> For the complete documentation index, see [llms.txt](https://developer.celigo.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://developer.celigo.com/mcp/readme.md).

# Celigo Platform MCP

Celigo Platform MCP is a hosted, remote MCP server that lets an AI agent operate your integrator.io account in plain language. Connect Claude, Cursor, VS Code, ChatGPT, or any client that speaks the [Model Context Protocol](https://modelcontextprotocol.io/), sign in with your Celigo credentials, and the agent can read, build, run, and troubleshoot your integrations through conversation.

There is nothing to install and no local process to run. The server is hosted by Celigo and calls the same [Celigo REST API](https://developer.celigo.com/api) under the hood, with tool descriptions written for AI agents.

|                   |                                                                                         |
| ----------------- | --------------------------------------------------------------------------------------- |
| **Endpoint (US)** | `https://api.integrator.io/celigo-mcp`                                                  |
| **Endpoint (EU)** | `https://api.eu.integrator.io/celigo-mcp`                                               |
| **Transport**     | Streamable HTTP                                                                         |
| **Auth**          | OAuth sign-in with your Celigo credentials, or `Authorization: Bearer <YOUR_API_TOKEN>` |

## What you can do

Ask in plain language and the agent picks the right tools:

* **"Audit my account health."** It counts your integrations and flows, finds the ones with errors, and hands back a prioritized list.
* **"Why did the Shopify to NetSuite flow fail last night?"** It pulls the latest job, reads the per-step errors, and explains the root cause.
* **"Build a flow that syncs new Salesforce accounts into NetSuite."** It checks your connections, fetches the right schemas, and scaffolds the export, import, and flow.
* **"Retry the failed records on that order import."** It lists the open errors and retries them after you confirm.
* **"Install the Shopify to NetSuite template."** It previews what the marketplace template creates, maps its connections to yours, and deploys it as a new integration.

See [What you can do](/mcp/use-cases.md) for worked examples, the exact prompts, and the tools each one fires.

## How it works

Your client connects to the endpoint over Streamable HTTP and authenticates one of two ways: an OAuth sign-in in your browser, or a bearer API token in the `Authorization` header. The agent discovers the available [tools](/mcp/tools.md) at runtime, then calls them on your behalf. Every call runs with the permissions of the signed-in user or the token, so the agent can do exactly what you could do through the API, and nothing more.

## How it relates to the API and CLI

Celigo Platform MCP, the [REST API](https://developer.celigo.com/api), and the [Celigo CLI](https://developer.celigo.com/cli) all operate on the same integrator.io resources. Platform MCP wraps the REST API with tool descriptions tuned for AI agents. It does not replace the API or CLI.

> **Not the same as the `mcpServers` resource.** Celigo Platform MCP is Celigo's first-party hosted server that AI agents connect to. The [`mcpServers` resource](https://developer.celigo.com/cli/extend/mcp-servers) is a builder-side primitive for creating your own MCP endpoints that expose your Tools and APIs to external clients. They are separate things.

## FAQ

**Does it use OAuth or an API token?** Either. OAuth is the default for interactive clients: add the endpoint with no credentials, and the client opens a browser sign-in. Sign in with your Celigo credentials (SSO and MFA included), then pick an account and environment. For scripts and clients without OAuth support, pass an API token or personal access token in the `Authorization` header. See [Connect a client](/mcp/connect.md).

**What can the agent access?** With OAuth, whatever your user can do in the account and environment you picked at sign-in — the same permissions you have in integrator.io. With a token, whatever the token's scope allows.

**US or EU?** The endpoint is region-specific and credentials do not cross regions. Use `api.integrator.io` for US, `api.eu.integrator.io` for EU. OAuth sign-in routes to the tenant that matches the endpoint.

**Which AI clients work?** Any MCP client that supports Streamable HTTP. OAuth sign-in additionally requires a client that implements the MCP authorization spec — current releases of Claude, Cursor, VS Code, Windsurf, and ChatGPT all do. Clients without OAuth support connect with an API token. See the [per-client guides](/mcp/connect.md).

## Next steps

* [Connect a client](/mcp/connect.md): get connected in a few minutes.
* [What you can do](/mcp/use-cases.md): examples and ready-to-use prompts.
* [Tools](/mcp/tools.md): how the toolset works, and the full catalog.


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter:

```
GET https://developer.celigo.com/mcp/readme.md?ask=<question>&goal=<endgoal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.
