Employment Data MCP
Provides read-only FastMCP tools for BLS nonfarm employment, ADP private employment, economic calendar, and employment release summaries.
README
Employment Data MCP
面向 Hermes 的只读 FastMCP 服务,提供六个工具:
get_bls_employment:BLS 非农就业总人数、月新增、失业率及实际发布日期;get_adp_employment:ADP 私营就业总人数、月新增及实际发布日期;get_employment_release_schedule:未来 BLS、ADP 发布日期及美东/中国时间;get_latest_employment_actual_vs_forecast:最近两次非农和 ADP 实际值、预期值对比;get_economic_calendar:指定日期的实际值、预期值、前值和 surprise;get_employment_release_summary:指定发布日期的 ADP、非农和失业率摘要。
Docker 部署(推荐)
cp .env.example .env
# 编辑 .env,至少设置 BLS_API_KEY 和 FRED_API_KEY
docker compose up -d --build
docker compose logs -f employment-data-mcp
日志除 Uvicorn access log 外,还会输出 JSON 格式的 MCP 审计事件。例如:
{"event":"mcp_request_start","request_id":"7c17f8cf982a","started_at":"2026-08-10T06:30:00.123+00:00","method":"tools/call","message_type":"request","tool":"get_adp_employment","arguments":{"start_year":2025,"end_year":2026}}
{"event":"mcp_request_success","request_id":"7c17f8cf982a","method":"tools/call","duration_ms":1240.51,"tool":"get_adp_employment","result_count":18}
tools/list、initialize 等协议请求也会记录。日志时间为 UTC;参数中的 key、token、
password、authorization、proxy、secret 等敏感字段会自动脱敏。使用同一个
request_id 可配对开始、完成或错误事件。
服务每次启动还会输出 {"event":"mcp_audit_ready",...}。如果看不到该行,
说明当前容器仍在运行旧镜像;用 docker compose up -d --build --force-recreate
重建并替换容器。
默认 MCP 地址是 http://127.0.0.1:8010/mcp,仅监听宿主机回环地址,
不会直接暴露到公网。
如果代理运行在 Docker 宿主机的 11111 端口,容器内不能使用
127.0.0.1:11111,请配置:
FINANCE_PROXY_URL=http://host.docker.internal:11111
compose.yaml 已配置 Linux 所需的 host-gateway。不需要代理时将该变量留空。
Hermes 配置
在 Hermes 的 config.yaml 中加入:
mcp_servers:
us_employment:
url: "http://127.0.0.1:8010/mcp"
enabled: true
timeout: 120
supports_parallel_tool_calls: false
不配置 tools.include 时,Hermes 会自动发现服务端当前所有工具,后续新增工具无需手动
维护白名单。
然后执行 hermes mcp test us_employment 验证连接;已经启动的 Hermes 会话中可用
/reload-mcp 重新加载。若 Hermes 自身也在容器中,127.0.0.1 指向 Hermes
容器,应把两个服务接入同一 Docker 网络,并将 URL 改为
http://employment-data-mcp:8010/mcp。
不使用 Docker
服务器已有 Python 3.11+ 和 uv 时:
uv sync --frozen
uv run employment-data-mcp
依赖安装在本项目的 .venv,不污染 Hermes 的 Python 环境。修改监听配置可设置
MCP_HOST、MCP_PORT 和 MCP_PATH。
开发验证
uv sync --extra dev
uv run pytest
API key 和代理地址只放在 .env,不要提交到仓库。经济日历预期值来自第三方日历源,
返回结果中会保留 source;BLS 与 ADP 历史序列可能随官方修订发生变化。
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.