Overview
The Heroku Platform MCP Server is an implementation of the Model Context Protocol (MCP) designed to enable seamless interaction between large language models and the Heroku Platform. It exposes a robust set of tools that let LLMs read, manage, and operate Heroku resources through structured commands and workflows. The server emphasizes secure and authenticated access to Heroku Platform APIs by leveraging the global Heroku CLI, allowing users to authenticate once and reuse the session for MCP operations. It supports several deployment and configuration methods, including starting the server with heroku mcp:start (which uses the current CLI context) or launching via npx with a dedicated HEROKU_API_KEY for programmatic access. Available tools cover core domains such as applications, dynos, add-ons, maintenance and logs, pipelines, teams and spaces, and PostgreSQL databases. The project is currently in early development, with ongoing enhancements to functionality and tooling. Guidance is provided for integrating the server with popular MCP clients (Claude Desktop, Zed, Cursor, Windsurf, VSCode, and others).
Features
Direct interaction with Heroku Platform resources
Enable LLMs to read, manage, and operate Heroku Platform resources through a suite of MCP tools.
Secure and authenticated access via the Heroku CLI
Provides secure authentication context using the global Heroku CLI, avoiding manual API key handling when using mcp:start.
Natural language interface
Offers a natural language interface for Heroku Platform interactions, enabling conversational control of apps and resources.
Broad toolset for core platform areas
Includes tools for application management, dyno control, add-ons, maintenance, logs, pipelines, teams and spaces, and PostgreSQL management.
Debugging and development workflow support
Integrated debugging via MCP Inspector and VS Code Run and Debug launcher, supporting local development and testing.
Client integrations and deployment options
Provides configuration guidance for multiple MCP clients and deployment methods (mcp:start and npx) to suit different workflows.
Who Is This For?
- Developers:Use MCP server to manage Heroku apps, dynos, add-ons, and databases via LLMs.
- ML Engineers:Enable LLM-driven workflows to read and operate Heroku resources through controlled tools.
- DevOps Teams:Orchestrate deployments, pipelines, and infrastructure changes using natural language interactions.
