Introducing Rover, AI coding agent manager

We are happy to announce today the release of Rover, an Open Source manager for AI coding agents.

Rover helps you get more done, faster, by allowing multiple agents to work on your codebase simultaneously. The agents work in the background in separate, isolated environments: they don’t interfere with your work or each other.

Rover does not change how you work: everything runs locally, under your control, and using your already installed tools and agents. It works with Claude Code, OpenAI Codex, Google Gemini and Qwen, with support for more on the way.

Getting started with Rover is as easy as installing an npm package and typing rover init in your code repository:

npm install -g @endorhq/rover@latest
rover init

From that point on, you can start creating “tasks” that need to be done, like implementing certain features or refactoring part of the code. Rover will take care of:

• 📦 Preparing local isolated environments using containers
• 🤖 Installing and configuring your preferred AI coding agent in that environment
• 🚀 Getting them going until they finish

Each environment is independent from each other and has its own copy of the code, so you can run multiple agents in parallel, without any extra effort on your part. Rover will take care of managing those agents and will reach out if it needs any input. You can always proactively monitor their progress and provide additional instructions or feedback as needed.

Once the agents are done, Rover will create a PR-ready version of the code changes as well as a set of developer-friendly documents about what they did and why, to help you review.

Rover is also available as a Visual Studio Code extension for those who prefer working with an IDE and is available from the built-in extension catalog.

We use coding agents heavily and we initially built Rover based on our experience and needs. We found it so useful that we started sharing it with other developer friends and ultimately decided to open source and make it available to everyone. We hope you find it as useful as we do and look forward to your feedback and contributions!