#25 · Primary category: AI Agents & Automation

daytona

agentic-workflow ai ai-agents ai-runtime ai-sandboxes code-execution code-interpreter developer-tools

Daytona is a Secure and Elastic Infrastructure for Running AI-Generated Code

Project last updated:07/24/26

GitHub Stars

71.9K

Forks

5.6K

Contributors

233

License

Other

Why we included this project

Teams building agents that generate and execute code need somewhere safe for that code to run, and Daytona is a solid reference for how to build that layer. Its core unit is the sandbox: a full isolated computer with its own kernel, filesystem, and network stack, plus dedicated CPU, RAM, and disk, that boots in under 90 milliseconds and can be driven programmatically through SDKs, an API, and a CLI. Fast startup combined with persistent state makes it a useful model for agent workflows that need files and processes to survive across sessions rather than firing off stateless function calls. The architecture docs and multi-language client libraries are worth reading if you are designing your own code-execution layer. One honest caveat: development moved to a private codebase in mid-2026, so this repository is no longer maintained. Treat it as a fork-and-build-on reference, not a dependency you can expect ongoing fixes from.

Articles for this project

No articles for this project yet.

To suggest a topic or contribute an article, contact us.

Related projects in this category