DB Query MCP

Q: What is db-query-mcp?

A: It's a versatile database query and export tool (MCP) that supports multiple databases, converts natural language to SQL, and offers secure read-only access by default. Key features include compatibility with ElasticSearch, MySQL, PostgreSQL, and more, plus CSV/JSON export options.

Q: How do I install db-query-mcp?

A:

• Basic install:

  pip install db-query-mcp
  

• For ElasticSearch:

  pip install db-query-mcp[elasticsearch]
  

• From GitHub:

  pip install git+https://github.com/NewToolAI/db-query-mcp
  

Q: What databases are supported?

A:

• SQLite (built-in)
• MySQL (requires pymysql)
• PostgreSQL (requires psycopg2-binary)
• Oracle (requires cx_Oracle)
• SQL Server (requires pyodbc or pymssql)
• ElasticSearch (added in v2025-06-02)

Q: How do I configure db-query-mcp for different databases?

A: Use a JSON config file. Examples:

{
  "mcpServers": {
    "sqlite_db_mcp": {
      "command": "db-query-mcp",
      "args": [
        "--db",
        "sqlite",
        "--uri",
        "sqlite:///sqlite_company.db",
      ]
    },
    "es_db_mcp": {
      "command": "db-query-mcp",
      "args": [
        "--db",
        "elasticsearch",
        "--uri",
        "https://user:password@localhost:9200?index=test_data_index&ca_certs=/home/user/http_ca.crt",
      ]
    }
  }
}

Q: Is db-query-mcp secure?

A: Yes! It defaults to read-only mode to protect your data.

Q: Can I export query results?

A: Absolutely! Export data to CSV or JSON formats directly from the tool.

Q: What’s next for db-query-mcp?

A: Future updates will add support for MongoDB and GraphDatabase, aiming to become a full-stack DB query MCP.

Q: How do I contribute?

A: PRs are welcome! Check the GitHub repo:
🔗 https://github.com/NewToolAI/db-query-mcp

Q: What’s the license?

A: Apache 2.0 (open-source and free to use).

Need more help? Open an issue on GitHub! 🚀