# Finout Docs MCP Integration (Beta)

### Overview

The Finout Docs MCP server gives AI assistants read-only access to the published Finout documentation. Once connected, you can ask how-to and conceptual questions about Finout in natural language and get answers grounded in the live docs — without leaving your AI tool.

Use it to:

* Ask how-to questions about Finout features (e.g., *"How do I connect OpenAI?"*)
* Look up prerequisites before starting an integration
* Get guidance on Virtual Tags, Dashboards, Cost Centers, and more
* Combine doc lookups with live Finout data when used alongside the [Finout MCP Integration (Beta)](https://docs.finout.io/ai-and-developer-tools/finout-mcp-integration-beta)

{% hint style="info" %}
Looking for live access to your Finout cost data instead of the documentation? See [Finout MCP Integration (Beta)](https://docs.finout.io/ai-and-developer-tools/finout-mcp-integration-beta). The two MCP servers are complementary — many users connect both.
{% endhint %}

### Prerequisites

Before connecting, make sure you have:

* An MCP-compatible AI assistant (e.g., [Claude](https://claude.ai/), [Cursor](https://www.cursor.com/), or VS Code with MCP support)
* The ability to configure MCP servers in your AI tool

No Finout account credentials are required. The Finout Docs MCP server provides access to publicly published documentation only.

### MCP Server URL

The Finout Docs MCP server is available at: <https://docs.finout.io/\\~gitbook/mcp>

{% hint style="warning" %}
Visiting this URL directly in a browser returns an error. The endpoint is intended for MCP-compatible AI tools, not browser access.
{% endhint %}

#### Getting Connected

{% stepper %}
{% step %}

### Configure the Finout Docs MCP Server

{% tabs %}
{% tab title="Claude" %}

1. Open **Settings → Connectors**
2. Click **Add Custom Connector**
3. Enter the following details and click **Add**:

| Field | Value                                 |
| ----- | ------------------------------------- |
| Name  | Finout Docs                           |
| URL   | `https://docs.finout.io/~gitbook/mcp` |

<figure><img src="/files/rj9wab5XBWHh2iVH0B8e" alt=""><figcaption></figcaption></figure>
{% endtab %}

{% tab title="Cursor" %}

1. Press **Cmd/Ctrl + Shift + P** to open the command palette
2. Type **MCP** and select **Open MCP Settings**
3. Add the following to your `mcp.json` file:

```json
{
  "mcpServers": {
    "finout-docs": {
      "url": "https://docs.finout.io/~gitbook/mcp"
    }
  }
}
```

4. Save the file
   {% endtab %}

{% tab title="Other clients" %}
Any MCP-compatible AI tool that supports remote connections over HTTP transport will work. Point it at `https://docs.finout.io/~gitbook/mcp`.
{% endtab %}
{% endtabs %}
{% endstep %}

{% step %}

### Start Asking Questions

Open a new chat and ask a question about Finout. Your AI assistant searches the Finout documentation, retrieves the relevant pages, and answers using the published content.

<figure><img src="/files/KzfQsJBpbS3n0ggITuik" alt=""><figcaption></figcaption></figure>
{% endstep %}
{% endstepper %}

### Example Questions

Once connected, ask questions like:

* "What are the prerequisites to connect Datadog to Finout?"
* "How do AI Virtual Tags work?"
* "Walk me through setting up an OpenAI Cost Center."
* "How does Finout handle unallocated Kubernetes costs?"
* "What's the difference between MegaBill and a Data Explorer?"

### Why use this instead of letting your AI assistant search the web?

The Finout documentation is publicly available, so AI tools with web access can already read it. The Docs MCP server is still useful in a few specific cases:

* **Tools without web access.** Many MCP clients — Cursor, VS Code, and internal agents — don't browse the web, or have it disabled by policy. The MCP server is often the only way for those tools to read Finout docs.
* **Direct, structured retrieval.** The MCP queries the documentation directly, rather than relying on a search engine's index. Results are current, scoped to the docs, and not mixed with third-party pages or stale caches.
* **Lower hallucination risk on Finout-specific questions.** Responses are grounded in retrieved pages your AI client can show you, rather than the model's training data or general web results.
* **Pairs with the** [**Finout MCP Integration (Beta)**](https://docs.finout.io/ai-and-developer-tools/finout-mcp-integration-beta)**.** Connect both, and your AI assistant can answer questions that combine how-to guidance with your live cost data in a single conversation — for example, *"What does the docs say about Reserved Instance coverage, and what does our actual coverage look like?"*

### FAQ

**My AI assistant can't connect**

Verify the URL is entered correctly (`https://docs.finout.io/~gitbook/mcp`) and that your tool supports MCP over HTTP transport. Check your AI tool's documentation for MCP-specific setup requirements.

**The assistant isn't finding relevant content**

Rephrase your question to be more specific. For example, instead of "how to set up an OpenAI connector," ask "how do I connect an OpenAI Cost Center in Finout."

Need more help? Contact Finout support at <support@finout.io>.


---

# Agent Instructions: 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:

```
GET https://docs.finout.io/ai-and-developer-tools/finout-docs-mcp-integration-beta.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
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.