内容
# Lambda Capture MCP Server
[](https://lambda-capture.com/) [](https://lambda-capture.com/)
MCP 实现了我们的标准 [宏观经济数据的语义搜索 API](https://github.com/lambda-capture/Semantic-Search-API)
## 远程 MCP Server (可流式传输的 HTTP)
检查服务器状态 [HERE](https://mcp.lambda-capture.com/)
### [OpenAI Responses API](https://platform.openai.com/docs/guides/tools-remote-mcp)
```python
from openai import OpenAI
client = OpenAI()
resp = client.responses.create(
model="gpt-4.1",
input="通货膨胀预期的关键变化",
tools=[
{
"type": "mcp",
"server_label": "lambda-capture",
"server_url": "https://mcp.lambda-capture.com/v1/mcp/",
"headers": {
"Authorization": "Bearer YOUR_ACCESS_TOKEN"
}
}
]
)
print(resp.output_text)
```
### Curl
```bash
curl -X POST "https://mcp.lambda-capture.com/v1/mcp/" \
-H "Content-Type: application/json" \
-H "Accept: application/json, text/event-stream" \
-H "Authorization: Bearer YOUR_ACCESS_TOKEN" \
-d '{
"jsonrpc": "2.0",
"method": "tools/call",
"id": 1,
"params": {
"name": "macroecon_semantic_search",
"arguments": {
"query_text": "inflation expectations",
"max_results": 3
}
}
}'
```
```bash
curl -X POST "https://mcp.lambda-capture.com/v1/mcp/" \
-H "Content-Type: application/json" \
-H "Accept: application/json, text/event-stream" \
-H "Authorization: Bearer YOUR_ACCESS_TOKEN" \
-d '{
"jsonrpc": "2.0",
"id": "1",
"method": "list_tools",
"params": {}
}'
```
### 配置你的 MCP 客户端 (Claude 桌面应用)
前往 Claude -> 设置 -> 开发者 -> 编辑配置。将以下内容添加到你的 `claude_desktop_config.json` 中。
#### 节点:
```json
{
"mcpServers": {
"lambda-capture-mcp": {
"command": "npx",
"args": [
"mcp-remote",
"https://mcp.lambda-capture.com/v1/mcp/",
"--header", "Authorization: Bearer YOUR_ACCESS_TOKEN"
],
"description": "使用 Lambda Capture 宏观经济数据 API 的 RemoteMCP"
}
}
}
```
## 本地 MCP Server
### 先决条件
- [Lambda Capture API 密钥](https://lambda-capture.com/)
- 对于 MCP Typescript: [Node.js 18+（包括 npx 和 npm）](https://nodejs.org/en/download/)
- 对于 MCP Python: [Python 3.11+](https://www.python.org/downloads/)
### 安装
1. 克隆仓库
#### 节点：
2. `npm install` 安装依赖
3. `npm run build` 构建项目
#### Python:
2. `python -m venv .venv` 创建虚拟环境
3. `source .venv/bin/activate` 激活虚拟环境
4. `pip install -r requirements.txt` 安装依赖项
### 配置你的 MCP 客户端 (Claude 桌面应用)
前往 Claude -> 设置 -> 开发者 -> 编辑配置。将以下内容添加到你的 `claude_desktop_config.json` 中。
#### 节点:
```json
{
"mcpServers": {
"lambda-capture-mcp": {
"command": "node",
"args": [
"/绝对路径到/mcp-server/dist/index.js"
],
"env": {
"LAMBDA_CAPTURE_API_KEY": "您的 API 密钥字符串"
},
"description": "运行 Node MCP 与 Lambda Capture 宏观经济数据 API"
}
}
}
```
#### Python:
```json
{
"mcpServers": {
"lambda-capture-mcp": {
"command": "/绝对路径到/.venv/bin/python",
"args": [
"/绝对路径到/mcp-server/main.py"
],
"env": {
"LAMBDA_CAPTURE_API_KEY": "您的 API 密钥字符串"
},
"description": "运行 Python MCP，使用 Lambda Capture 宏观经济数据 API"
}
}
}
```
### 上下文窗口大小
根据模型的上下文窗口大小调整 `maxTokens` (.ts) 或 `max_tokens` (.py) 变量（不计算元数据，仅计算内容令牌）
© 2025 Lambda Capture Limited (注册号 15845351) 52 Tabernacle Street, London, EC2A 4NJ - 保留所有权利
MCP 市场