The DuckDB / MotherDuck Local MCP Server connects AI assistants to various data sources using DuckDB's SQL engine, allowing for SQL read- and write-queries and database management.
From the registry: SQL analytics and data engineering for AI Assistants and IDEs

Please install the `mcp-server-motherduck` MCP server into my current AI client (that's you).
Required prerequisites (do these first if not already done):
- **Install uv** — Install uv Python package manager. Run: `pip install uv` (https://docs.astral.sh/uv/)
Canonical MCP server config (stdio transport):
- command: `uvx`
- args: ["mcp-server-motherduck","--db-path",":memory:","--read-write","--allow-switch-databases"]
- optional environment variables:
- `motherduck_token`: MotherDuck access token. Required for MotherDuck connections (md:). (example: `<your-motherduck-token>`)
Note: Default config uses in-memory DuckDB. For MotherDuck, use --db-path md: and set motherduck_token env var.
Add this MCP server to my current client's config in the correct format for you. If you need secrets or credentials I haven't provided, ASK me — do not invent values or leave raw placeholders. After adding it, tell me how to verify the server is connected.motherduck_tokenrequiredMotherDuck access token (alternative to --motherduck-token)AWS_ACCESS_KEY_IDAWS access key for S3 database connectionsAWS_SECRET_ACCESS_KEYAWS secret key for S3 database connectionsAWS_SESSION_TOKENAWS session token for temporary credentials (IAM roles, SSO, EC2 instance profiles)AWS_DEFAULT_REGIONAWS region for S3 connectionsAWS_ENDPOINTAWS endpoint for S3 connectionsMinimal, token-efficient Database MCP Server for PostgreSQL, MySQL, SQL Server, SQLite, MariaDB