# MCP server

> A Model Context Protocol (MCP) server exposes tools, resources, and prompts to AI agent clients through a structured protocol, as LocalCloud does at /mcp.
>
> Source: https://local.cloud/glossary/mcp-server/

Glossary

A Model Context Protocol server exposes tools, resources, or prompts that agent clients can call through a structured protocol instead of ad-hoc shell instructions.

[Read docs](https://local.cloud/docs/) [Check compatibility](https://local.cloud/compatibility/) [SDK examples](https://local.cloud/docs/sdk-examples/)

In LocalCloud

## MCP server with LocalCloud

- LocalCloud exposes its runtime-owned MCP server at the canonical /mcp endpoint.
- Use the localcloud mcp stdio bridge when an MCP client cannot connect with Streamable HTTP.

Copy-ready setup

## Commands and prompts

### Copy a prompt for your coding agent

#### Start LocalCloud

Give an agent one URL and have it start the local GCP sandbox.

```
Fetch https://local.cloud/ai/agents.md and follow the instructions to start LocalCloud on my machine. Install the LocalCloud CLI if it is missing, verify Docker, start or reuse LocalCloud, export emulator environment variables, and run one local GCP SDK/API smoke check. Do not ask for or use real GCP credentials.
```

Production boundary

## Before production

Use LocalCloud for non-production development, testing, and CI under the [Public Preview License](https://local.cloud/license/), check [each service's local boundaries](https://local.cloud/compatibility/), and validate against real Google Cloud before you deploy.

Related

## Next steps

- [Agent resources](https://local.cloud/ai/) — Runtime endpoints, agent rules, and the AGENTS.md template.
- [Agent setup guides](https://local.cloud/agents/) — Run coding agents against LocalCloud.