mcp .sqlite

This MCP server provides a SQLite-based SQL execution environment accessible via the MCP protocol. It exposes tools such as list_tables to enumerate all tables in the SQLite database and read_query to execute SELECT statements and retrieve results. The server runs inside a Docker image (mcp/sqlite) and can be integrated into your MCP client by configuring the mcpServers section with the docker command and the appropriate volume mount and database path. Once connected, you can send MCP actions to list tables or run queries like SELECT * FROM users to synthesize insights from your data. The tooling is designed to be interactive, allowing iterative querying and schema inspection within the MCP framework.

Prerequisites

• Docker installed and running on your machine
• Basic familiarity with MCP client configuration

Installation steps

1. Clone the repository (or obtain the Docker image): git clone https://github.com/javsanmar5/mcp-server.sqlite.git cd mcp-server.sqlite

2. Build the Docker image locally: docker build -t mcp/sqlite .

3. Configure your MCP client to connect to the SQLite MCP server by adding the following to its configuration (example shows the docker-based setup): { "mcpServers": { "sqlite": { "command": "docker", "args": [ "run", "--rm", "-i", "-v", "mcp-test:/mcp", "mcp/sqlite", "--db-path", "test_db.sqlite3" ] } } }

4. Restart your MCP client to load the new server configuration.

Notes

• The configuration uses a Docker-based deployment. If you prefer not to build locally, ensure you pull the mcp/sqlite image from a registry compatible with your setup.
• Ensure the path to the SQLite database is set correctly via --db-path (test_db.sqlite3 in the example).

Optional If you already have a running Docker environment, you can skip the build step and pull the prebuilt image (if available) and use the same docker run command in your client configuration.

Tips and caveats:

• The Docker command mounts a named volume (mcp-test) at /mcp inside the container to store state; ensure the volume has appropriate permissions.
• If you’re using Windows, you may need to adjust line endings or CLI quoting; the container run command should be executed as part of the MCP client configuration rather than manually in some clients.
• The MCP tools exposed by this server primarily include list_tables and read_query for interacting with the SQLite database. You can extend queries by sending standard SQL through read_query.
• If you encounter network or container lifecycle issues, restarting the MCP client or recreating the Docker container can help. Make sure the container has access to the database path specified by --db-path.
• Since this server is Docker-based in the repository, there is no npm package published for this server at the moment.