# Fivetran MCP connector

Connect Fivetran to Elaichi and Claude, ChatGPT, Cursor, Microsoft Copilot, any MCP client and the Elaichi Agent can manage Fivetran users, roles, groups, teams and connection certificates, each person signed in as themselves with every call logged.

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

## Facts

| | |
| --- | --- |
| Application | Fivetran |
| Category | BI |
| AI tools | 179 |
| Authentication | Connects with an API key |
| Bring your own OAuth app | No |
| MCP endpoint | https://api.elaichi.ai/mcp |
| Works with | Claude, ChatGPT, Cursor, Microsoft Copilot, 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 Fivetran is connected

- List every Fivetran user with an account administrator role
- Add Priya to the analytics team in Fivetran
- Create a connect card link for the Salesforce source

## Connect Fivetran in Elaichi

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

1. Open Connections, choose Add connection, and pick Fivetran.
2. Optionally set Share with, then press Connect.
3. Paste a Fivetran API key. One person generates a token in Fivetran 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 Fivetran tools are exposed, rename them, or freeze arguments before anyone points a client at it.

## Fivetran 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.

## Fivetran MCP connector for Claude Code

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

1. In a terminal, run `claude mcp add` with the endpoint.
2. Start Claude Code and run `/mcp`.
3. Pick elaichi, then sign in and approve in the browser.

```bash
claude mcp add --transport http elaichi --scope user https://api.elaichi.ai/mcp
```

Use `--scope project` instead to share it with the repository through `.mcp.json`.

## Fivetran 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. The plugin directory lives at chatgpt.com/plugins.

## Fivetran 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.

## Fivetran MCP connector for Microsoft Copilot

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

1. In Copilot Studio, open your agent, then Tools, Add a tool, New tool and Model Context Protocol.
2. Paste the endpoint into Server URL, choose OAuth 2.0, then Dynamic discovery, and press Create.
3. Create a new connection, sign in and approve, then press Add to agent.
4. Under Channels, open Teams and Microsoft Copilot and make the agent available in Microsoft Copilot.

Microsoft Copilot was called Microsoft 365 Copilot. Keep the tool on end-user credentials, so each person signs in as themselves.

## Connect Fivetran 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

Every scope the client asks for starts ticked except "Delete things", which you tick yourself. Calling a Fivetran tool needs "Use your connected apps". 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. Think hard before granting Use your connected apps, which can also delete in connected apps and cannot be undone.

## What teams do with Fivetran through Elaichi

### Onboard a new analyst in minutes

Data engineering. Create the Fivetran user, give them the right role and add them to the group and team that own their destinations, all from one ask.

### Audit who still has Fivetran access

IT. List every Fivetran user with their role, spot people who have moved on or no longer need it, and remove them without opening the admin console.

### Rotate connection certificates before they expire

Security. Pull the certificates and fingerprints behind each connection, add replacements and bulk delete the stale ones in a single sitting.

### Check which group owns a destination

Analytics. Ask which Fivetran groups and teams exist and who sits in each before requesting a new pipeline, so the request goes to the right owner.

### Hand a source owner a connect card

Operations. Generate a Fivetran connect card link so the person who owns a source can authorize it themselves, without being given a Fivetran login.

### Reorganize teams as the company grows

Platform. Create, rename and update Fivetran teams and groups when departments split or merge, and keep ownership of destinations tidy.

## Elaichi vs Zapier MCP vs Composio for Fivetran

All three can connect Fivetran 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 Fivetran tools | Allow or restrict single Fivetran 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 Fivetran 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 | Free for one person; $15 per user per month on Gold. | 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 Fivetran to Claude?

Two steps. First connect Fivetran in Elaichi by pasting a Fivetran API key, there is no OAuth application to register and no client ID or secret to generate. Then in Claude open Customize, then Connectors, then Add, and paste https://api.elaichi.ai/mcp. Claude now reaches Fivetran through that one endpoint, and you sign in as yourself.

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

Yes. Fivetran is connected once in Elaichi, and the same endpoint, https://api.elaichi.ai/mcp, works in Claude, ChatGPT, Cursor, Microsoft Copilot, any other MCP client and the Elaichi Agent. There is nothing to set up again per client.

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

It can list and look up Fivetran users, roles, groups and teams, create or update a user, add someone to a team, and remove people who have left. It can also pull the certificates and fingerprints behind a Fivetran connection, add or bulk delete certificates, and create a connect card link for someone to authorize a source. Short, concrete asks such as 'list every Fivetran user with an admin role' work better than long paragraphs.

### Does connecting Fivetran give the AI my whole account?

No. Every call into Fivetran runs inside the access of the person who signed in to Elaichi, so an agent can never see more of Fivetran than that person already could. Elaichi can narrow that access further with roles and restrictions, and it never widens it.

### Can my team share one Fivetran connection?

Yes. One person connects Fivetran in Elaichi and shares the connection with a team, and nobody else ever handles the Fivetran API key. Each teammate still signs in to Elaichi as themselves, so the audit log names the actual person behind every Fivetran change.

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

Yes. Restrictions in Elaichi work per action, so you can restrict deleting Fivetran users, removing certificates or editing groups for a role or for one person while leaving listing and lookups open. A restricted action still appears in Elaichi's tool search, flagged as restricted, but no client can run it, whatever the prompt says.

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

Offboarding a person in Elaichi ends their access to Fivetran through every client at once. If they connected Fivetran for a shared team, that connection keeps working for everyone else it was shared with. When you want Fivetran gone entirely, disconnecting it once in Elaichi removes it from Claude, ChatGPT, Cursor, Microsoft Copilot and every other client.

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

Yes. Fivetran 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, Microsoft Copilot, 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 Fivetran?

Yes. Both let Claude, ChatGPT or Cursor use Fivetran. 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 Fivetran call.

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

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 Fivetran: one address, restrictions per role or user, and $15 per user per month. Both have role permissions and a log of every call.

## All 179 Fivetran tools

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

- **List all Fivetran users** (List). List all users in your Fivetran account. Returns each user's id, email, role, active flag, logged_in_at, and created_at. Max 1000 per page.
- **Get single Fivetran user by ID** (Get). Get a single Fivetran user by id. Returns the user object including id, email, verified, role, active, and created_at. Required: id.
- **Delete a Fivetran user by ID** (Delete). Delete a Fivetran user by id. Returns a confirmation message on success. Required: id. The last remaining account owner cannot be deleted.
- **Update a Fivetran user by ID** (Update). Update an existing Fivetran user by id. Returns the updated user object including id, email, verified, role, active, and created_at. Required: id.
- **Create a Fivetran user** (Create). Invite a new user to your Fivetran account. Regular users get an invitation email they must accept; service accounts are created immediately. Returns: id, email, verified, role, active, given_name, family_name, created_at, and 6 more. Required: email, given_name, family_name. A service account user can't create another service account user.
- **List all Fivetran roles** (List). List all predefined and custom roles in your Fivetran account. Returns: name, description, scope, is_custom, is_deprecated, replacement_role_name. Max 1000 per page.
- **List all Fivetran groups** (List). List all groups (destinations) in your Fivetran account. Returns: id, name, created_at. Max 1000 per page.
- **Get single Fivetran group by ID** (Get). Get a single Fivetran group by id. Returns: id, name, created_at. Required: id.
- **Create a Fivetran group** (Create). Create a new group in your Fivetran account. Returns: id, name, created_at. Groups and destinations map 1:1, so a group must exist before you can create a destination in it.
- **Delete a Fivetran group by ID** (Delete). Delete a group from your Fivetran account by id. Returns a confirmation message that the group has been deleted. Required: id.
- **Update a Fivetran group by ID** (Update). Update the name of an existing group in your Fivetran account by id. Returns: id, name, created_at. Required: id.
- **List all Fivetran teams** (List). List all teams in your Fivetran account. Returns each team's id, name, description, and role.
- **Get single Fivetran team by ID** (Get). Get details of a single team in your Fivetran account by id. Returns: id, name, description, role. Required: id.
- **Create a Fivetran team** (Create). Create a new team in your Fivetran account with a name, description, and role. Returns the created team's id, name, description, and role.
- **Delete a Fivetran team by ID** (Delete). Delete a team from your Fivetran account by id. Returns the deletion status message on success. Required: id.
- **Update a Fivetran team by ID** (Update). Update an existing team in your Fivetran account by id to change its name, description, or role. Returns the updated team's id, name, description, and role. Required: id.
- **List all Fivetran account** (List). Get Fivetran account information for the authenticated API key. Returns: account_id, account_name, user_id, system_key_id.
- **Create a Fivetran certificate** (Create). Approve a certificate for a Fivetran connection or destination so Fivetran trusts the source/destination database certificate during setup tests. Deprecated. Returns a confirmation message once approved. Required: hash, encoded_cert.
- **List all Fivetran connection certificates** (List). List certificates approved for a Fivetran connection. Returns: id, name, hash, type, sha1, sha256, public_key, validated_date, validated_by. Max 1000 per page. Required: connection_id, connectionId.
- **Create a Fivetran connection certificate** (Create). Approve a certificate for a Fivetran connection so its setup tests trust the source database certificate. Returns: id, name, hash, sha1, sha256, validated_date. Only needed for MySQL, PostgreSQL, or SQL Server sources. Required: connection_id, connectionId, hash, encoded_cert.
- **Fivetran connection certificates bulk delete** (Delete). Revoke a certificate from a Fivetran connection by its hash. Returns an empty 204 response on success. Required: connection_id, hash.
- **List all Fivetran connection certificates connections** (List). Retrieve the details of a single certificate for a Fivetran connection by its hash. Returns: id, name, hash, type, sha1, sha256, public_key, validated_date, validated_by. Required: connection_id, hash.
- **Create a Fivetran connection connect card** (Create). Create a Connect Card for a Fivetran connection, generating the Connect Card URI for end-user setup. Returns: connector_id, connect_card, connect_card_config. Required: connection_id, connectionId, connect_card_config.
- **List all Fivetran connection fingerprints** (List). List the approved SSH fingerprints for a Fivetran connection. Returns: hash, attributes. Required: connection_id, connectionId.
- **Create a Fivetran connection fingerprint** (Create). Approve an SSH fingerprint for a Fivetran connection, letting Fivetran trust it when connecting to the source database through an SSH tunnel. Returns the approved fingerprint record: hash, attributes. Required: connection_id, connectionId, hash, public_key.
- **Fivetran connection fingerprints bulk delete** (Delete). Revoke an approved SSH fingerprint from a Fivetran connection by its hash. Returns an empty 204 response on success. Required: connection_id, hash.
- **List all Fivetran connection fingerprints connections** (List). Retrieve details of a specific connection fingerprint for a Fivetran connection. Returns: id, hash, public_key, validated_date, validated_by. Required: connection_id, hash.
- **Create a Fivetran connection move** (Create). Move a Fivetran connection to a different destination group. Returns: connection_id, destination_group_id, sync_behavior, moved_at, job_id. Connection must be paused first; job_id tracks async progress. sync_behavior: CONTINUE, BACKFILL, CONTINUE_WITH_DATA (Snowflake Lakehouse only). Required: connection_id, connectionId, destination_group_id.
- **Get single Fivetran connection move by ID** (Get). Get the status of a Fivetran connection move job by id. Returns: id, status, connection_id, source_group_id, destination_group_id, created_at, started_at, completed_at, error_message. Used to monitor a move initiated with CONTINUE_WITH_DATA. Required: id, connection_id, connectionId, jobId.
- **Create a Fivetran connection resync** (Create). Trigger a full historical re-sync of a Fivetran connection, optionally limited to specific tables. Returns the confirmation message once the sync is triggered. Fails with a conflict error if a sync is already in progress; paused connections schedule the sync until re-enabled. Required: connection_id, connectionId.
- **List all Fivetran connection schemas** (List). Retrieve the top-level schema configuration for a Fivetran connection, including global flags and the schema-to-table selection tree. Returns: schemas, schema_change_handling, enable_new_by_default, row_filtering_supported. Unedited columns are omitted from the response. Required: connection_id, connectionId.
- **Create a Fivetran connection schema** (Create). Set up a schema config for a new Fivetran connection before the schema is captured from the source. Returns: schemas, schema_change_handling, enable_new_by_default, row_filtering_supported. Settings for entities that cannot be excluded are ignored. Required: connection_id, connection_id, schemas.
- **Update a Fivetran connection schema by ID** (Update). Update the database schema config for a single schema of a Fivetran connection. Returns: schemas, schema_change_handling, enable_new_by_default, row_filtering_supported. Required: connection_id, schema_name, connection_id, enabled.
- **Fivetran connection schemas bulk update** (Update). Update a Fivetran connection's schema config, including schemas, tables, sync modes, and row filters. Returns: schemas, schema_change_handling, enable_new_by_default, row_filtering_supported. Required: connection_id. Column drops take effect at the next sync.
- **Delete a Fivetran schemas drop column by ID** (Delete). Drop multiple blocked columns from your Fivetran destination tables; columns are marked for deletion now and physically removed during the next sync. Returns a success confirmation. Required: connection_id, connectionId, schemas.
- **Create a Fivetran schemas reload** (Create). Reload the full schema configuration for an existing Fivetran connection from its data source. Returns the reload status message. May take a long time to complete depending on the schema size and the number of databases, tables, and columns. Required: connection_id, connectionId.
- **Create a Fivetran tables resync** (Create). Trigger a historical re-sync of the specified tables within a Fivetran connection. Returns: code, message. Does not override the connection's standard sync frequency set in the Fivetran dashboard. Required: connection_id, connectionId.
- **Fivetran schema tables bulk update** (Update). Update a table's sync configuration within a Fivetran connection schema. Returns: enable_new_by_default, schemas, row_filtering_supported, schema_change_handling. Required: connection_id, schema_name, table_name, enabled.
- **List all Fivetran table columns** (List). Retrieve the real-time column configuration for a source table in a Fivetran connection. Returns: columns (enabled and hashed flags per column). Only for connections in a Connected state; the table must report supports_columns_config. Required: connection_id, schema, table, connectionId.
- **Fivetran table columns bulk update** (Update). Update a column's configuration for a Fivetran connection (enable, hash, primary key, target data type). Returns: schemas, schema_change_handling, enable_new_by_default, row_filtering_supported. Required: connection_id, schema_name, table_name, column_name, enabled.
- **Fivetran table columns bulk delete** (Delete). Drop a blocked column from the destination for a Fivetran connection table. Returns an empty 204 response on success. Required: connection_id, schema_name, table_name, column_name.
- **Get single Fivetran connection state by ID** (Get). Retrieve the connection state for a Fivetran Function or Connector SDK connection. Returns: state. Required: id. Other connection types return a 400 Bad Request error.
- **Fivetran connection states bulk update** (Update). Update the stored connection state for a Fivetran Function or Connector SDK connection. Returns the updated state with its code and message. Required: connection_id. Only supported for Function and Connector SDK connectors; pause the connection first, and the update fails with 409 if a sync is in progress.
- **Create a Fivetran connection sync** (Create). Trigger an on-demand data sync for a Fivetran connection without waiting for the next scheduled run. Returns a confirmation message that the sync was triggered. This does not override the connection's normal sync schedule; for manual-schedule connections, syncs only run through this endpoint. Required: connection_id, connectionId.
- **Get single Fivetran connection sync history by ID** (Get). Retrieve sync history records for a Fivetran connection by id. Returns: sync_id, status, start, end, reason, stages. Max 7-day time window; use start_time and end_time to narrow it (duration is deprecated). Required: id, connectionId.
- **Create a Fivetran connection test** (Create). Run the setup tests for an existing Fivetran connection without changing its configuration. Returns the connection object including its id, service, status, setup_tests, and sync_frequency. Required: connection_id, connectionId.
- **List all Fivetran connection warnings** (List). List all active warnings for a Fivetran connection, grouped by destination table; each warning includes a code, message, error details, and creation timestamp. Returns: items, next_cursor. Required: connection_id, connectionId.
- **Fivetran connection warnings bulk delete** (Delete). Dismiss connection warnings of a specific type for a Fivetran connection. Returns an empty 204 response on success. Required: connection_id, warning_type.
- **List all Fivetran connections** (List). List all connections in your Fivetran account. Returns each connection's id, service, schema, paused, status, sync_frequency, and created_at.
- **Create a Fivetran connection** (Create). Create a new connection in Fivetran. Runs setup tests and returns the created connection with id, service, schema, paused, status, and created_at. Requires group_id and service; destination_schema_names will soon be required. Required: group_id, service.
- ...and 129 more tools. Call `tools/list` via the MCP endpoint, or see the full catalog via the API, for the complete set.
