# MotherDuck MCP connector

The MotherDuck connector lets Claude, ChatGPT, Cursor and the Elaichi Agent query your MotherDuck databases, browse tables and shares, and work with dives and flights, with every query running inside the signed-in person's own access.

Source: https://elaichi.ai/connectors/motherduck/

## Facts

| | |
| --- | --- |
| Application | MotherDuck |
| Category | Database |
| AI tools | 39 |
| Authentication | Connects over OAuth |
| Bring your own OAuth app | No |
| Native MCP | Yes. MotherDuck builds and runs this MCP server. Elaichi adds sign-in, access controls and an audit log on top |
| Support for its tools | motherduck.com/contact-us |
| MCP endpoint | https://api.elaichi.ai/mcp |
| Works with | Claude, ChatGPT, Cursor, any MCP client, and the Elaichi Agent |
| Tools advertised by name | No. Connected tools are never listed one by one, however few there are. The endpoint advertises `search_tools` and `execute_tool` instead |

## What you can ask once MotherDuck is connected

- Which tables in the sales database were updated this week?
- Show me last month's revenue by region from the orders table.
- Save this churn analysis as a dive and share it with Finance.

## Connect MotherDuck in Elaichi

This happens once for the organization, before any client is involved.

1. Open Connections, choose Add connection, and pick MotherDuck.
2. Optionally set Share with, then press Connect.
3. Approve it in MotherDuck. MotherDuck's own window opens. Whoever approves it decides what this connection can reach.

Credentials are vaulted and nobody, including the AI, reads them back. The connection becomes a toolbox immediately, so you can curate which MotherDuck tools are exposed, rename them, or freeze arguments before anyone points a client at it.

## MotherDuck MCP connector for Claude

Endpoint: https://api.elaichi.ai/mcp

1. Open Customize, then Connectors.
2. Press Add.
3. Name it, paste the MCP server URL, then Continue.
4. Sign in and approve.

On Team and Enterprise, an Owner adds it once. Everyone else turns it on for themselves.

## MotherDuck MCP connector for ChatGPT

Endpoint: https://api.elaichi.ai/mcp

1. Open Plugins, then press the + button.
2. Name it and paste the endpoint into Server URL.
3. Leave Authentication on OAuth, then tick the risk acknowledgement.
4. Press Create, then sign in and approve.

Works on the web today. The plugin directory lives at chatgpt.com/plugins.

## MotherDuck MCP connector for Cursor

Endpoint: https://api.elaichi.ai/mcp

1. Open `~/.cursor/mcp.json`.
2. Add the endpoint under `mcpServers`.
3. Reload Cursor, then sign in and approve.

Set up per machine, so repeat it on each computer you work from.

## Connect MotherDuck to any MCP client

Endpoint: https://api.elaichi.ai/mcp

1. Add the endpoint as a remote MCP server.
2. Sign in and approve.

The Elaichi Agent already has these tools, with nothing to set up.

## What the consent screen decides

Only Read is granted by default, which is not enough to call a MotherDuck tool. Over MCP there is no trusted place to confirm a write in the moment, so the consent screen is the standing approval rather than a formality. Grant Read and Run tools. Think hard before granting Delete, which reaches into connected apps and cannot be undone.

## What teams do with MotherDuck through Elaichi

### Answer a metrics question without writing SQL

Analytics. Ask for last quarter's revenue by region and the agent finds the right table in MotherDuck, runs the query and explains the result in plain language.

### Check month-end numbers against the warehouse

Finance. Pull totals from the billing and orders tables in MotherDuck and compare them with the close spreadsheet before anyone signs off.

### Pull pipeline figures from the sales database

Sales. Get open pipeline by owner and stage straight from MotherDuck ahead of the Monday forecast call, without waiting on a data request.

### Measure a campaign from the events tables

Marketing. Ask how many signups came from last week's campaign and the agent queries the events and attribution tables in MotherDuck for you.

### Find the right table before building anything

Data. Search the MotherDuck catalog for a column or view, see what each one holds, and ask the docs how a feature works instead of guessing.

### Save a recurring check as a dive

Operations. Turn a weekly inventory or SLA query into a saved dive in MotherDuck, update it as the question changes, and share the results with the team.

## Elaichi vs Zapier MCP vs Composio for MotherDuck

All three can connect MotherDuck to an AI assistant, and all three have admin controls. They differ in where access lives and how you pay.

| What to check | Elaichi | Zapier MCP | Composio |
| --- | --- | --- | --- |
| Where the AI connects | One address for the whole organization. Endpoint: https://api.elaichi.ai/mcp | A server per member, created at sign-in. | An MCP endpoint per team, or an SDK. |
| Control over MotherDuck tools | Allow or restrict single MotherDuck tools, per role or user. | App and action restrictions on the account. | Role permissions, down to the action. |
| Record of calls | One audit entry per MotherDuck call. | A History tab of tool calls. | A log of every tool call. |
| Single sign-on | SAML or OIDC, plus SCIM, on Gold. | SAML on Enterprise. | SAML and OIDC on Enterprise. |
| Price | $15 per user per month. | 2 tasks per successful call. | Billed per tool call. |

Sources: Zapier MCP [docs](https://docs.zapier.com/mcp/get-started/quickstart), [security](https://docs.zapier.com/mcp/manage/security), [usage](https://docs.zapier.com/mcp/features/usage); Composio [docs](https://docs.composio.dev/docs/composio-connect), [gateway](https://composio.dev/mcp-gateway), [enterprise](https://composio.dev/enterprise), [pricing](https://composio.dev/pricing). Checked September 2026.

Longer take: [Zapier MCP alternative](/blog/zapier-mcp-alternative/) and [when you don't need an MCP gateway](/blog/when-you-dont-need-an-mcp-gateway/).

## Frequently asked questions

### How do I connect MotherDuck to Claude?

Connect MotherDuck in Elaichi first; it signs in over OAuth, so you approve access in your MotherDuck account and there is no client ID or secret to generate. Then in Claude open Customize, then Connectors, then Add, and paste https://api.elaichi.ai/mcp as the endpoint. Claude signs you in through Elaichi and MotherDuck is ready to query.

### Does MotherDuck work with ChatGPT and Cursor as well as Claude?

Yes. Once MotherDuck is connected in Elaichi, the same endpoint, https://api.elaichi.ai/mcp, works in Claude, ChatGPT, Cursor, any other MCP client and the Elaichi Agent. You connect MotherDuck once and every client picks it up.

### What can an AI agent actually do with my MotherDuck data?

An agent can browse your MotherDuck databases, shares, tables, views and macros, search the catalog for the right column, run queries and explain the answers, and, where allowed, write results back. It can also save, update and share dives, create and update flights, and look up the MotherDuck docs when it is unsure. Short, concrete asks such as "revenue by region last month" work better than long paragraphs.

### Does connecting MotherDuck give the AI access to every database and share?

No. Every query runs as the person who signed in, so the agent sees only the MotherDuck databases, shares and tables that person can already reach. Elaichi can narrow that further for a team or a role, but it can never widen what MotherDuck itself allows.

### Can my team share one MotherDuck connection?

Yes. One person connects MotherDuck in Elaichi and shares it with a team, and nobody else ever handles a credential. Each teammate still signs in to Elaichi as themselves, so the audit log names the actual person behind every MotherDuck query.

### Can I stop an agent from deleting or changing things in MotherDuck?

Yes. In Elaichi you restrict actions one by one, so you can allow reading from MotherDuck tables while blocking writes, dive deletion or flight changes. A restricted action is never shown to Claude, ChatGPT or Cursor at all, so no prompt can reach it.

### What happens to a MotherDuck connection when someone leaves?

Offboarding a person in Elaichi ends their access to MotherDuck through every client at once. If they had shared a MotherDuck connection with a team, it keeps working for everyone else. Disconnecting MotherDuck once in Elaichi removes it from Claude, ChatGPT, Cursor and every other client together.

### Does the MotherDuck MCP connector work with Gemini, Codex, Claude Code or other MCP clients?

Yes. MotherDuck is reached over the same MCP endpoint every client uses, so anything that speaks MCP can call it — Gemini, Codex, Claude Code, Windsurf, Cline, Zed and OpenCode among them — alongside Claude, ChatGPT, Cursor, and the Elaichi Agent. The tools on offer and the access behind them are identical whichever client asks. Only the setup screen differs.

### Is Elaichi an alternative to Zapier MCP for MotherDuck?

Yes. Both let Claude, ChatGPT or Cursor use MotherDuck. Zapier MCP fits a team that already automates in Zapier, since each person signs in and acts as themselves in that account. Elaichi fits when IT wants one address for the whole company, per-tool rules by role, and a record of every MotherDuck call.

### How is Elaichi different from Composio for MotherDuck?

Composio gives AI agents tools and sign-in handling across 1,000+ apps, for developers building agents or people using an assistant, billed per tool call. Elaichi gives a company's own people governed access to MotherDuck: one address, restrictions per role or user, and $15 per user per month. Both have role permissions and a log of every call.

## All 39 MotherDuck tools

Every tool below is callable through https://api.elaichi.ai/mcp once MotherDuck is connected, subject to the toolbox it is in and the restrictions on the caller.

- **List databases** (List). List all databases in your MotherDuck account
- **List shares** (List). List database shares that have been shared with you
- **List tables** (List). List tables and views in a MotherDuck database
- **List columns** (List). List columns of a table or view with types and comments
- **List views** (List). List views in a MotherDuck database with schema, comment, and column count
- **List macros** (List). List table and scalar macros in a MotherDuck database with their parameters
- **Search catalog** (Search). Fuzzy search across databases, schemas, tables, columns, and shares
- **Query** (Search). Execute SQL queries against MotherDuck databases
- **Query rw** (Search). Execute SQL queries that can modify data or schema in MotherDuck
- **Ask docs question** (Action). Ask questions about DuckDB or MotherDuck documentation
- **Get dive guide** (Get). Load instructions for creating MotherDuck Dives
- **List dives** (List). List all Dives in your MotherDuck workspace
- **Read dive** (Action). Read a specific Dive by ID, including its full component code
- **View dive** (Action). Render a MotherDuck Dive as a live, interactive MCP app inside the host client.
- **Save dive** (Action). Save a new Dive to your MotherDuck workspace
- **Update dive** (Update). Update an existing Dive's title, description, or content
- **Share dive data** (Share). Share the data for a Dive with your organization
- **Delete dive** (Delete). Permanently delete a Dive by ID
- **Get flight guide** (Get). Load the canonical instructions for authoring, scheduling, running, and troubleshooting MotherDuck Flights.
- **List flights** (List). List Flights with summary metadata, optionally filtered by keywords.
- **Get flight** (Get). Fetch a Flight's metadata and version snapshot by UUID, optionally at a specific historical version.
- **List flight versions** (List). List the version history of a Flight, newest first.
- **Create flight** (Create). Create a new Flight from Python source code, requirements, and an optional schedule.
- **Update flight** (Update). Update a Flight's source, requirements, config, token, secrets, name, or schedule.
- **Edit flight source** (Action). Edit a Flight's source code with one or more find-and-replace operations, producing a new version.
- **Delete flight** (Delete). Permanently delete a Flight, its versions, schedule, and run history.
- **Run flight** (Run). Trigger an on-demand execution of a Flight using its current version.
- **List flight runs** (List). List the execution history of a Flight, newest first.
- **Get flight run logs** (Get). Fetch the logs and run record for a single Flight run.
- **Cancel flight run** (Cancel). Cancel an in-progress Flight run.
- **Get query guide** (Search). Load your organization's query guidance and an overview of available Guides before writing SQL
- **List guides** (List). Browse your organization's Guides level by level, grouped by topic
- **Get guide** (Get). Load a Guide's full content by UUID, optionally pinning a historical version
- **Create guide** (Create). Create a new Guide — a markdown document that AI agents use to answer your organization's data questions correctly
- **Update guide** (Update). Append a version to an existing Guide with updated content or references
- **Edit guide content** (Action). Apply targeted string replacements to a Guide and save the result as a versioned update
- **Update guide metadata** (Update). Change a Guide's title, description, or topic without creating a content version
- **Set guide access** (Set). Change a Guide's visibility between private and org-wide
- **Delete guide** (Delete). Soft-delete a Guide while preserving its version history