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

# MCP

{% @scale-docs-page-title/page-title title="MCP" %}

{% hint style="warning" %}
**Confirm before publishing:** supported MCP transport and server URL, authentication handshake, exact tool names and schemas, public REST operation set, scopes, approval rules and client onboarding. The product MCP servers below are designed surfaces; the current internal agent tools are not a public MCP contract.
{% endhint %}

Model Context Protocol (MCP) lets a compatible AI client discover and call approved Scale tools. The MCP server is a thin adapter over the versioned API: each tool maps to a scoped REST operation, and the REST gateway remains responsible for identity, access, limits and audit.

## Choose a server

| Server                                           | Caller                                   | Data boundary                                    |
| ------------------------------------------------ | ---------------------------------------- | ------------------------------------------------ |
| [ScaleApp MCP](/api/mcp/scaleapp-mcp.md)         | Signed-in ScaleApp customer              | Own profile, households and permitted strategies |
| [ScaleConnect MCP](/api/mcp/scaleconnect-mcp.md) | Broker or approved brokerage integration | Permitted brokerage and client book              |

A server should advertise only tools the connected identity may call. A tool's description and JSON Schema come from the selected public API contract, not the whole R16 Swagger dump.

## Connection flow

1. Choose the product server and authenticate as the person or integration using it.
2. Review the requested scopes and tools.
3. Connect from the MCP client using the server URL and transport in that server's guide.
4. Discover tools and call a read-only tool first.
5. Confirm the returned resource belongs to the expected account or brokerage.
6. Grant write tools only where the workflow needs them.

The final connection block belongs on each product page once server URLs and auth are implemented. Do not use an internal service token, product session copied from a browser or unrestricted database credential.

## Tool execution

Each call has a named input schema, a documented result and a clear failure. Reads may be retried safely; writes require the same validation and authorization as their REST operation. Clients should show a proposed write to the user before executing it when the action changes client data or sends a message. A failed REST permission check must remain a failed MCP call; the adapter must not widen access or retry under a more privileged identity.

MCP resources and prompts can be added later as separate contracts. This section currently specifies tools.


---

# 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 by asking a question.

Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters:

```
GET https://docs.scaleconnect.com.au/api/mcp.md?ask=<question>&goal=<user_goal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case.

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.