Nb MCP Server 6a49aba0

MCPAI代理

提供方 netbeez

AI代理 未标注 Token 社区 / 第三方 Claude DesktopClaudeCursorWindsurf

NetBeez MCP服务器是一个连接LLM客户端和NetBeez网络监控平台的协议服务器,支持AI辅助的网络故障排除、监控分析和事件调查。

一句话选型结论

按资料字段整理

Nb MCP Server 6a49aba0 属于AI代理类 MCP 服务器,传输方式未标注,Token,可直接接入 Claude Desktop、Claude、Cursor 等,没有现成安装命令,需按官方文档接入。接入前先确认权限范围;涉及写入、删除或付款的操作,建议先在测试环境验证。

分类AI代理 传输方式未标注 鉴权Token 部署方式未标注
GitHub 星标 2 仓库热度,无公开仓库时显示未标注
工具数 32 暴露给模型的工具数量
资源 / 提示词 0 / 0 资源与提示词模板数量
维护方 社区 / 第三方 核验等级 仅资料核对

工具与能力

AI代理
数量来自来源目录的标注,具体工具清单以官方文档为准
工具数量32暴露给模型可调用的工具个数
资源数量0可读取的资源条目
提示词数量0内置提示词模板数量

能力标签

支持客户端

适用场景

网络监控TypeScript本地部署故障排除AI辅助协议服务器网络分析

安装与配置

未标注 Token
上游未给出传输方式
  1. 1
    准备运行环境

    按官方文档确认运行环境(常见为 Node.js 或 Python),装好后再执行安装命令。

  2. 2
    安装服务

    复制上面的安装命令执行;命令里带占位符的位置,换成自己的密钥或路径。

  3. 3
    写入客户端配置

    在客户端的 MCP 配置里填好启动命令和参数,保存后完全退出客户端再打开,让新配置生效。

  4. 4
    验证是否可用

    连接成功后客户端会列出约 32 个工具,随便调一个只读工具确认能返回结果。

典型使用场景

3 类
按分类与标签整理的常见用法,具体以官方文档为准
给代理接外部能力

让代理在任务中途调用搜索、数据库等真实工具。

多步任务编排

把一串操作拆成步骤,由代理按顺序执行。

自动化重复流程

把固定流程交给代理跑,减少手工操作。

选择建议

选之前先看这几条
按条目字段给出的判断,不替你决定用不用

值得选的地方

  • 资料标注 32 个工具,覆盖面较广,能满足多种调用。
  • 已适配 Claude Desktop、Claude、Cursor,换客户端不用重新研究配置。

先想清楚的地方

  • 没有现成安装命令,需要按官方文档自己填启动命令和参数。
  • 使用前要准备好Token对应的凭据,并按最小权限配置。
  • 版本更新与行为细节以官方仓库为准。

更适合

  • 需要在AI代理场景里让模型直接调用外部能力的人。
  • 已经在用 Claude Desktop、Claude 的用户,接入路径最短。
  • 想先快速试一下、能接受在本地或自有环境跑服务的人。

不太适合

  • 不方便管理密钥或第三方授权的人。
  • 需要厂商级 SLA 和技术支持的生产环境。

同类对比

6 个同类
同分类里挑最接近的几条,字段以各自详情页为准
MCP 服务器工具数传输方式部署方式安装命令最后核验
Nb MCP Server 6a49aba0当前 32 未标注 未标注 无 2026-05-17
Filesystem 560c4363 13 未标注 本地部署 无 2026-05-17
Server Everything (everything) 11 stdio · 本地进程 本地部署 有 2026-05-17
Time 7365cb6d 2 SSE · 服务端推送 远程部署 有 2026-05-18
Everything Claude Code 未标注 stdio · 本地进程 本地部署 有 2026-05-17
Openmemory 557e1585 未标注 stdio · 本地进程 本地部署 有 2026-05-17

服务器档案

AI代理
字段来自项目仓库与公开索引,中文说明由本站标注
提供方netbeez上游标注的维护方 netbeez
分类AI代理本站按上游分类展示
连接方式未标注上游未给出传输方式
鉴权方式Token远程服务建议用最小权限凭据
维护方社区 / 第三方核验等级 仅资料核对
包名 / 运行时未标注 · 未标注本地安装时用到的包标识与运行时
部署类型未标注上游给出的部署形态
最近核验2026-05-17上游最后核验时间

详细介绍

14.8 KB
来自项目仓库与上游目录的原文,未做删改

NetBeez MCP Server

A Model Context Protocol (MCP) server that connects LLM clients (Cursor, Claude Desktop, etc.) to the NetBeez network monitoring platform. Enables AI-assisted network troubleshooting, monitoring analysis, and incident investigation.

Features

  • 32 tools for querying and managing agents, agent groups, targets, tests, scheduled test templates, alerts, incidents, WiFi profiles, statistics, path analysis, and running ad-hoc Iperf/speed/VoIP/custom command tests (including multiagent run status)
  • 5 resources providing the LLM with NetBeez data model knowledge, cross-agent correlation methodology, troubleshooting workflows, full API reference, and ad-hoc test guidance
  • 5 prompt templates for common workflows: troubleshoot target, analyze agent health, investigate incident, network overview, run ad-hoc tests
  • Dual transport: stdio (for Cursor/Claude Desktop) and HTTP (for remote clients)

Quick Start (Recommended)

Run the one-line installer — it handles everything: Node.js, dependencies, build, credentials, and MCP client configuration:```bash
curl -fsSL https://raw.githubusercontent.com/netbeez/nb-mcp-server/main/install.sh | bash


The installer will prompt you for:
1. Your **NetBeez instance URL** (e.g. `https://demo1.netbeezcloud.net`)
2. Your **API key** (from Dashboard → Settings → API Keys)
3. **SSL verification** preference (disable for self-signed certs)
4. Which **MCP clients** to configure (Cursor, Claude Desktop, Windsurf, Codex, Kiro)

The script is idempotent — re-run it any time to update to the latest version or change your configuration.

## Quick Start (Windows)

Run the one-line installer from PowerShell — it handles everything: Node.js, dependencies, build, credentials, and MCP client configuration:```powershell
irm https://raw.githubusercontent.com/netbeez/nb-mcp-server/main/install.ps1 | iex

The installer will prompt you for the same values as the Unix installer (instance URL, API key, SSL preference, and which MCP clients to configure). It is idempotent — re-run it any time to update or change your configuration.

If execution policy blocks a local run, use:```powershell
powershell -ExecutionPolicy Bypass -File .\install.ps1


### Development install (no git push required)

From a clone of this repo, run the installer with `--dev` to point Cursor, Claude Desktop, Codex, Windsurf, and Kiro at your local build:```bash
./install.sh --dev
# or from anywhere:
bash ~/path/to/nb-mcp-server/install.sh --dev
``````powershell
Set-ExecutionPolicy -Scope Process -ExecutionPolicy Bypass
.\install.ps1 --dev

Do not use cat install.sh · bash or irm install.ps1 · iex for development mode — those invocations receive no --dev flag and will download to the default install directory.

This skips cloning; it uses the current directory, runs npm install and npm run build, then configures the same MCP clients to use ./dist/index.js. After making changes, run npm run build and restart the client — no need to push to git to test.

What the installer does

Step Detail
Node.js Checks for Node.js 18+; installs via Homebrew, apt, or nvm (macOS/Linux) or winget (Windows) if missing
Download Downloads the latest source from GitHub (no git required) and extracts to ~/.netbeez-mcp (%USERPROFILE%\.netbeez-mcp on Windows)
Build Runs npm install and npm run build
Configure Prompts for credentials and writes ~/.netbeez-mcp/.env
MCP clients Merges the server entry into Cursor / Claude Desktop / Windsurf / Codex / Kiro config

Prerequisites

If you prefer to install manually, you'll need:

  • Node.js 18+
  • A NetBeez BeezKeeper instance with API access
  • An API key (Dashboard → Settings → API Keys)

The one-line installer only requires Node.js 18+ and curl (macOS/Linux) or PowerShell 5.1+ (Windows). No git required.

Manual Installation

1. Download & build```bash

curl -fsSL https://github.com/netbeez/nb-mcp-server/archive/refs/heads/main.tar.gz -o nb-mcp-server.tar.gz
tar -xzf nb-mcp-server.tar.gz && cd nb-mcp-server-main
npm install
npm run build


### 2. Configure

Copy the example environment file and fill in your values:```bash
cp .env.example .env

Edit .env:```
NETBEEZ_BASE_URL=https://your-instance.netbeezcloud.net
NETBEEZ_API_KEY=your-api-key-here


### 3. Run```bash
npm start

MCP Client Configuration

Cursor

Add to your Cursor MCP settings (~/.cursor/mcp.json on macOS/Linux, $env:USERPROFILE\.cursor\mcp.json on Windows):```json
{
"mcpServers": {
"netbeez": {
"command": "node",
"args": ["~/.netbeez-mcp/dist/index.js"],
"env": {
"NETBEEZ_BASE_URL": "https://your-instance.netbeezcloud.net",
"NETBEEZ_API_KEY": "your-api-key-here"
}
}
}
}


### Claude Desktop

Add to your Claude Desktop config (`~/Library/Application Support/Claude/claude_desktop_config.json` on macOS, `~/.config/Claude/claude_desktop_config.json` on Linux, `$env:APPDATA\Claude\claude_desktop_config.json` on Windows):```json
{
  "mcpServers": {
    "netbeez": {
      "command": "node",
      "args": ["~/.netbeez-mcp/dist/index.js"],
      "env": {
        "NETBEEZ_BASE_URL": "https://your-instance.netbeezcloud.net",
        "NETBEEZ_API_KEY": "your-api-key-here"
      }
    }
  }
}

Windsurf

Add to your Windsurf MCP config (~/.codeium/windsurf/mcp_config.json on macOS/Linux, $env:USERPROFILE\.codeium\windsurf\mcp_config.json on Windows):```json
{
"mcpServers": {
"netbeez": {
"command": "node",
"args": ["~/.netbeez-mcp/dist/index.js"],
"env": {
"NETBEEZ_BASE_URL": "https://your-instance.netbeezcloud.net",
"NETBEEZ_API_KEY": "your-api-key-here"
}
}
}
}


### Codex

Add to your Codex config (`~/.codex/config.toml` on macOS/Linux, `$env:USERPROFILE\.codex\config.toml` on Windows):```toml
[mcp_servers.netbeez]
command = "node"
args = ["~/.netbeez-mcp/dist/index.js"]

[mcp_servers.netbeez.env]
NETBEEZ_BASE_URL = "https://your-instance.netbeezcloud.net"
NETBEEZ_API_KEY = "your-api-key-here"

Kiro

Add to your Kiro user MCP config (~/.kiro/settings/mcp.json on macOS/Linux, $env:USERPROFILE\.kiro\settings\mcp.json on Windows):```json
{
"mcpServers": {
"netbeez": {
"command": "node",
"args": ["~/.netbeez-mcp/dist/index.js"],
"env": {
"NETBEEZ_BASE_URL": "https://your-instance.netbeezcloud.net",
"NETBEEZ_API_KEY": "your-api-key-here"
}
}
}
}


Kiro also supports workspace-level MCP config at `.kiro/settings/mcp.json`.

> **Tip:** The installer (`curl · bash` or `irm · iex` above) writes these config files automatically.

## Updating

Re-run the installer to pull the latest version and rebuild:```bash
curl -fsSL https://raw.githubusercontent.com/netbeez/nb-mcp-server/main/install.sh | bash
``````powershell
irm https://raw.githubusercontent.com/netbeez/nb-mcp-server/main/install.ps1 | iex

Your existing credentials will be shown as defaults — press Enter to keep them.

If you installed with --dev, run npm run build in the repo and restart your MCP client to pick up changes.

Uninstalling```bash

rm -rf ~/.netbeez-mcp
``````powershell
Remove-Item -Recurse -Force "$env:USERPROFILE.netbeez-mcp"


Then remove the `"netbeez"` entry from your MCP client config file(s).

## Environment Variables

| Variable | Required | Default | Description |
|----------|----------|---------|-------------|
| `NETBEEZ_BASE_URL` | Yes | — | BeezKeeper instance URL |
| `NETBEEZ_API_KEY` | Yes | — | API key for authentication |
| `NETBEEZ_SSL_VERIFY` | No | `true` | Set to `false` for self-signed certs |
| `MCP_TRANSPORT` | No | `stdio` | Transport mode: `stdio`, `http`, or `both` |
| `MCP_HTTP_PORT` | No | `3000` | HTTP transport port |

## Tools (32)

### Agent Tools (6)

| Tool | Description |
|------|-------------|
| `list_agents` | List all agents with filtering by name, category, class, active status |
| `get_agent` | Get detailed info for a specific agent by ID |
| `search_agents` | Search agents by name (exact or regex) |
| `get_agent_logs` | Get connection/disconnection event logs (including WiFi/DHCP for wireless) |
| `get_agent_performance_metrics` | Get CPU, memory, disk usage over time |
| `get_agent_access_point_connections` | Get WiFi access point connection history (SSID/BSSID, signal, DHCP) for wireless troubleshooting |

### Agent Group Tools (4)

| Tool | Description |
|------|-------------|
| `list_agent_groups` | List agent groups with filtering and pagination |
| `get_agent_group` | Get a single agent group with relationships |
| `create_agent_group` | Create an agent group with member agents and optional target assignment |
| `update_agent_group` | Update an existing agent group (name, members, targets, auto_assign) |

### Target Tools (6)

| Tool | Description |
|------|-------------|
| `list_targets` | List monitoring targets with filters (agent groups, open incident, WiFi/wired, etc.) |
| `get_target` | Get target details with test templates |
| `create_target` | Create a custom monitoring target (raw JSON:API payload) |
| `create_target_saas` | Create a target from a built-in SaaS application |
| `create_target_from_template` | Create a target from simplified templates (Website, DNS, VPN, Gateway) |
| `update_target` | Update an existing target by ID (raw JSON:API payload) |

### Test & Results Tools (3)

| Tool | Description |
|------|-------------|
| `list_tests` | List all monitoring tests (ping/dns/http/traceroute/path_analysis) |
| `get_test_results` | Get test results by type with time range and agent filters |
| `get_path_analysis_results` | Get hop-by-hop path analysis data |

### Scheduled Test Tools (5)

| Tool | Description |
|------|-------------|
| `list_scheduled_test_templates` | List scheduled Iperf, Network Speed, and VoIP templates with filters |
| `get_scheduled_test_template` | Get a scheduled test template by ID |
| `create_scheduled_test_template` | Create a scheduled Iperf/Speed/VoIP template |
| `update_scheduled_test_template` | Update an existing scheduled test template |
| `get_scheduled_test_results` | Get results for a scheduled template (bandwidth, VoIP quality, Iperf) by time range and agents |

### Alert & Incident Tools (2)

| Tool | Description |
|------|-------------|
| `list_alerts` | List alerts with extensive filtering (severity, status, time, agents, targets) |
| `list_incidents` | List incidents with timeline, acknowledgement status, and event logs |

### Statistics Tools (3)

| Tool | Description |
|------|-------------|
| `get_test_statistics` | Aggregated test performance statistics over time |
| `get_agent_statistics` | Agent uptime/availability statistics |
| `get_access_point_metrics` | WiFi signal quality metrics (signal, quality, rates) |

### Ad-hoc & Other (3)

| Tool | Description |
|------|-------------|
| `run_adhoc_test` | Run an on-demand Iperf test (agent-to-agent or agent-to-server via IP/FQDN), network speed test, VoIP test (agent-to-agent), or custom command test (script on one or more agents); returns multiagent run ID |
| `get_multiagent_test_run_status` | Get status and results of a multiagent test run by ID (use after `run_adhoc_test` to poll or retrieve results) |
| `list_wifi_profiles` | List WiFi profiles and their incident status |

## Resources (5)

| Resource | URI | Description |
|----------|-----|-------------|
| NetBeez Data Model | `netbeez://data-model` | Entity relationships, data shapes, timeseries data catalog |
| Correlation Guide | `netbeez://correlation-guide` | Cross-agent correlation methodology, alert interpretation |
| Troubleshooting Guide | `netbeez://troubleshooting-guide` | Step-by-step troubleshooting workflows |
| API Reference | `netbeez://api-reference` | Comprehensive endpoint reference for JSON:API and legacy stats APIs |
| Ad-hoc Tests Guide | `netbeez://ad-hoc-tests` | Ad-hoc Iperf/Speed/VoIP/Custom Command payloads, polling, and result parsing |

## Prompts (5)

| Prompt | Arguments | Description |
|--------|-----------|-------------|
| `troubleshoot-target` | `target_name` or `target_id` | Guided workflow to diagnose target issues |
| `analyze-agent-health` | `agent_name` or `agent_id` | Comprehensive agent health check |
| `investigate-incident` | `incident_id` | Deep dive into a specific incident |
| `network-overview` | (none) | Overall network health summary |
| `run-adhoc-test` | `test_type`, `agent_hint`, `destination_hint` | Guided ad-hoc execution flow for Iperf, Speed, VoIP, and Custom Command tests |

## Development```bash
# Build
npm run build

# Watch mode
npm run dev

# Test with MCP Inspector
npm run inspect

Architecture```

src/
├── index.ts # Entry point, transport setup
├── server.ts # MCP server definition, registration
├── config.ts # Environment variable loading
├── api/
│ ├── base-client.ts # Shared HTTP client (fetch, retries, errors)
│ ├── jsonapi-client.ts # JSON:API client (Bearer auth, filters, pagination)
│ ├── legacy-client.ts # Legacy API client (API-VERSION v1)
│ └── types.ts # TypeScript types for all entities
├── tools/
│ ├── agents.ts # Agent tools (6)
│ ├── agent-groups.ts # Agent group tools (4)
│ ├── targets.ts # Target tools (6)
│ ├── tests.ts # Test tools (2)
│ ├── incidents.ts # Incident tools (1)
│ ├── alerts.ts # Alert tools (1)
│ ├── wifi.ts # WiFi tools (1)
│ ├── scheduled-tests.ts # Scheduled test template & results tools (5)
│ ├── statistics.ts # Statistics tools (3)
│ ├── path-analysis.ts # Path analysis tools (1)
│ ├── actions.ts # Ad-hoc test tools (2): run_adhoc_test, get_multiagent_test_run_status
│ └── ad-hoc.ts # Re-exports actions for ad-hoc tests
├── resources/
│ ├── data-model.ts # Entity relationships and data shapes
│ ├── correlation-guide.ts # Cross-agent correlation patterns
│ ├── troubleshooting-guide.ts
│ ├── api-reference.ts # API endpoint reference
│ └── ad-hoc-tests.ts # Ad-hoc tests payloads and result parsing guide
└── prompts/
├── troubleshoot-target.ts
├── analyze-agent-health.ts
├── investigate-incident.ts
├── network-overview.ts
└── run-adhoc-test.ts
```

API Coverage

The server communicates with two NetBeez API layers:

  • JSON:API (primary): All entity and relationship endpoints using Authorization: Bearer authentication. Supports filtering, pagination, includes, and ordering.
  • Legacy API: Three statistics endpoints (nb_test_statistics, nb_agent_statistics, access_point_metrics) using Authorization: + API-VERSION: v1 headers.

License

MIT

常见问题

6 问
答案由本站依据条目字段整理,不引用上游原文
Nb MCP Server 6a49aba0 怎么安装?

资料没有给出安装命令,请到官方仓库或文档按说明接入;这类条目多数是生态应用,或尚未发布可安装包。

Nb MCP Server 6a49aba0 怎么在 Claude Desktop、Claude、Cursor 里配置?

它的传输方式未标注,在客户端的 MCP 配置里填好启动命令和参数,保存后完全退出客户端再打开,配置才会生效。

Nb MCP Server 6a49aba0 收费吗?

本站不代收费用,也没有标注具体价格。是否收费取决于服务本身和背后的接口额度,正式使用前按官方说明确认。

Nb MCP Server 6a49aba0 需要 API Key 吗?

资料标注需要Token认证,使用前准备好对应凭据,并按最小权限配置。

Nb MCP Server 6a49aba0 支持哪些客户端?

资料标注支持 Claude Desktop、Claude、Cursor、Windsurf;其他客户端只要能配置 MCP Server,也能用同样的命令或地址接入。

Nb MCP Server 6a49aba0 能提供哪些能力?

资料显示约 32 个工具;能力标签包括 网络监控、TypeScript、Claude。具体工具清单以官方文档为准。

热门专题

按标签挑服务器
全部专题 →

服务器信息以作者仓库与官方文档为准;分类、传输方式与鉴权方式的中文说明为中文整理,最近核验 2026-05-17。