# Coda MCP connector

The Coda connector brings your Coda docs, pages, tables and rows to Claude, ChatGPT, Cursor and any MCP client, so once connected, each person can ask about and work in Coda while signed in as themselves.

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

## Facts

| | |
| --- | --- |
| Application | Coda |
| Category | Knowledge Management |
| AI tools | 18 |
| Authentication | Connects with an API key |
| Needs your own OAuth app | No |
| MCP endpoint | https://api.elaichi.ai/mcp |
| Works with | Claude, ChatGPT, Cursor, any MCP client, and the Elaichi Agent |
| Tools advertised by name | Yes |

## What you can ask once Coda is connected

- Summarize the pages in our Q3 planning Coda doc.
- Which Coda tables track owners without a due date?
- Create a Coda doc for next week's launch review.

## Connect Coda in Elaichi

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

1. Open Connections, choose Add connection, and pick Coda.
2. Optionally set Share with, then press Connect.
3. Paste a Coda API key. One person generates a token in Coda and pastes it once. Everyone else works through Share with, and never sees it.

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

## Connect Coda to 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.

## Connect Coda to 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.

## Connect Coda to 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 Coda 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 Coda 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 Coda through Elaichi

### Find the right doc without browsing

Operations. Ask which Coda docs cover vendor onboarding or the office move and get the list back with links, instead of scrolling through the workspace folder by folder.

### Read the roadmap tracker in plain language

Product. Ask what is marked blocked in the launch tracker table and the agent reads the columns and rows in that Coda doc and answers in a sentence or two.

### Check the account plan before a call

Sales. Pull the rows for one customer from the account planning table in Coda and get a short brief on owners, next steps and open questions.

### Reuse a handbook page elsewhere

People. Export a page from the team handbook in Coda and turn it into an email, a slide or a shorter version for new joiners.

### Start a new doc from a conversation

Leadership. After a planning session, have the agent create a Coda doc for the initiative and rename it later when the project name settles.

### Answer from the runbook, not from memory

Support. Ask how a refund escalation is handled and the agent finds the page in the support runbook doc in Coda and quotes the steps.

## Frequently asked questions

### How do I connect Coda to Claude?

Connect Coda in Elaichi first: Coda connects with an API key, so you generate one in your Coda account settings and paste it in, with no OAuth application to register and no client ID or secret to generate. Then open Claude, go to Customize, then Connectors, then Add, and paste https://api.elaichi.ai/mcp as the endpoint. Claude asks you to sign in to Elaichi and your Coda docs are available from then on.

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

Yes. Once Coda is connected in Elaichi, the same endpoint, https://api.elaichi.ai/mcp, is pasted into Claude, ChatGPT, Cursor, any other MCP client or used by the Elaichi Agent. Coda is connected once and every client sees it.

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

With Coda connected, an agent can list the docs you have access to, open a doc and its pages, read the tables, columns and rows inside it, and export a page so its content can be summarized or reused. It can also create a new Coda doc, rename or update an existing one, and delete a doc if that action is allowed. It cannot add or edit individual rows in Coda through this connector.

### Does connecting Coda give the AI every doc in the workspace?

No. Every call to Coda runs under the person who signed in, so the agent only sees the Coda docs, pages and tables that person could already open in Coda. Elaichi can narrow that further with roles and restrictions, but it never widens what someone can reach in Coda.

### Can my team share one Coda connection?

Yes. One person connects Coda in Elaichi and shares the connection with a team, so nobody else needs to handle a Coda API key. Each teammate still signs in to Elaichi as themselves, so every action on a Coda doc is logged under their own name.

### Can I stop an agent from deleting or changing docs in Coda?

Yes. Restrictions in Elaichi apply per action, so you can allow reading Coda docs, pages and rows while blocking deleting or updating a doc. A blocked action is never advertised to Claude, ChatGPT, Cursor or any other client, so no prompt can reach it.

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

When a person is offboarded in Elaichi, their access to Coda through Elaichi ends at once, in every client they had set up. If they had shared a Coda connection with a team, it keeps working for everyone else. Disconnecting Coda once in Elaichi removes it from Claude, ChatGPT, Cursor and every other connected client.

## All 18 Coda tools

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

- **List all Coda users** (List). List all users within an organization in Coda. Requires organization_id. Returns each user's id, email, name, pictureUrl, status, registeredAt, deactivatedAt, and ownedDocCount, along with pagination details.
- **List all Coda organizations** (List). List organizations the caller is an administrator for in Coda. Returns an array of organizations with associated href, nextPageToken, and nextPageLink fields in the response.
- **Get single Coda organization by ID** (Get). Get organization in Coda using id. Returns type (always 'organization'), id (organization identifier), and name (organization name) in the response.
- **List all Coda me** (List). Get information about the current user in Coda. Returns name, loginId, type, scoped, tokenName, href, workspace (with id, organizationId, browserLink, name), and pictureLink in the response.
- **List all Coda docs** (List). List available docs in Coda accessible by the user. Returns key fields including id, name, owner, createdAt, updatedAt, and workspace details.
- **Get single Coda doc by ID** (Get). Get metadata for a specific doc in Coda using id. Returns fields such as id, name, type, href, browserLink, owner, ownerName, createdAt, updatedAt, and docSize that describe the doc and its details.
- **Create a Coda doc** (Create). Create a new doc in Coda. Returns id, type, href, browserLink, name, owner, ownerName, createdAt, updatedAt, workspace, folder, and other metadata about the created doc. Requires being a Doc Maker in the workspace.
- **Update a Coda doc by ID** (Update). Update metadata for a doc in Coda using id. Returns updated doc information including title and iconName. Requires the user to be a Doc Maker if updating the title.
- **Delete a Coda doc by ID** (Delete). Delete a specific doc in Coda using id. Returns a result object confirming that the doc was deleted.
- **List all Coda pages** (List). List pages in a Coda doc using doc_id. Returns each page's id, name, type, href, browserLink, createdAt, and updatedAt fields in the response.
- **Get single Coda page by ID** (Get). Get details about a specific page in Coda. Requires doc_id and id. Returns fields like id, name, type, href, browserLink, contentType, isHidden, isEffectivelyHidden, children, parent, authors, createdAt, and updatedAt.
- **List all Coda tables** (List). List tables in a specific doc in Coda. Requires doc_id. Returns each table's id, name, type, tableType, href, browserLink, and parent details.
- **Get single Coda table by ID** (Get). Get details about a specific table or view in Coda. Requires doc_id and id. Returns fields such as id, name, href, browserLink, layout, rowCount, createdAt, updatedAt, and tableType for identifying and describing the table.
- **List all Coda columns** (List). List columns in a specific table in Coda. Requires doc_id and table_id_or_name. Returns fields like href, nextPageToken, nextPageLink, and an array of column objects representing each table column.
- **Get single Coda column by ID** (Get). Get details about a specific column in Coda using doc_id, table_id_or_name, and id. Returns fields such as id (column identifier), name (column name), type (resource type), href (API link), and format (column format).
- **List all Coda rows** (List). List rows in a specific table in Coda. Requires doc_id and table_id_or_name. Returns fields such as id, name, href, index, createdAt, updatedAt, and values describing each row.
- **Get single Coda page export by ID** (Get). Get page export status in Coda using doc_id, page_id_or_name, and id. Returns id (export request identifier), status (export state), href (status URL), downloadLink (file download URL), and error (failure message if any).
- **Create a Coda page export** (Create). Begin a page content export in Coda for a specific page using doc_id and id. Returns id (export request identifier), status (export progress), and href (URL for polling export status).