Cover image
Try Now
2025-04-14

https://www.claudemcp.com/docs/quickstart

3 years

Works with Finder

0

Github Watches

0

Github Forks

0

Github Stars

🧠 MCP Quickstart: Claude + SQLite + Docker

TL;DR: Build a secure local Claude brain using Docker, SQLite, and uvx.


🛡️ Step 1: Build & Run the Docker SQLite Server

docker-compose build
docker-compose up

Your server will be live at http://localhost:8080, ready to accept MCP-style queries.


🧠 Step 2: Install Astral (uv / uvx)

Astral is the MCP-compatible package manager.

curl -Ls https://astral.sh/uv/install.sh | sh

Verify it:

uvx mcp-server-sqlite --help

🔧 Step 3: Make uvx Available to Claude Desktop

Claude launched via Spotlight/Dock doesn’t see your shell config. So we have to tell macOS:

launchctl setenv PATH "$HOME/.local/bin:/usr/local/bin:/usr/bin:/bin:/usr/sbin:/sbin"

You can also verify this:

launchctl getenv PATH

🔗 Add Symlink for uv (required by uvx)

Claude expects uv at /usr/local/bin/uv, but Astral installs it in ~/.local/bin/uv. Fix that with:

sudo ln -s $HOME/.local/bin/uv /usr/local/bin/uv

Verify:

uvx mcp-server-sqlite --help

⚙️ Step 4: Configure Claude to Use Your SQLite DB

Create or edit:

~/Library/Application Support/Claude/claude_desktop_config.json

Insert:

{
  "mcpServers": {
    "sqlite": {
      "command": "uvx",
      "args": [
        "mcp-server-sqlite",
        "--db-path",
        "/Users/YOUR_USERNAME/Desktop/mcp_quickstart/data/test.db"
      ]
    }
  }
}

✅ Make sure to replace YOUR_USERNAME with your actual username.


🧪 Step 5: Test It Manually

Try this from the terminal:

echo '{"jsonrpc":"2.0","id":"debug","method":"query","params":{"sql":"SELECT * FROM products LIMIT 1"}}' \
  | uvx mcp-server-sqlite --db-path /Users/YOUR_USERNAME/Desktop/mcp_quickstart/data/test.db

You should get back:

{
  "jsonrpc": "2.0",
  "id": "debug",
  "result": {
    "columns": ["id", "name", "price"],
    "rows": [[1, "Widget", 19.99]]
  }
}

🧰 Troubleshooting

🧹 Check if the DB is accessible

docker exec -it $(docker ps --filter "name=mcp-server" -q) sqlite3 /data/test.db

Then run:

SELECT * FROM products LIMIT 1;

🔄 Rebuild the Docker container

docker-compose down -v && docker-compose build && docker-compose up

💬 Bonus: Simulate Claude Queries with a Test JSON

echo '{"jsonrpc":"2.0","id":"test123","method":"query","params":{"sql":"SELECT * FROM products"}}' > test-mcp.json

cat test-mcp.json | uvx mcp-server-sqlite --db-path /Users/YOUR_USERNAME/Desktop/mcp_quickstart/data/test.db

📚 Resources


✅ You now have:

  • 🐻 Dockerized SQLite server
  • 🧪 Secure query interface via FastAPI + Uvicorn
  • 📡 Claude MCP bridge using uvx + mcp-server-sqlite
  • 🧠 A local Claude brain that talks SQL

相关推荐

  • av
  • Exécutez sans effort LLM Backends, API, Frontends et Services avec une seule commande.

  • 1Panel-dev
  • 🔥 1Panel fournit une interface Web intuitive et un serveur MCP pour gérer des sites Web, des fichiers, des conteneurs, des bases de données et des LLM sur un serveur Linux.

  • rulego
  • ⛓️RULEGO est un cadre de moteur de règle d'orchestration des composants de nouvelle génération légère, intégrée, intégrée et de nouvelle génération pour GO.

  • WangRongsheng
  • 🧑‍🚀 全世界最好的 LLM 资料总结 (数据处理、模型训练、模型部署、 O1 模型、 MCP 、小语言模型、视觉语言模型) | Résumé des meilleures ressources LLM du monde.

  • Byaidu
  • PDF Traduction de papier scientifique avec formats conservés - 基于 AI 完整保留排版的 PDF 文档全文双语翻译 , 支持 Google / Deepl / Olllama / Openai 等服务 , 提供 CLI / GUI / MCP / DOCKER / ZOTERO

  • hkr04
  • SDK C ++ MCP (Protocole de contexte modèle léger)

  • sigoden
  • Créez facilement des outils et des agents LLM à l'aide de fonctions Plain Bash / JavaScript / Python.

  • RockChinQ
  • 😎简单易用、🧩丰富生态 - 大模型原生即时通信机器人平台 | 适配 QQ / 微信 (企业微信、个人微信) / 飞书 / 钉钉 / Discord / Telegram / Slack 等平台 | 支持 Chatgpt 、 Deepseek 、 Dify 、 Claude 、 GEMINI 、 XAI 、 PPIO 、 OLLAMA 、 LM Studio 、阿里云百炼、火山方舟、 Siliconflow 、 Qwen 、 Moonshot 、 ChatGlm 、 Sillytraven 、 MCP 等 LLM 的机器人 / Agent | Plateforme de bots de messagerie instantanée basés sur LLM, prend en charge Discord, Telegram, WeChat, Lark, Dingtalk, QQ, Slack

  • dmayboroda
  • Rag de conversation sur site avec des conteneurs configurables

  • modelscope
  • Commencez à construire des applications multi-agents LLM, plus facilement.

    Reviews

    1 (1)
    Avatar
    user_QyvSChbl
    2025-04-24

    As a dedicated user of mcp_quickstart, I have to say that this tool has exceeded my expectations. Created by chrisjmendez, it offers a seamless and efficient solution for quick project setup. The user-friendly interface and thorough documentation make it approachable even for beginners. If you’re looking for a reliable starting point for your projects, mcp_quickstart is definitely worth a try.