Vibe Maintainer steve-yegge.medium.com
Steve Yegge on maintaining two fast-growing open source projects, Beads and Gas Town, that between them take around 50 contributor PRs a day, roughly 99% of them AI-generated. His answer is not to filter harder but to accept nearly everything and put agents on the repair work. Over five months: 1,000+ contributors, 4,000+ PRs with 2,300+ merged, an 88% eventual merge rate, and a median resolution time of about 15 hours, at 15 to 20 hours a week of his own time.
The argument underneath is about what rejection now costs. Anyone turned away can have an agent fork the project and maintain the fork, so a No is no longer the end of a conversation:
Forking used to be a declaration of war. Now it’s simply a declaration that someone liked your software enough to want to change it, but you said No.
The workflow is a triage pipeline. An agent sorts open PRs into easy wins, fix-merge candidates, and needs-review. Easy wins (bug fixes, docs, dependency bumps) go through automatically every couple of hours. Harder ones get an agent’s analysis and a recommended outcome from a fixed set: merge, fix-merge, cherry-pick, split-merge, redesign, retire, reject. Somewhere between half and two thirds of PRs never need him, and he keeps human judgement for the last 5 to 10% where the call is genuinely subjective.
The idea worth keeping is that maintainer effort has moved from gatekeeping to remediation, which only works because fixing a near-miss PR is now cheap. His own summary of the typical case: it’s Claude telling you, hey, this PR is mostly healthy but it’s missing a kidney, and you say, please add the kidney. DHH argues the principled version of the same position in Let the agents democratize open source; Yegge is the maintainer actually absorbing the volume.