Build any custom app and action with Heroku.

With Heroku, developers can extend Agentforce with custom logic, dynamic data, and secure execution. Use Heroku to connect any service, build agent actions using your favorite programming language, and ground AI agents in your application’s real-time context — all on the AI PaaS trusted by Salesforce teams.

See how these businesses bring custom apps to life with Heroku.

Custom Agents with Agentforce + Heroku MCP

Power agents with real-time data, scalable compute, and trusted integration using Heroku MCP (Model Context Protocol).

Heroku works seamlessly with Salesforce, making it easy to bring cross-cloud intelligence to Service Cloud, Sales Cloud, and Agentforce. Empower your teams to deploy AI agents that are context-aware, action-ready, and grounded in the systems that drive your business.

Heroku provides a secure, fully managed runtime for executing AI-driven workflows. With a MCP server hosted on Heroku, you can define what tools your AI agents can use and enforce fine-grained controls for how actions are performed, ensuring every agent task runs safely within your architecture.

MCP on Heroku empowers developers to define agent tools as code, deploy them via Git, and manage them using standard CLI workflows. Teams can build and iterate on agentic capabilities faster, while staying aligned with enterprise governance and DevOps standards.

Heroku FAQ

Resource-based pricing means customers only pay for the resources they’ve used and deployed. Instead of evaluating your usage based on which features or tools you use, it evaluates how many resources you consume over a given time period.

All Heroku applications run in a collection of lightweight Linux containers called dynos. Dynos are isolated, virtualized containers that are automatically managed by the platform and designed to execute code based on a user-specified command. The usage of dynos is calculated into our resource-based pricing model.

Heroku add-ons are components that support your application, such as data storage, monitoring, analytics, and data processing. These are fully maintained for you by either a third-party provider or by Heroku. Add-ons exist so developers can focus on their own application logic and not the additional complexity of keeping supporting services running at full production capacity.

Heroku Connect is an add-on that synchronizes data between your Salesforce organization and a Heroku Postgres database. Using Heroku Connect with Heroku Postgres, you can build applications that interact with your Salesforce data using your favorite tools – in all languages and frameworks supported by Heroku.

The Heroku Dev Starter Package is designed to help enterprises quickly start using Heroku. This package includes 20 dynos, 1,000 add-on credits, and 5,000,000 rows of Heroku Connect. Resources in this package are not limited to just development and can be leveraged for QA, staging, or production workloads. Some small applications and services may be able to run production workloads leveraging the Heroku Dev Starter Package. The Heroku resources you need will be dependent on the applications and services you are running on Heroku.

The Heroku Production Starter Package is designed to help enterprises get started using Heroku for production applications. This package includes 200 dynos, 10,000 add-on credits, 1 billion rows of Heroku Connect, and one Heroku Private Space. Resources in this package can be used for development, QA, staging, and/or production workloads. This package could run multiple production applications or services. The Heroku resources you need will be dependent on the applications and services you are running on Heroku.