Enhanced Coolify

You can run an MCP server that lets you control your Coolify infrastructure with natural language. This server handles application deployment, infrastructure automation, monitoring, environment management, and more, so you can manage complex workflows through simple conversations with AI.

How to use

You connect your MCP client to the Enhanced Coolify MCP Server to perform operations across projects, apps, databases, services, and deployments. Use conversational prompts to create projects, deploy applications, add infrastructure, update environment variables, monitor resource usage, view logs, and manage deployments. Start by authenticating with your Coolify server, then issue intents like creating a project, deploying a fullstack setup, or deploying a prebuilt Docker Compose service. The MCP server translates your natural language requests into actionable tasks on your Coolify instance.

How to install

# Prerequisites
# - Node.js 18+ (install from https://nodejs.org)
# - Git
# - Access to a Coolify server with an API token

# Option A: Interactive setup (recommended for beginners)
git clone https://github.com/dazeb/coolify-mcp-enhanced.git
cd coolify-mcp-enhanced
npm install && npm run build
node setup-guide.js

# Option B: Manual setup

# Step 1: Get your Coolify API token (from your Coolify dashboard)

# Step 2: Install the Enhanced MCP Server

git clone https://github.com/dazeb/coolify-mcp-enhanced.git
cd coolify-mcp-enhanced
npm install
npm run build

# Step 3: Configure your environment
export COOLIFY_ACCESS_TOKEN=0|your-actual-token-here
export COOLIFY_BASE_URL=https://your-coolify-server.com

# Step 4: Test the connection
node test-mcpaas.js

# Step 5: Quick verification (runs the MCP server locally)
echo '{"jsonrpc":"2.0","id":1,"method":"tools/call","params":{"name":"list_servers","arguments":{}}}' | node dist/index.js

Additional setup and considerations

Configuration revolves around your Coolify server URL and a valid API token. Use environment variables to provide the connection details to the MCP server. You can integrate the MCP server with AI agents or IDEs by pointing them at the local or remote MCP endpoint and supplying the same authentication parameters.

If you need to run the server within an IDE or automation tool, you can configure the MCP client with the same command and environment variables used on your workstation. For example, in a workflow automation or AI IDE, set the command to node and pass the path to the MCP entry point along with the required tokens and base URL.

Security and troubleshooting notes

Keep your API token secure. Do not commit it to version control or expose it in logs. If you suspect token compromise, regenerate the token in your Coolify dashboard and update the MCP environment accordingly.

If a connection fails, verify the base URL is reachable and the token has the required permissions (read servers, manage projects, manage applications, manage services, and manage databases). Check network access and ensure there are no firewall rules blocking access to your Coolify instance.

For troubleshooting, enable detailed logging to diagnose issues with authentication, token scopes, or MCP command execution by running the server with debug logging enabled if supported by your environment.

Tools and commands overview

The Enhanced MCP Server exposes a comprehensive set of actions to manage your Coolify environment, including listing resources, creating projects and apps, updating environment variables, deploying infrastructure stacks, and handling deployments.

Available tools

list_servers

List all Coolify servers and their basic info

list_projects

List all projects in your Coolify accounts

create_project

Create a new project with specified name and settings

list_applications

List all applications across projects

create_application

Deploy a new application from a repository or template

get_application_logs

Retrieve logs for a specific application

get_application_resources

Monitor resource usage for an application (CPU, memory, etc.)

list_services

List all services within a project or environment

create_service

Create a new service with given parameters

create_docker_compose_service

Deploy a service from a Docker Compose configuration

list_databases

List databases available to your projects

create_database

Create a new database instance (PostgreSQL, MySQL, etc.)

update_database

Update settings for an existing database instance

get_application_environment_variables

Get environment variables for an application

update_application_environment_variables

Update environment variables for an application

get_deployments

Show deployment histories for applications or services

cancel_deployment

Cancel an in-progress deployment

deploy_application

Trigger deployment of an application with specified version or source