japan-healthcare-mcp
A local-first MCP server and toolkit for Japanese healthcare reference data, enabling lookup and search without bundling official datasets.
README
japan-healthcare-mcp
Japan Healthcare MCP is a local-first toolkit and MCP server for Japanese healthcare reference data.
The repository includes TypeScript workspace packages for the core library, CLI, MCP server, optional REST API, synthetic sample data, and local demo workflows. It does not bundle official production datasets.
Japanese documentation is available in README.ja.md.
Goals
- Provide a local-first workspace for working with Japanese healthcare reference data.
- Keep source attribution and data provenance visible in search results.
- Offer an MCP server interface for reference-data lookup tools.
- Avoid bundling official datasets unless their terms explicitly allow redistribution.
Requirements
- Node.js 22 LTS or newer
- pnpm 10 or newer
- Corepack, recommended for activating the pinned pnpm version
Initial setup
corepack enable
corepack prepare pnpm@10.12.4 --activate
pnpm install
pnpm check:prerequisites
Development commands
pnpm lint
pnpm format:check
pnpm typecheck
pnpm test
pnpm build
Docker demo
The Docker demo builds the MCP server image and prepares a SQLite database from bundled synthetic samples:
docker compose build
docker compose run --rm -T mcp-server
See docs/docker-demo.md for Docker, docker-compose, MCP Inspector, demo DB, and terminal recording details.
Development baseline
- The project uses pnpm as the package manager.
- The root
package.jsonrecords the expected package manager and Node.js engine range. pnpm-workspace.yamlis prepared for packages underpackages/*.- Generated files such as
node_modules/, build output, coverage output, local.envfiles, and local SQLite databases are ignored.
For more details, see docs/development-environment.md.
Monorepo packages
The workspace is organized into TypeScript packages:
packages/core— core data models, importers, search, source attribution, and safety response modules.packages/cli—jhccommand line interface for local database initialization, imports, searches, diffs, and MCP startup.packages/mcp-server— MCP server and healthcare reference data tools.packages/api— optional REST API server and OpenAPI schema support.
OSS operations
- Security policy:
SECURITY.md - Contribution guide:
CONTRIBUTING.md - Code of conduct:
CODE_OF_CONDUCT.md - Changelog:
CHANGELOG.md - Repository topics and settings notes:
docs/repository-settings.md - v0.1.0 release notes:
docs/release-notes/v0.1.0.md
Suggested GitHub topics are mcp, model-context-protocol, healthcare, japan, japanese-healthcare, reference-data, sqlite, typescript, drug-master, reimbursement, and pmda.
Data policy status
Official healthcare datasets are not bundled. The included sample CSV files and Docker demo database contain fictional synthetic data only.
Future data handling work must preserve source attribution and avoid redistributing official datasets unless their license explicitly allows it.
See NOTICE for the repository-level notice and official data redistribution policy.
Disclaimer
This project is for reference data tooling only. Do not rely on it as the only source for regulated or professional decisions.
See DISCLAIMER.md for the project disclaimer.
License
This project is licensed under the Apache License, Version 2.0. See LICENSE for details.
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。