格拉法纳MCP分析器
项目介绍
Grafana MCP Analyzer 🤖
让AI直接理解您的监控数据，并成为您专属的智能运维助手！
✨ 项目简介
想象这样一个场景：
- 您问AI：“我的服务器现在运行得怎么样？”
- AI直接检查您的Grafana监控并回答：“CPU使用率较高，建议检查这些进程。”
复杂的监控图表，AI辅助一键解读。您不再需要逐屏查看图像，真正实现从图表到洞察的闭环分析体验！
🚀 核心特性
Grafana MCP Analyzer基于**MCP（模型上下文协议）**协议，赋予像Claude和ChatGPT这样的AI助手以下超级能力：
- 自然语言查询 -轻松访问监控数据，AI一键输出专业分析
- 多轮对话支持 -支持复杂多轮对话分析，能够根据上下文进行深入分析
- curl支持 -通过直接从浏览器复制的curl合成查询
- 全数据源支持 -Prometheus、MySQL、ES等都支持
- 专业DevOps建议 -不仅展示数据，还提供可执行的优化解决方案以提高DevOps效率
💡 新架构模型会话缓存 → 逐步数据获取 → 逐步深入分析 → 缓存复用，使AI分析更准确高效。
🛠️ 快速开始
第一步：快速安装（30秒）
npm install -g grafana-mcp-analyzer
环境要求：Node.js 18+ | 安装指南
第二步：集成AI助手（30秒）
光标设置 → "MCP" → 服务配置（以光标为例）：
{
"mcpServers": {
"grafana": {
"command": "grafana-mcp-analyzer",
"env": {
"CONFIG_PATH": "https://raw.githubusercontent.com/SailingCoder/grafana-mcp-analyzer/main/config/grafana-config-play.js",
"MAX_CHUNK_SIZE": "100"
}
}
}
}
注意：CONFIG_PATH支持绝对路径和远程路径，推荐使用远程路径进行快速体验。
第三步：编写配置文件（1分钟）
如果您需要连接自己的数据，可以在CONFIG_PATH路径下创建一个配置文件：（示例为grafana-config-play.exe）👉 点击这里查看
<details> <summary>点击展开查看示例</summary>如果您只想快速体验示例，可以跳过此步骤，直接进入第四步。
/**
* 基于Grafana Play演示实例的配置文件
* 以下配置文件内容来源：https://raw.githubusercontent.com/SailingCoder/grafana-mcp-analyzer/main/config/grafana-config-play.js
* 请求配置方式：支持 http api 和 curl
*/
const config = {
// Grafana服务器地址
baseUrl: 'https://play.grafana.org',
// 默认请求头
defaultHeaders: {
'Content-Type': 'application/json',
'Accept': 'application/json, text/plain, */*'
},
// 健康检查配置
healthCheck: {
url: 'api/health'
},
// 查询定义
queries: {
// Price Only, Hollow Candles
// 使用HTTP API格式
// 数据源：https://play.grafana.org/d/candlestick/candlestick?orgId=1&from=2021-07-13T22:13:30.740Z&to=2021-07-13T22:46:18.921Z&timezone=utc&viewPanel=panel-7
candlestick_priceOnly_hollowCandles: {
url: 'api/ds/query',
method: 'POST',
params: {
ds_type: 'grafana-testdata-datasource',
requestId: 'SQR279'
},
headers: {
'accept': 'application/json, text/plain, */*',
'accept-language': 'zh-CN,zh;q=0.9',
'cache-control': 'no-cache',
'content-type': 'application/json',
'origin': 'https://play.grafana.org',
'pragma': 'no-cache',
'priority': 'u=1, i',
'referer': 'https://play.grafana.org/d/candlestick/candlestick?orgId=1&from=2021-07-13T22:13:30.740Z&to=2021-07-13T22:46:18.921Z&timezone=utc&viewPanel=panel-7',
'sec-ch-ua': '"Not)A;Brand";v="8", "Chromium";v="138", "Google Chrome";v="138"',
'sec-ch-ua-mobile': '?0',
'sec-ch-ua-platform': '"macOS"',
'sec-fetch-dest': 'empty',
'sec-fetch-mode': 'cors',
'sec-fetch-site': 'same-origin',
'traceparent': '00-f0f1243b82acf0e362fd1f836565154a-fc3a173d3190c9df-01',
'user-agent': 'Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/138.0.0.0 Safari/537.36',
'x-dashboard-title': 'Candlestick',
'x-dashboard-uid': 'candlestick',
'x-datasource-uid': 'PD8C576611E62080A',
'x-grafana-device-id': '49c7d4ecdeee88ab5dde64deffa8ea2e',
'x-grafana-org-id': '1',
'x-panel-id': '7',
'x-panel-plugin-id': 'candlestick',
'x-panel-title': 'Price Only, Hollow Candles',
'x-plugin-id': 'grafana-testdata-datasource'
},
data: {
queries: [{
csvFileName: "ohlc_dogecoin.csv",
refId: "A",
scenarioId: "csv_file",
datasource: {
type: "grafana-testdata-datasource",
uid: "PD8C576611E62080A"
},
datasourceId: 454,
intervalMs: 2000,
maxDataPoints: 1180
}],
from: "1626214410740",
to: "1626216378921"
},
systemPrompt: `您是狗狗币K线图分析专家。
**分析重点**：
1. 价格趋势识别 - 识别主要趋势方向(上涨/下跌/横盘)
2. 关键价位分析 - 找出支撑位和阻力位
3. 交易机会评估 - 基于K线形态识别入场时机
4. 风险评估 - 提供风险提示和投资建议
**输出格式**：
## 图表概览
- 时间范围：[具体时间]
- 价格范围：[最高价-最低价]
- 主要趋势：[上涨/下跌/横盘]
## 技术分析
- 支撑位：[价格水平]
- 阻力位：[价格水平]
- 关键行为：[重要价格行为]
## 交易建议
- 短期方向：[看涨/看跌/中性]
- 关键价位：[关注价位]
- 风险提示：[重要提醒]`
},
// faro-shop-control-plane - Overall CPU Utilization
// 使用 cUrl 格式
// 数据源：https://play.grafana.org/d/cNMLIAFK/cpu-utilization-details-cores?var-interval=$__auto&orgId=1&from=now-3h&to=now&timezone=browser&var-host=faro-shop-control-plane&var-cpu=$__all&viewPanel=panel-22
overall_cpu_utilization: {
curl: `curl 'https://play.grafana.org/api/ds/query?ds_type=prometheus&requestId=SQR112' \
-H 'accept: application/json, text/plain, */*' \
-H 'accept-language: zh-CN,zh;q=0.9' \
-H 'cache-control: no-cache' \
-H 'content-type: application/json' \
-b '_ga=GA1.2.1909983567.1753671369; _gid=GA1.2.532774264.1753671369; rl_page_init_referrer=RudderEncrypt%3AU2FsdGVkX1%2B2lASJjXBqxv6%2FOpvlv5ClRT5vw%2BELHuE%3D; rl_page_init_referring_domain=RudderEncrypt%3AU2FsdGVkX19MSXh%2BQbiHW5f9mLAaP3ghy%2FcJZIk9zhI%3D; intercom-id-agpb1wfw=219eac14-cc23-4ca5-aa16-c299fab8c0ab; intercom-session-agpb1wfw=; intercom-device-id-agpb1wfw=fd9a6df6-d6c8-4b40-958b-568fc7f30ae2; rl_group_id=RudderEncrypt%3AU2FsdGVkX196IBi0ppflecKuY9333Hf3E8fCWy4xJNU%3D; rl_group_trait=RudderEncrypt%3AU2FsdGVkX19%2Fc4msmFb6pg0d4rM%2BpLKI9zqEnxxFrPE%3D; rl_anonymous_id=RudderEncrypt%3AU2FsdGVkX186iymdvmvCOhwF2sff5XEHniCdK0idYHYA4P%2BUpg8hnPVqFbQpqF%2Fn5dfeDz3BxORb9hPn8cIvwQ%3D%3D; rl_user_id=RudderEncrypt%3AU2FsdGVkX1%2B7qEm%2BjVUpWQfQIZgdXaAXNAGDqx%2ByBo3qzXCeyxQWfQNHP9CFM4cX; rl_trait=RudderEncrypt%3AU2FsdGVkX19zSSOXFUxzg3KWR6VQOAkavGgxHg9JdbDKn6hPh3%2BBm3nDBP%2F6tM0wl0b6r0f1A2MZ2SeB6p9f%2FeeaUcrUzR%2FQDfqJHZGhOCdpwmOXZVVQncG%2Ff3ITY6GU%2BvGu9sfYHNgcpS5UHphpBA%3D%3D; _ga_Y0HRZEVBCW=GS2.2.s1753671369$o1$g1$t1753671728$j23$l0$h0; rl_session=RudderEncrypt%3AU2FsdGVkX18BkXGTwuY7KtE7Zr6WjpDFDtkvh9%2Btz4dc8BJeXT1%2FrqgdzGnXydN9EMwRRVR%2FQzGVBtyZ%2FNhg27pvhkbqL2QVLD%2F79GRtbxM8qDKCDo4c%2FfokCEdeF8AoiuRXQzPkAC7UEy7g1swC9w%3D%3D' \
-H 'origin: https://play.grafana.org' \
-H 'pragma: no-cache' \
-H 'priority: u=1, i' \
-H 'referer: https://play.grafana.org/d/cNMLIAFK/cpu-utilization-details-cores?var-interval=$__auto&orgId=1&from=now-3h&to=now&timezone=browser&var-host=faro-shop-control-plane&var-cpu=$__all&viewPanel=panel-22&inspect=panel-22&inspectTab=query' \
-H 'sec-ch-ua: "Not)A;Brand";v="8", "Chromium";v="138", "Google Chrome";v="138"' \
-H 'sec-ch-ua-mobile: ?0' \
-H 'sec-ch-ua-platform: "macOS"' \
-H 'sec-fetch-dest: empty' \
-H 'sec-fetch-mode: cors' \
-H 'sec-fetch-site: same-origin' \
-H 'traceparent: 00-fea7a897de47671f57a42d15b26043a5-578babdc8cb152e0-01' \
-H 'user-agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/138.0.0.0 Safari/537.36' \
-H 'x-dashboard-title: CPU Utilization Details (Cores)' \
-H 'x-dashboard-uid: cNMLIAFK' \
-H 'x-datasource-uid: grafanacloud-prom' \
-H 'x-grafana-device-id: 49c7d4ecdeee88ab5dde64deffa8ea2e' \
-H 'x-grafana-org-id: 1' \
-H 'x-panel-id: 22' \
-H 'x-panel-plugin-id: timeseries' \
-H 'x-panel-title: $host - Overall CPU Utilization' \
-H 'x-plugin-id: prometheus' \
--data-raw $'{"queries":[{"calculatedInterval":"2s","datasource":{"type":"prometheus","uid":"grafanacloud-prom"},"datasourceErrors":{},"errors":{},"expr":"clamp_max((avg by (mode) ( (clamp_max(rate(node_cpu_seconds_total{instance=\\"faro-shop-control-plane\\",mode\u0021=\\"idle\\"}[1m]),1)) or (clamp_max(irate(node_cpu_seconds_total{instance=\\"faro-shop-control-plane\\",mode\u0021=\\"idle\\"}[5m]),1)) )),1)","format":"time_series","hide":false,"interval":"1m","intervalFactor":1,"legendFormat":"{{mode}}","metric":"","refId":"A","step":300,"exemplar":false,"requestId":"22A","utcOffsetSec":28800,"scopes":[],"adhocFilters":[],"datasourceId":171,"intervalMs":60000,"maxDataPoints":1180},{"datasource":{"type":"prometheus","uid":"grafanacloud-prom"},"expr":"clamp_max(max by () (sum by (cpu) ( (clamp_max(rate(node_cpu_seconds_total{instance=\\"faro-shop-control-plane\\",mode\u0021=\\"idle\\",mode\u0021=\\"iowait\\"}[1m]),1)) or (clamp_max(irate(node_cpu_seconds_total{instance=\\"faro-shop-control-plane\\",mode\u0021=\\"idle\\",mode\u0021=\\"iowait\\"}[5m]),1)) )),1)","format":"time_series","hide":false,"interval":"1m","intervalFactor":1,"legendFormat":"Max Core Utilization","refId":"B","exemplar":false,"requestId":"22B","utcOffsetSec":28800,"scopes":[],"adhocFilters":[],"datasourceId":171,"intervalMs":60000,"maxDataPoints":1180}],"from":"1753660994019","to":"1753671794019"}'`,
systemPrompt: `您是系统性能分析专家，专注于CPU使用率历史趋势分析。
**数据特点**：这是总体CPU使用率的历史时间序列数据，包含：
- **user**: 用户模式CPU使用率
- **system**: 系统模式CPU使用率
- **iowait**: I/O等待时间
- **softirq**: 软件中断
- **Max Core Utilization**: 单核最大使用率
**分析重点**：
1. **历史趋势分析** - 识别CPU使用率的变化趋势和模式
2. **性能瓶颈识别** - 分析哪个CPU模式占用最多资源
3. **峰值分析** - 识别CPU使用率的峰值时间和原因
4. **系统健康评估** - 基于历史数据