Scalable, observable, governed AI SDLC.

With Facility you set up your own AI Software Development Lifecycle (AI SDLC) platform, open source and self-hosted. Stories go in, reviewed pull requests come out, and people hold on to the decisions that matter.

We built it because the bottleneck moved. Agents can produce more code than a team can read; the scarce work now is deciding what gets built and checking what came back.

A story can start wherever the work starts: an issue, a pull request, a request from the team. Scheduled work follows the same path. Facility keeps the project's context attached and gives the story a persistent workspace. A person chooses what gets built.

You define the crew in your repository: agents can plan against the real codebase, implement in the story's workspace, run checks, review changes, and respond to CI failures. Engineers steer agents as they work and try the running product in Facility, then review pull requests on GitHub. Their job becomes brokering many streams of work asynchronously. One person's attention covers far more than their typing ever could.

Facility keeps the conversation, files, and review feedback with the story, so a person or another agent can continue the same work.

The guardrails remain structural. Each story runs in an isolated workspace with credentials scoped to the project. Agents can push directly to the repository; required reviews, checks, and branch protection decide what can merge. Treat repository content and external messages as untrusted input.

Every story leaves a record: agent activity, Git changes, pull requests, reviews, and checks. Project views bring together recorded model and workspace costs. You can follow what happened and see which work needs attention.

Your code stays in your repositories. Suspending or archiving a story keeps its files and sessions available; you choose when to delete the workspace. We use Facility to ship our own products at The Agile Monkeys and published it early so people can use it and argue with it in the open. Expect rough edges.