Elasticsearch MCP Server
by jesee
README.md
# Elasticsearch MCP Server
Elasticsearch MCP 服务器,支持 Claude Code、VSCode (Cline)、Cursor 等工具。
## 安装
```bash
npm install -g es-mcp-server
```
## 配置
### 环境变量
| 变量 | 说明 | 默认值 |
|------|------|--------|
| `ES_HOST` | ES 地址 | `localhost:9200` |
| `ES_USERNAME` | 用户名 | - |
| `ES_PASSWORD` | 密码 | - |
| `ES_API_KEY` | API Key | - |
### Claude Code 配置
```json
{
"mcpServers": {
"elasticsearch": {
"command": "es-mcp-server",
"env": {
"ES_HOST": "localhost:9200"
}
}
}
}
```
### VSCode (Cline) 配置
路径: `~/.config/Code/User/globalStorage/hybridtalentcomputing.cline-chinese/settings/cline_chinese_mcp_settings.json`
```json
{
"mcpServers": {
"elasticsearch": {
"command": "es-mcp-server",
"env": {
"ES_HOST": "localhost:9200"
}
}
}
}
```
### Cursor 配置
```json
{
"mcpServers": {
"elasticsearch": {
"command": "es-mcp-server",
"env": {
"ES_HOST": "localhost:9200"
}
}
}
}
```
## 可用工具
| 工具 | 说明 |
|------|------|
| `es_health_check` | 集群健康检查 |
| `es_list_indices` | 列出所有索引 |
| `es_create_index` | 创建索引 |
| `es_delete_index` | 删除索引 |
| `es_index_document` | 索引文档 |
| `es_get_document` | 获取文档 |
| `es_search` | 搜索文档 |
| `es_update_document` | 更新文档 |
| `es_delete_document` | 删除文档 |
| `es_cluster_info` | 集群信息 |
## 使用示例
### 搜索文档
```json
{
"index": "user_info",
"query": {
"match": { "content": "总监" }
},
"size": 10
}
```
### 索引文档
```json
{
"index": "products",
"document": {
"name": "iPhone",
"price": 999
}
}
```
## 许可证
MIT
TDQS
B3.1/5.0
Scored across 10 tools
Disambiguation5/5
Each tool has a distinct, non-overlapping purpose: cluster info/health, index CRUD, document CRUD, listing, and search. No ambiguity exists.
Naming Consistency5/5
All tools follow a consistent 'es_verb_noun' pattern in snake_case, making them predictable and easy to understand (e.g., es_create_index, es_delete_document).
Tool Count5/5
With 10 tools, the set is well-scoped for basic Elasticsearch operations. Each tool serves a specific function, and the count is neither too few nor excessive.
Completeness4/5
The tools cover essential CRUD for indices and documents, plus cluster health and search. Missing operations like bulk indexing or mapping updates are minor gaps, but core workflows are supported.