> For the complete documentation index, see [llms.txt](https://docs.complycube.com/documentation/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.complycube.com/documentation/integration-resources/model-context-protocol-mcp.md).

# Model Context Protocol (MCP)

### Overview

Use the **ComplyCube Model Context Protocol (MCP) server** to connect **Claude**, **Cursor**, **ChatGPT**, **Gemini**, **Perplexity**, **VS Code extensions,** **Windsurf**, and other **MCP-compatible AI assistants** to ComplyCube documentation.

Use it to search ComplyCube docs, read full pages, and **generate code** grounded in current ComplyCube guidance.

This guide explains how ComplyCube MCP works, how to connect it, and how to use it for API, SDK, workflow, and webhook integrations.

{% hint style="info" %}
The ComplyCube MCP server provides read-only access to publicly available documentation. It cannot view or interact with your API keys, Portal account, client records, or verification data.
{% endhint %}

### What is MCP?

[Model Context Protocol](https://modelcontextprotocol.io/) is an open standard that lets AI assistants connect to external tools and data sources.

The ComplyCube MCP server gives your AI assistant access to this documentation. Once connected, the assistant can search the docs and read full pages while it answers questions or drafts code.

This helps your assistant use current ComplyCube documentation instead of relying only on older training data.

The ComplyCube MCP server is read-only. It only serves documentation. It does not have access to your API keys, Portal, clients, checks, cases, or Sandbox and Live account data.

{% hint style="info" %}
Use the MCP server to give your AI assistant access to ComplyCube documentation. It provides product knowledge and API context, while your application and backend continue to handle business logic and verification workflows.
{% endhint %}

### Works with every ComplyCube product

The MCP server covers the full ComplyCube documentation, so you can use it across the platform.

#### Verification checks

Ask about specific checks and verification flows:

* [Extensive AML screening](/documentation/product-guides/watchlist-pep-and-adverse-media/aml-screening-check.md) for sanctions, PEPs, and adverse media checks.
* [Document checks](/documentation/product-guides/identity-verification/document-check.md) for document authenticity and data extraction.
* [Identity checks](/documentation/quick-guides/api-quick-guide/run-an-identity-check.md) for liveness and biometric face match.
* [Proof of address checks](/documentation/quick-guides/api-quick-guide/perform-proof-of-address-check.md) for address document validation.
* [Multi-bureau checks](/documentation/product-guides/identity-verification/multi-bureau-check.md) for bureau and trusted data-source matching.

#### Integration paths

Use MCP to compare integration options and implementation details:

* [Hosted Solution](/documentation/quick-guides/hosted-solution-quick-guide.md) for a fully hosted flow.
* [Web SDK](/documentation/quick-guides/web-sdk-quick-guide.md) and [Mobile SDKs](/documentation/quick-guides/mobile-sdk-guide.md) for embedded experiences.
* [API Reference](https://docs.complycube.com/documentation/api-reference/) for an API integrations with granular controls.

#### Developer tools and operations

Use MCP to find exact implementation details fast:

* [Webhooks](/documentation/integration-resources/webhooks.md) for event handling and async updates.
* [Postman](/documentation/integration-resources/postman.md) and [testing data](/documentation/integration-resources/testing-data.md) for sandbox testing and debugging.
* [Compliance guidance](/documentation/trust-and-compliance/compliance.md) for trust, controls, and privacy coverage.

You can ask your AI assistant about any of these areas and it will search the right pages for you.

### What you can do

* Search the docs in plain language.
* Read full pages when a snippet is not enough.
* Draft integration code grounded in current ComplyCube guides.
* Debug headers, webhook payloads, and parameter names without leaving your editor.

In most AI clients, you do not need to call these directly. Ask your question normally and let the assistant choose when to **search** or **read**.

### Connect your AI assistant to the ComplyCube MCP server

You do not need to install anything locally. The ComplyCube MCP server runs online.

Copy this URL into your MCP client:

{% code title="Server URL" %}

```
https://docs.complycube.com/documentation/~gitbook/mcp
```

{% endcode %}

Most MCP-compatible clients let you add a remote server from their settings or connector menu.

Use these connection details:

* **Name:** ComplyCube Docs
* **Authentication:** None

Then:

1. Open your AI assistant's **MCP**, **connectors**, or **integrations** settings.
2. Add a new **remote MCP server**.
3. Enter the **ComplyCube Docs** name and URL.
4. **Save** the connection.
5. **Enable** the server in the workspace or chat where you want to use it.

{% hint style="info" %}
MCP setup steps vary by client and change often. If your assistant supports remote MCP servers, use the server URL above and follow that client's current setup flow.
{% endhint %}

### How to use the ComplyCube MCP server

After you connect the server, ask your assistant questions in **plain language**. You do not need to call the documentation tools yourself.

#### Find information in ComplyCube docs

Try prompts like:

* "How do I create a client and run a document check in Sandbox?"
* "Which ComplyCube integration path fits a hosted onboarding flow with minimal code?"
* "Which webhook events should I handle after a check completes?"

#### Generate ComplyCube integration code

Ask for code grounded in **current docs**, for example:

* "Write Node.js code that creates a client and starts a hosted verification flow".
* "Write a Python example that creates a check and retrieves the result".

For concrete implementation patterns, start with the [API Quick Guide](/documentation/quick-guides/api-quick-guide.md), [Hosted Solution Guide](/documentation/quick-guides/hosted-solution-quick-guide.md), [Web SDK Guide](/documentation/quick-guides/web-sdk-quick-guide.md), or [Mobile SDK Guide](/documentation/quick-guides/mobile-sdk-guide.md).

#### Debug your ComplyCube integration

Use the MCP server to inspect exact **field names** and **flows**:

* "Why does my webhook signature validation fail?"
* "Which parameters are required when I generate an SDK token?"

### Best practices for better prompts

* State whether you want a **workflow-based** or **check-based** integration path.
* Be specific about whether you use [Hosted Solution](/documentation/quick-guides/hosted-solution-quick-guide.md), [Web SDK](/documentation/quick-guides/web-sdk-quick-guide.md), [Mobile SDKs](/documentation/quick-guides/mobile-sdk-guide.md), or [direct API integration](/documentation/quick-guides/api-quick-guide.md).
* Ask for exact endpoint names, payload fields, and sequence steps.
* Keep **API keys**, **secrets**, and **personal data** out of prompts.
* Search first, then ask for full-page detail when needed.
* Always review generated code before you ship it.

### Privacy and scope

The MCP server only serves content from [ComplyCube documentation](/documentation/master.md).

It does not have access to your **Portal**, **sessions**, **API keys**, or **customer data**.

Your prompts and the returned documentation flow through your AI assistant provider. Check your provider's data policy before sharing sensitive information.


---

# 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://docs.complycube.com/documentation/integration-resources/model-context-protocol-mcp.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.