Run an AI software factory that opens, reviews, and merges PRs
A guide to building agents that open, review, and merge pull requests.
Done with a custom-built or unnamed agent
- What they did
- The guide breaks a software factory into five stages: intake, isolation, tools, verification and a merge gate. It draws on published setups from Stripe, Spotify, Shopify and Ramp. It gives configs for each stage, including an issue filter, git worktrees per agent, a shared MCP tool block, a screenshot-verification skill and tiered merge rules. A human reviews at the end.
- How it went
- Microsoft's dotnet/runtime data shows agent PRs merging 67.9% of the time against 87.1% for engineers, and 55.1% when fully autonomous. Review capacity becomes the bottleneck, and proven results are mostly mechanical migrations.
- Worth knowing
- Build the gates before the fleet, and start with one gate where agents cause the most rework. Worktrees start empty, with no .env, dependencies or free port, and non-interactive runs don't clean up after themselves.
Try it yourself with Codex
Set up a workflow for the repository [repo name] where an agent picks up issues labeled [label], opens a pull request with a fix and tests, and a second agent reviews it. Write the configuration and a short guide for how it runs. Do not enable automatic merging yet; show me the merge rules you propose and wait for my approval.
Discussion on HN · Sep 11, 2026
Five of these in your inbox every morning
The best things people got an AI agent to do, each with the prompt to try it.
More like this
- Claude
Build a C compiler with a team of parallel agents
Anthropic's engineering write-up on tasking Opus 4.6 agent teams with building a C compiler.
- Custom agent
Catch cloud outage risks before they happen
An autonomous ReAct agent that queries real data to stop cloud outages before they occur, built for the Sanity Challenge.
Hand narrow code changes to Devin for pull requests
A team used Devin for a month on small fixes like chart formatting and API edge cases, with tests and PRs; a complex full-stack task went poorly.
Migrate toolchain, architecture and site design with Claude Code
A developer had Claude Code switch a monorepo to oxlint/oxfmt, restructure the architecture and redesign a site, validating in a loop; the redesign took days instead of weeks.
Build and deploy a Game of Life simulation site
In a vendor launch demo, Devin built an interactive Game of Life website, added requested features step by step, and deployed it to Netlify.
Resolve a Slack-reported issue with a coding agent
A Codex agent resolved an issue passed to it in about 40 minutes, and an ops agent verified the fix back in Slack.