HR Solx MCP Server
Acts as a bridge between AI models and the HR Solx REST API, enabling natural language management of users, geographic data, and professional skills. It provides tools for performing health checks, retrieving reference data, and creating or listing user accounts.
README
HR Solx MCP Server
A Model Context Protocol (MCP) server that exposes HR API endpoints as AI-accessible tools. Built with Express.js and the official MCP SDK.
What It Does
This server acts as a bridge between AI models (Claude, GPT, etc.) and the HR Solx REST API. AI clients can discover and invoke tools through the MCP protocol, enabling natural language interactions with HR data like users, countries, skills, roles, and more.
Features
- 12 MCP Tools — Health checks, geographic data, reference data, and user management
- Streamable HTTP Transport — Stateless JSON-RPC 2.0 communication
- TypeScript — Full type safety with Zod validation
- Modular Architecture — Organized codebase with separated concerns
- Authentication — API key protection for MCP endpoint
- Rate Limiting — IP-based request throttling
- Error Handling — Typed errors with contextual messages
System Architecture
graph LR
AI["AI Client\n(Claude, GPT)"] -->|"POST /mcp\nJSON-RPC 2.0"| Express["Express Server\n(MCP Protocol)"]
Express -->|"fetch"| API["HR API\napi.hr-solx-mobile.com"]
API -->|"JSON Response"| Express
Express -->|"SSE Stream"| AI
subgraph Middleware
Express
end
classDef ai fill:#e1f5fe,stroke:#01579b
classDef server fill:#fff3e0,stroke:#e65100
classDef api fill:#e8f5e9,stroke:#2e7d32
class AI ai
class Express server
class API api
Quick Start
npm install
npm run dev
Server runs at http://localhost:4000/mcp
Environment Variables
| Variable | Default | Description |
|---|---|---|
MCP_SERVER_PORT |
4000 |
Server listening port |
MCP_API_URL |
https://api.hr-solx-mobile.com |
Upstream HR API URL |
MCP_API_KEY |
— | API key to protect MCP endpoint |
API_TOKEN |
— | Bearer token for upstream API auth |
RATE_LIMIT_WINDOW_MS |
900000 |
Rate limit window (15 min) |
RATE_LIMIT_MAX_REQUESTS |
100 |
Max requests per window |
Copy .env.example to get started:
cp .env.example .env
Available Tools
Health Checks
| Tool | Description |
|---|---|
basic-health-check |
Check if API is reachable |
detailed-health-check |
Comprehensive system health |
Geographic Data
| Tool | Description |
|---|---|
get-countries |
List all countries |
get-states |
List all states |
get-cities |
List all cities |
Reference Data
| Tool | Description |
|---|---|
get-skills |
List available skills |
get-languages |
List available languages |
get-working-statuses |
List working statuses |
get-roles |
List available roles |
User Management
| Tool | Description | Params |
|---|---|---|
get-users |
List all users | — |
create-user |
Create a new user | name, email, mobile |
Usage Examples
List Available Tools
curl -X POST http://localhost:4000/mcp \
-H "Content-Type: application/json" \
-H "Accept: application/json, text/event-stream" \
-d '{"jsonrpc":"2.0","method":"tools/list","params":{},"id":1}'
Call a Tool
curl -X POST http://localhost:4000/mcp \
-H "Content-Type: application/json" \
-H "Accept: application/json, text/event-stream" \
-d '{
"jsonrpc": "2.0",
"method": "tools/call",
"params": {
"name": "get-users",
"arguments": {}
},
"id": 1
}'
Create a User
curl -X POST http://localhost:4000/mcp \
-H "Content-Type: application/json" \
-H "Accept: application/json, text/event-stream" \
-d '{
"jsonrpc": "2.0",
"method": "tools/call",
"params": {
"name": "create-user",
"arguments": {
"name": "John Doe",
"email": "john@example.com",
"mobile": "+1234567890"
}
},
"id": 1
}'
With Authentication
curl -X POST http://localhost:4000/mcp \
-H "Content-Type: application/json" \
-H "Accept: application/json, text/event-stream" \
-H "X-API-Key: your-api-key" \
-d '{"jsonrpc":"2.0","method":"tools/list","params":{},"id":1}'
Project Structure
├── docs/ # Detailed documentation
│ ├── ARCHITECTURE.md # System architecture & design
│ ├── MCP-PROTOCOL.md # MCP protocol guide
│ ├── TOOLS-REFERENCE.md # Complete tool catalog
│ ├── DEVELOPER-GUIDE.md # How to extend & test
│ ├── SECURITY.md # Security considerations
│ ├── TROUBLESHOOTING.md # Common issues & fixes
│ └── REQUEST-FLOW.md # Request lifecycle
├── src/
│ ├── index.ts # Server entry point
│ ├── types/
│ │ ├── api.ts # API response interfaces
│ │ └── errors.ts # Custom error types
│ ├── client/
│ │ └── api-client.ts # Upstream API client
│ ├── tools/
│ │ ├── echo.ts # Echo tool/resource/prompt
│ │ ├── health.ts # Health check tools
│ │ ├── reference.ts # Reference data tools
│ │ └── users.ts # User management tools
│ └── middleware/
│ ├── auth.ts # API key authentication
│ └── rate-limit.ts # Rate limiting
├── .env.example # Environment template
├── package.json
└── tsconfig.json
Module Dependencies
graph TD
Index["src/index.ts\n(Server Entry)"] --> Types["src/types/\n(api.ts, errors.ts)"]
Index --> Client["src/client/\n(api-client.ts)"]
Index --> Tools["src/tools/\n(echo, health, reference, users)"]
Index --> Middleware["src/middleware/\n(auth, rate-limit)"]
Tools --> Client
Tools --> Types
Client --> Types
classDef entry fill:#fff3e0,stroke:#e65100
classDef module fill:#e3f2fd,stroke:#1565c0
class Index entry
class Types,Client,Tools,Middleware module
See docs/ARCHITECTURE.md for full details.
Adding New Tools
- Define the TypeScript interface in
src/types/api.ts - Register the tool in the appropriate module under
src/tools/ - Import and register in
src/index.ts
See docs/DEVELOPER-GUIDE.md for step-by-step instructions.
Security
The server supports two layers of authentication:
- MCP endpoint — Protected via
X-API-Keyheader (MCP_API_KEY) - Upstream API — Authenticated via Bearer token (
API_TOKEN)
See docs/SECURITY.md for full security guide and production checklist.
Troubleshooting
Common issues and solutions are documented in docs/TROUBLESHOOTING.md.
Quick checks:
# Verify server is running
curl http://localhost:4000/mcp
# Test upstream API
curl https://api.hr-solx-mobile.com/health
# Check environment variables
echo $MCP_API_URL
echo $MCP_SERVER_PORT
License
MIT
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。