About us

We're building better incident response, in the open

DeAgent is an open-source DevOps agent for SRE teams who want faster recovery, less alert fatigue, and full control over what runs in production.

On-call is hard enough without the same incidents showing up every week: a pod in CrashLoopBackOff, a red CI pipeline, a disk about to fill. Most teams either handle everything manually or bolt on automation they can't fully inspect.

We started DeAgent to close that gap: an agent that watches your stack, diagnoses with real collector output, applies fixes through safety-checked executors, and logs every decision to org-scoped storage.

It runs on your infrastructure. You bring your own API keys. And because it's open source, anyone can verify how it behaves before trusting it with production.

open-devops-agent / open-devops-agent

Open-source codebase for DeAgent

mainMIT2

Branch strategy

  • main: Default branch. Stable, production-ready code
  • feat/*: Feature branches for new capabilities
  • fix/*: Fixes, hardening, and security patches

What we believe

Safety first

Never delete production data. Dry-run by default, approval gates for risky actions, and a complete audit trail.

Transparency

MIT licensed and fully auditable. Your security team can read every collector, tool, and safety check.

Built with the community

Contributors shape collectors, integrations, and docs. We merge what helps teams run production reliably.

Respect for on-call

The goal is less toil, not replacing engineers. Automate the repetitive work so people can focus on hard problems.

People building this

DeAgent is maintained by contributors who care about reliable infrastructure. Here are the people who have shipped code to the project so far.

How to contribute

You don't need to be a core maintainer to help. Documentation fixes, new collectors, test coverage, and bug reports all move the project forward.

1

Explore the repo

Read the README, architecture docs, and security guarantees.

View repository
2

Find something to work on

Look for good-first-issue or help-wanted labels, or open a discussion about what your team needs.

Browse issues
3

Fork, branch, and PR

Use feat/* or fix/* branches. Follow CONTRIBUTING.md for setup, tests, and the pull request process.

Read contributing guide
  • Add or improve platform collectors (AWS, K8s, CI/CD, and more)
  • Extend MCP tools for Cursor, Claude Desktop, and custom agents
  • Improve docs, examples, and the website you're reading now
  • Report production edge cases. That's how safety gets better

Want to talk, not just code?

Request a demo if you're evaluating DeAgent for your team, or head to GitHub if you're ready to contribute.