DEV Community

Google Developer Experts profile image xbill
xbill for Google Developer Experts

Posted on • Originally published at Medium on

MCP Development with Dart/Flutter, Cloud Run, and Gemini CLI

#aitools #mcpserver #googlegeminicli #googlecloudrun

Leveraging Gemini CLI and the underlying Gemini LLM to build Model Context Protocol (MCP) AI applications using Flutter/Dart with a local development environment.

Why not just use Python?

Python has traditionally been the main coding language for ML and AI tools. One of the strengths of the MCP protocol is that the actual implementation details are independent of the development language. The reality is that not every project is coded in Python- and MCP allows you to use the latest AI approaches with other coding languages.

Flutter/Dart? For AI? Does that even work?

The goal of this article is to provide a minimal viable basic working MCP stdio server in Dart that can be run locally without any unneeded extra code or extensions.

The key MCP Dart package is here:

mcp_dart | Dart package

What Is Flutter / Dart?

Flutter is Google’s open-source UI toolkit and framework used for building natively compiled, multi-platform applications (mobile, web, desktop, and embedded systems) from a single codebase. Dart is the programming language , also developed by Google, in which Flutter apps are written.

The full details of the language can be found here:

Intro to Dart

Installing Dart

Step by Step instructions vary by platform — a good starting point is here:

Install Flutter manually

For a Debian style Linux Distro these steps are similar to:

sudo apt-get update -y && sudo apt-get upgrade -y
sudo apt-get install -y curl git unzip xz-utils zip libglu1-mesa
Enter fullscreen mode Exit fullscreen mode

The download and extract the latest Flutter Bundle and extract to a common directory like ~/develop :

tar -xf <sdk_zip_path> -C <destination_directory_path>
Enter fullscreen mode Exit fullscreen mode

As of writing the current version is:

tar -xf /mnt/chromeos/MyFiles/Downloads/flutter_linux_3.38.5-stable.tar.xz -C ~/develop/
Enter fullscreen mode Exit fullscreen mode

You can validate the working tools with the — version command:

xbill@penguin:~$ dart --version
Dart SDK version: 3.10.4 (stable) (Tue Dec 9 00:01:55 2025 -0800) on "linux_x64"
xbill@penguin:~$ flutter --version
Flutter 3.38.5 • channel stable • https://github.com/flutter/flutter.git
Framework • revision f6ff1529fd (3 weeks ago) • 2025-12-11 11:50:07 -0500
Engine • hash c108a94d7a8273e112339e6c6833daa06e723a54 (revision 1527ae0ec5) (23 days ago) • 2025-12-11 15:04:31.000Z
Tools • Dart 3.10.4 • DevTools 2.51.1
Enter fullscreen mode Exit fullscreen mode

Gemini CLI

If not pre-installed you can download the Gemini CLI to interact with the source files and provide real-time assistance:

npm install -g @google/gemini-cli
Enter fullscreen mode Exit fullscreen mode

Testing the Gemini CLI Environment

Once you have all the tools and the correct Node.js version in place- you can test the startup of Gemini CLI. You will need to authenticate with a Key or your Google Account:

gemini
Enter fullscreen mode Exit fullscreen mode

Dart Gemini CLI Extension

To simplify working with Dart/Flutter with Gemini CLI- a Gemini CLI extension is available:

Flutter extension for Gemini CLI

The installation will look similar to this:

xbill@penguin:~/gemini-cli-codeassist/mcp-stdio-flutter$ 
gemini extensions install https://github.com/gemini-cli-extensions/flutter.git --auto-update
Installing extension "flutter".
**The extension you are about to install may have been created by a third-party developer and sourced from a public repository. Google does not vet, endorse, or guarantee the functionality or security of extensions. Please carefully inspect any extension and its source code before installing to understand the permissions it requires and the actions it may perform.**
This extension will run the following MCP servers:
  * dart (local): dart mcp-server
This extension will append info to your gemini.md context using flutter.md
Do you want to continue? [Y/n]: 
Extension "flutter" installed successfully and enabled.
xbill@penguin:~/gemini-cli-codeassist/mcp-stdio-flutter$
Enter fullscreen mode Exit fullscreen mode

Node Version Management

Gemini CLI needs a consistent, up to date version of Node. The nvm command can be used to get a standard Node environment:

GitHub - nvm-sh/nvm: Node Version Manager - POSIX-compliant bash script to manage multiple active node.js versions

Where do I start?

The strategy for starting MCP development is a incremental step by step approach.

First, the basic development environment is setup with the required system variables, and a working Gemini CLI configuration.

Then, a minimal Hello World Style Dart MCP Server is built with HTTPS transport. This server is validated with Gemini CLI in the local environment.

This setup validates the connection from Gemini CLI to the local process via MCP. The MCP client (Gemini CLI) and the Dart MCP server both run in the same local environment.

Next- the basic MCP server is extended with Gemini CLI for deployment to Google Cloud Run. This remote HTTP connection is validated with a local Gemini CLI instance.

Setup the Basic Environment

At this point you should have a working Dart/Flutter environment and a working Gemini CLI installation. The next step is to clone the GitHub samples repository with support scripts:

cd ~
git clone https://github.com/xbill9/gemini-cli-codeassist
Enter fullscreen mode Exit fullscreen mode

Then run init.sh from the cloned directory.

The script will attempt to determine your shell environment and set the correct variables:

cd gemini-cli-codeassist
source init.sh
Enter fullscreen mode Exit fullscreen mode

If your session times out or you need to re-authenticate- you can run the set_env.sh script to reset your environment variables:

cd gemini-cli-codeassist
source set_env.sh
Enter fullscreen mode Exit fullscreen mode

Variables like PROJECT_ID need to be setup for use in the various build scripts- so the set_env script can be used to reset the environment if you time-out.

Hello World with HTTP Transport

One of the key features that the standard MCP libraries provide is abstracting various transport methods.

The high level MCP tool implementation is the same no matter what low level transport channel/method that the MCP Client uses to connect to a MCP Server.

The simplest transport that the SDK supports is the stdio (stdio/stdout) transport — which connects a locally running process. Both the MCP client and MCP Server must be running in the same environment.

The HTTP(S) transport allows the client and server to be on the same host or distributed over the Internet.

The connection over streaming http will look similar to this:

      final server = StreamableMcpServer(
        serverFactory: (sessionId) => createServer(),
        host: host,
        port: port,
        path: path,
      );
Enter fullscreen mode Exit fullscreen mode

Reviewing the Dart Code

First- switch the directory with the Dart MCP sample code:

cd ~/gemini-cli-codeassist/mcp-https-flutter
Enter fullscreen mode Exit fullscreen mode

Current Version of Dart dependencies:

name: mcp_https_flutter
description: A Model Context Protocol (MCP) server implementation in Dart supporting HTTPS and SSE transport.
version: 1.0.0
repository: https://github.com/xbill9/gemini-cli-codeassist/mcp-https-flutter

environment:
  sdk: ^3.10.4

# Add regular dependencies here.
dependencies:
  path: ^1.9.0
  mcp_dart: ^1.2.0
  args: ^2.5.0
  logging: ^1.2.0
Enter fullscreen mode Exit fullscreen mode

Gemini CLI is used to review the project:

✦ I have completed the review of the project. Overall, the codebase is well-structured and follows many Dart best practices,
  especially regarding logging and server configuration.

  Key Strengths
   * Structured Logging: Excellent implementation using the logging package with JSON output to stderr, which is ideal for cloud
     environments and monitoring.
   * Robust Output Handling: Using runZoned to redirect print statements to the logger ensures that any stray output is captured
     and formatted correctly.
   * Clean CLI Interface: Proper use of ArgParser for configuration and StreamableMcpServer for the HTTP/SSE transport.
   * Clear Separation: The server logic is separated from the entry point, facilitating testing.
Enter fullscreen mode Exit fullscreen mode

Installing and Running the Dart Code

Run the build make target on the local system:

xbill@penguin:~/gemini-cli-codeassist/mcp-stdio-flutter$ make release
dart analyze
Analyzing mcp-https-flutter... 0.2s
No issues found!
dart test
00:00 +3: All tests passed!                                                                                                     
dart compile exe bin/mcp_https_flutter.dart -o bin/mcp_https_flutter
Generated: /home/xbill/gemini-cli-codeassist/mcp-https-flutter/bin/mcp_https_flutter
Enter fullscreen mode Exit fullscreen mode

To test the Dart Code:

xbill@penguin:~/gemini-cli-codeassist/mcp-https-flutter$ make check
dart analyze
Analyzing mcp-https-flutter... 0.2s
No issues found!
dart test
00:00 +3: All tests passed!
Enter fullscreen mode Exit fullscreen mode

Gemini CLI settings.json

The default Gemini CLI settings.json has an entry for the source:

 {
    "mcpServers": {
    "hello-https-flutter": {
      "httpUrl": "http://127.0.0.1:8080/mcp"
    }
  }
}
Enter fullscreen mode Exit fullscreen mode

Starting the Local HTTP Server

The local server is started with the make run target:

xbill@penguin:~/gemini-cli-codeassist/mcp-https-flutter$ make run
dart run bin/mcp_https_flutter.dart --port 8080 --host 0.0.0.0 --path /mcp
Enter fullscreen mode Exit fullscreen mode

The 0.0.0.0 address tells the development server to listen on all network interfaces. Port 8080 is a commonly used port for the connection.

You will see basic JSON logging:

{"timestamp":"2026-01-05T21:07:24.833764Z","level":"INFO","logger":"mcp_server","message":"Starting MCP server on http://0.0.0.0:8080/mcp"}
{"timestamp":"2026-01-05T21:07:24.854915Z","level":"INFO","logger":"StreamableMcpServer","message":"MCP Streamable HTTP Server listening on http://0.0.0.0:8080/mcp"}
Enter fullscreen mode Exit fullscreen mode

Validation with Gemini CLI

Leave the server running and open new terminal window. Gemini CLI is restarted and the MCP connection over stdio to the Dart Code is validated, The full Gemini CLI Session will start:

gemini

> /mcp list

Configured MCP servers:

🟢 hello-https-flutter - Ready (1 tool)
  Tools:
  - greet

> greet Dash!

✦ I will read the project files to understand how the greet tool is implemented.
╭───────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────╮
│ ✓ ReadFolder lib │
│ │
│ Listed 1 item(s). │
│ │
│ ✓ ReadFile bin/mcp_https_flutter.dart │
│ │
╰───────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────╯
✦ I will read lib/mcp_https_flutter.dart to find the implementation of greetHandler.
╭───────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────╮
│ ✓ ReadFile lib/mcp_https_flutter.dart │
│ │
╰───────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────╯
✦ I will call the greet tool with the parameter "Dash!".
╭───────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────╮
│ ? greet (hello-https-flutter MCP Server) {"param":"Dash!"} ← │
│ │
│ MCP Server: hello-https-flutter │
│ Tool: greet │
│ │
│ Allow execution of MCP tool "greet" from server "hello-https-flutter"? │
│ │
│ 1. Allow once │
│ 2. Allow tool for this session │
│ ● 3. Allow all server tools for this session │
│ 4. No, suggest changes (esc) │
│ │
╰───────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────╯
✦ I will call the greet tool with the parameter "Dash!".
╭───────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────╮
│ ✓ greet (hello-https-flutter MCP Server) {"param":"Dash!"} │
│ │
│ Dash! │
╰───────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────╯
✦ Dash
Enter fullscreen mode Exit fullscreen mode

Using the Dart Gemini CLI Extension

The Gemini CLI Extension provides tools for working with Dart Code:

> run_tests

✦ I will run the tests for the project to ensure everything is working correctly.

╭───────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────╮
│ ✓ run_tests (dart MCP Server) {"roots":[{"root":"file:///home/xbill/gemini-cli-codeassist/mcp-https-flutter"}]} │
│ │
│ dart|flutter test in /home/xbill/gemini-cli-codeassist/mcp-https-flutter: │
│ \u001b[32m+3\u001b[0m: All tests passed!\u001b[0m │
╰───────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────╯
✦ All tests passed successfully. What would you like to do next?
Enter fullscreen mode Exit fullscreen mode

Deploying to Cloud Run

After the local HTTP version of the MCP server has been tested locally — it can be deployed remotely to Google Cloud Run.

First- switch to the directory with the HTTP MCP sample code:

cd ~/gemini-cli-codeassist/mcp-https-flutter
Enter fullscreen mode Exit fullscreen mode

Deploy the project to Google Cloud Run with the pre-built cloudbuild.yaml and Dockerfile:

cd ~/gemini-cli-codeassist/mcp-https-flutter
xbill@penguin:~/gemini-cli-codeassist/mcp-https-flutter$ make deploy
Enter fullscreen mode Exit fullscreen mode

The Cloud Build will start:


Submitting build to Google Cloud Build...
Creating temporary archive of 17 file(s) totalling 7.0 MiB before compression.
Some files were not included in the source upload.

Check the gcloud log [/home/xbill/.config/gcloud/logs/2026.01.05/16.14.12.250917.log] to see which files and the contents of the
default gcloudignore file used (see `$ gcloud topic gcloudignore` to learn
more).
Enter fullscreen mode Exit fullscreen mode

It can take 5–10 minutes to complete the build.

When the build is complete- an endpoint will be returned:

Step #2: Already have image (with digest): gcr.io/cloud-builders/gcloud
Step #2: Deploying container to Cloud Run service [mcp-https-flutter] in project [comglitn] region [us-central1]
Step #2: Deploying...
Step #2: Setting IAM Policy.............done
Step #2: Creating Revision.....................................................................................................................................................................done
Step #2: Routing traffic.....done
Step #2: Done.
Step #2: Service [mcp-https-flutter] revision [mcp-https-flutter-00003-l94] has been deployed and is serving 100 percent of traffic.
Step #2: Service URL: https://mcp-https-flutter-1056842563084.us-central1.run.app
Finished Step #2
Enter fullscreen mode Exit fullscreen mode

The service endpoint in this example is :

https://mcp-https-flutter-1056842563084.us-central1.run.app
Enter fullscreen mode Exit fullscreen mode

The actual endpoint will vary based on your project settings.

Review Service in Cloud Run

Navigate to the Google Cloud console and search for Cloud Run -

and then you can detailed information on the Cloud Run Service:

Cloud Logging

The remote server writes logs to stderr in standard JSON format. These logs are available from the deployed Cloud Run Service:

Validate HTTP connection

Once you have the Endpoint — you can attempt a connection- navigate to in your browser:

https://mcp-https-flutter-1056842563084.us-central1.run.app
Enter fullscreen mode Exit fullscreen mode

You will need to adjust the exact URL to match the URL returned from Cloud Build.

You will get an error- this connection is expecting a message in the MCP format:

Not Found
Enter fullscreen mode Exit fullscreen mode

Gemini CLI settings.json.cloudrun

Replace the default Gemini CLI configuration file —  settings.json with a pre-configured sample- settings.json.cloudrun to use the Cloud Run version of the connection:

 {
    "mcpServers": {
    "hello-cloudrun-flutter": {
      "httpUrl": "https://mcp-https-flutter-$PROJECT_NUMBER.us-central1.run.app/mcp"
    }
  }
}
Enter fullscreen mode Exit fullscreen mode

Copy the Cloud Run version of the Gemini CLI configuration file:

xbill@penguin:~/gemini-cli-codeassist/mcp-https-php$ cd .gemini
cp settings.json.cloudrun settings.json
xbill@penguin:~/gemini-cli-codeassist/mcp-https-php/.gemini$
Enter fullscreen mode Exit fullscreen mode

Validation with Gemini CLI

The final connection test uses Gemini CLI as a MCP client with the deployed Cloud Run Service in PHP providing the MCP server. Startup Gemini CLI with the updated settings :

gemini

> /mcp list

Configured MCP servers:

🟢 hello-cloudrun-flutter - Ready (1 tool)
  Tools:
  - greet

 > greet Cloud Dash!

✦ I will greet Cloud Dash for you.

╭───────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────╮
│ ? greet (hello-cloudrun-flutter MCP Server) {"param":"Cloud Dash!"} ← │
│ │
│ MCP Server: hello-cloudrun-flutter │
│ Tool: greet │
│ │
│ Allow execution of MCP tool "greet" from server "hello-cloudrun-flutter"? │
│ │
│ 1. Allow once │
│ 2. Allow tool for this session │
│ ● 3. Allow all server tools for this session │
│ 4. No, suggest changes (esc) │
│ │
╰───────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────╯

╭───────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────╮
│ ✓ greet (hello-cloudrun-flutter MCP Server) {"param":"Cloud Dash!"} │
│ │
│ Cloud Dash! │
╰───────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────╯
✦ OK. Cloud Dash
Enter fullscreen mode Exit fullscreen mode

Summary

MCP development with Flutter/Dart using Gemini CLI was validated with an incremental step by step approach.

A minimal streaming HTTP transport MCP Server was built from source code and validated with Gemini CLI running as a MCP client in the same local environment.

Then — the MCP server was wrapped in a container and submitted to Google Cloud Build for deployment to Google Cloud Run. The remote MCP server was validated with a standard browser, and Gemini CLI.

Finally- remote MCP operations were performed from the local Gemini CLI installation to the Flutter/Dart MCP server hosted in Google Cloud Run.

This approach can be extended to more complex deployments and Cloud based options.

Top comments (0)