Introducing the Official Heroku MCP Server

We’re excited to announce the launch of the Heroku MCP Server, designed to bridge the gap between agent-driven development and Heroku’s AI PaaS. Having defined the platform experience for apps in the cloud, Heroku extends our developer and operator experience to AI capabilities. With the Heroku MCP Server, you can now expose Heroku’s robust platform capabilities as a set of intuitive actions accessible to AI agents through Model Context Protocol (MCP).

The Heroku MCP server enables AI-powered applications like Claude Desktop, Cursor, and Windsurf to directly interface with Heroku, unlocking new levels of automation, efficiency, and intelligence for managing your custom applications.

How Does the Heroku MCP Server Work?

Under the hood, the Heroku MCP Server makes intelligent use of the toolchain developers already trust: the Heroku CLI. It uses the CLI as the primary engine for executing actions, ensuring consistency and benefiting from its existing command orchestration logic.

To maximize performance and responsiveness, especially for sequences of operations, the server runs the Heroku CLI in REPL (Read-Eval-Print Loop) mode. This maintains a persistent CLI process, enabling significantly faster command execution and making multi-tool operations much more efficient compared to launching a new CLI process for every action.

What Can Your Agents Do Today?

The initial release of the Heroku MCP Server focuses on core developer workflows:

App Lifecycle Management: Empower agents to handle deploying, scaling, restarting, viewing logs, and monitoring your applications.
Database Operations: Enable actions on your Heroku Postgres databases.
Add-on Management: Allow agents to discover available add-ons and attach or detach resources to your apps.
Scaling and Performance: Facilitate intelligent scaling of your application resources.

Access the full list of tools here.

Here’s How to Get Started

Authentication Setup

Generate a Heroku authorization token by using the following CLI command.

heroku authorizations:create

Copy the token and use it as your HEROKU_API_KEY in the following steps.

Configuration with MCP Clients

MCP clients maintain the MCP config file in different locations:

Cursor - mcp.json
Windsurf - mcp_config.json
Claude - claude_desktop_config.json

Add the following to the appropriate config file:

{
  "mcpServers": {
    "heroku": {
      "command": "npx -y @heroku/mcp-server",
      "env": {
        "HEROKU_API_KEY": "<YOUR_HEROKU_AUTH_TOKEN>"
      }
    }
  }
}

Other Clients

For integration with other MCP compatible clients, please refer to the client specific config documentation.

Turbo Charge Your IDE and Agents

Heroku's core mission has always been to simplify the complexities of app development and deployment, and a key part of that is meeting developers right where they work: inside their IDE. We've championed this with tools like the Heroku VS Code extension, which brings the power of the Heroku Dashboard and the versatility of the CLI directly into your AI editor like Cursor, reducing the need to switch contexts for many common tasks.

As AI-native developer workflows emerge, the friction between coding environments and cloud platforms will disappear entirely. Developers want to stay focused, leveraging AI assistance without interrupting their flow or needing deep platform-specific expertise for routine tasks.

The Heroku MCP Server builds directly on our philosophy of seamless IDE and agent integration. While the VS Code extension provides excellent visual affordances and manual control for developers, the MCP Server addresses the rise of agent-driven development. It provides an intuitive way for your agents to manage your Heroku applications, databases, and infrastructure, making it an essential part of any AI PaaS (AI Platform as a Service) strategy.

What's Next?

This is just the beginning! We're actively working on exposing even more of the Heroku platform's capabilities through the MCP server. Our goal is to continuously enhance the AI-driven developer experience on Heroku, making it richer, more powerful, and even more intuitive. Stay tuned for updates as we expand the range of actions your agents can perform.

Join the Conversation

The Heroku MCP Server is just one piece of Heroku's plan for providing an excellent AI-driven developer experience, and to provide the primitives necessary to build, manage, and scale AI applications and agents. Stay tuned for next month's GA of our Managed Inference and Agents product, which comes complete with support for a range of MCP tools, and upcoming enhancements to broad MCP support across the platform.

Video Transcript