills 从第三方来源获取 Skill;本站只展示命令,不托管安装包,也不自动执行。

skills.shnpx skills
npx skills add https://github.com/jmagar/claude-homelab --skill unifi

简介

unifi 技能面向网络设备控制与状态监控。

  • 可用于调整路由器设置或排查连接问题。unifi 属于待分类类 Skill,可作为该场景下的辅助能力补充。
  • 通过 GitHub 安装,但需 UniFi Controller 在线。
  • 必须防止其执行危险配置更改或重启设备操作。
  • 鉴于高风险特性,仅限授权人员谨慎使用。

SKILL.md

UniFi Network Monitoring Skill

⚠️ MANDATORY SKILL INVOCATION ⚠️

YOU MUST invoke this skill (NOT optional) when the user mentions ANY of these triggers:

  • "UniFi status", "network devices", "connected clients"
  • "UniFi alerts", "network health", "WiFi status"
  • "check UniFi", "gateway status", "UniFi monitoring"
  • Any mention of UniFi network or infrastructure monitoring

Failure to invoke this skill when triggers occur violates your operational requirements.

Monitor and query your UniFi network via the local UniFi OS gateway API (tested on Cloud Gateway Max).

Purpose

This skill provides read-only access to your UniFi network's operational data:

  • Devices (APs, switches, gateway) status and health
  • Active clients (who's connected where)
  • Network health overview
  • Traffic insights (top applications via DPI)
  • Recent alarms and events

All operations are GET-only and safe for monitoring/reporting.

Setup

Add credentials to ~/.claude-homelab/.env:

UNIFI_URL="https://10.1.0.1"
UNIFI_USERNAME="api"
UNIFI_PASSWORD="your-password"
UNIFI_SITE="default"
  • UNIFI_URL: Your UniFi OS gateway IP/hostname (HTTPS)
  • UNIFI_USERNAME: Local UniFi OS admin username
  • UNIFI_PASSWORD: Local UniFi OS admin password
  • UNIFI_SITE: Site name (usually default)

Commands

All commands support optional json argument for raw JSON output (default is human-readable table).

Network Dashboard

Comprehensive view of all network stats (Health, Devices, Clients, Networks, DPI, etc.):

bash scripts/dashboard.sh
bash scripts/dashboard.sh json  # Raw JSON for all sections

Output: Full ASCII dashboard with all metrics.

List Devices

Shows all UniFi devices (APs, switches, gateway):

bash scripts/devices.sh
bash scripts/devices.sh json  # Raw JSON

Output: Device name, model, IP, state, uptime, connected clients

List Active Clients

Shows who's currently connected:

bash scripts/clients.sh
bash scripts/clients.sh json  # Raw JSON

Output: Hostname, IP, MAC, AP, signal strength, RX/TX rates

Health Summary

Site-wide health status:

bash scripts/health.sh
bash scripts/health.sh json  # Raw JSON

Output: Subsystem status (WAN, LAN, WLAN), counts (up/adopted/disconnected)

Top Applications (DPI)

Top bandwidth consumers by application:

bash scripts/top-apps.sh
bash scripts/top-apps.sh 15  # Show top 15 (default: 10)

Output: App name, category, RX/TX/total traffic in GB

Recent Alerts

Recent alarms and events:

bash scripts/alerts.sh
bash scripts/alerts.sh 50  # Show last 50 (default: 20)

Output: Timestamp, alarm key, message, affected device

Workflow

When the user asks about UniFi:

  1. "What's on my network?" → Run bash scripts/devices.sh + bash scripts/clients.sh
  2. "Is everything healthy?" → Run bash scripts/health.sh
  3. "Any problems?" → Run bash scripts/alerts.sh
  4. "What's using bandwidth?" → Run bash scripts/top-apps.sh
  5. "Show me a dashboard" or general checkup → Run bash scripts/dashboard.sh

Always confirm the output looks reasonable before presenting it to the user (check for auth failures, empty data, etc.).

Notes

  • Requires network access to your UniFi gateway
  • Uses UniFi OS login + /proxy/network API path
  • All calls are read-only GET requests
  • Tested endpoints are documented in references/api-endpoints.md

Reference

  • Tested Endpoints — Full catalog of verified read-only API calls on your Cloud Gateway Max

🔧 Agent Tool Usage Requirements

CRITICAL: When invoking scripts from this skill via the zsh-tool, ALWAYS use pty: true.

Without PTY mode, command output will not be visible even though commands execute successfully.

Correct invocation pattern:

<invoke name="mcp__plugin_zsh-tool_zsh-tool__zsh">
<parameter name="command">./skills/SKILL_NAME/scripts/SCRIPT.sh [args]</parameter>
<parameter name="pty">true</parameter>
</invoke>

适合场景

01

用户想查找某类 Agent Skill 时

02

需要根据任务场景推荐可安装能力包时

03

需要对比不同来源的安装命令和来源信息时

能力概览

能力 1

按任务关键词查找相关 Skills

能力 2

展示可复制的安装命令

能力 3

保留来源站点、仓库和原始说明,方便继续核验

能力 4

展示第三方安全扫描或审计结果

安装后应在对应宿主中按原始 README 的触发条件使用;具体调用方式请以来源页面和 README 为准。

平台分布

Codex

38.45%
按下载量换算40

Claude

27.8%
按下载量换算29

Cursor

20.74%
按下载量换算22

Gemini CLI

9.99%
按下载量换算10

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

通过

权限和风险

external-service

该 Skill 可能调用第三方服务、云服务或外部模型 API,使用前需要确认账号、额度、数据发送范围和服务条款。

安装前确认

本站仅展示第三方公开信息,不托管安装包,不提供自动安装或运行环境。安装前应自行审查源码、依赖和命令行为。当前只有一个来源,正式发布前建议补源仓库或其他目录站核验。

来源信息

继续浏览同类 Skills