> ## Documentation Index
> Fetch the complete documentation index at: https://docs.springwinter.dev/llms.txt
> Use this file to discover all available pages before exploring further.

# Deploy MCPO

> Set up MCPO and connect its dependencies with Spring Winter.

## Deploy with an agent

Copy agent prompt

Help me deploy MCPO. Read [https://docs.springwinter.dev/deploy/open-source/mcpo](https://docs.springwinter.dev/deploy/open-source/mcpo.md), [https://github.com/open-webui/mcpo](https://github.com/open-webui/mcpo), and [https://docs.springwinter.dev/api-reference/agents](https://docs.springwinter.dev/api-reference/agents.md). Fetch the public OpenAPI specification from my Spring Winter origin. Use my local SPRINGWINTER\_TOKEN or existing SPRINGWINTER\_API\_TOKEN. Keep credentials out of chat, logs and source.

Confirm my project, AWS connection, Region and budget. Pin a release. Check its required services, container port, health path, network access and durable storage. Data requirements for this deployment: No SQL database for the proxy. An upstream MCP server may have its own dependencies.

1. Prepare one approved MCP tool server first. For remote MCP, confirm that MCPO can reach its authenticated Streamable HTTP or SSE endpoint. For standard input and output, include the tool and its runtime in the same image.
2. Use ghcr.io/open-webui/mcpo or a GitHub build on port 8000. Supply the upstream URL or command and configure MCPO authentication. Use runtime secret injection for upstream credentials and the proxy key.
3. Restrict network access and expose only reviewed tools. If any tool requires a database, create and connect it in the upstream server before starting MCPO.
4. Start the upstream service first if it runs remotely. Start MCPO, check the generated OpenAPI schema for the expected tools, and then connect the consumer.

Show me the planned resources, database and migration order, secret names and estimated cost. Identify components that require separate hosting. Wait for approval before making changes. Use documented Spring Winter operations and wait for each workflow to finish. Invoke one harmless tool as an authorized client and verify that an unauthorized request is denied. Report resource links and anything you could not verify.

## Before you start

[Connect your AWS account](https://docs.springwinter.dev/connect-aws.md) and choose a project. Review the [upstream deployment instructions](https://github.com/open-webui/mcpo) for the release you plan to pin.

## Data dependencies

No SQL database for the proxy. An upstream MCP server may have its own dependencies.

## Deployment order

1. Prepare one approved MCP tool server first. For remote MCP, confirm that MCPO can reach its authenticated Streamable HTTP or SSE endpoint. For standard input and output, include the tool and its runtime in the same image.
2. Use `ghcr.io/open-webui/mcpo` or a GitHub build on port 8000. Supply the upstream URL or command and configure MCPO authentication. Use runtime secret injection for upstream credentials and the proxy key.
3. Restrict network access and expose only reviewed tools. If any tool requires a database, create and connect it in the upstream server before starting MCPO.
4. Start the upstream service first if it runs remotely. Start MCPO, check the generated OpenAPI schema for the expected tools, and then connect the consumer.

Use [Runtime variables](https://docs.springwinter.dev/concepts/environment-variables.md) for runtime credentials. For a [Spring Winter web server](https://docs.springwinter.dev/deploy/web-servers.md), set an app-supported health path that returns HTTP 200 without credentials. Check **History** before running the application check.

## Verify the deployment

Invoke one harmless tool as an authorized client and verify that an unauthorized request is denied.


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.