GitHub: agents started more than 40% of pull requests in September

Monday, 2 November 2026By Sridhar Mukkandi1 min read

GitHub published its Octoverse 2026 report on 28 October. By September 2026, more than 40% of pull requests on GitHub started with a first draft from a coding agent, up from 17% in March: 110 million in September alone, about six times March's volume. GitHub counts only the agents it can detect, so it calls the figure a floor.

Agent pull requests are also merging more often. The share merged within two weeks rose from 83% in February to 90% in September, while pull requests without detected agent involvement held at about 86%. Over the 12 months to 30 September 2026, GitHub counted 17.9 billion pushes, 34.3 billion commits and 1.3 billion merged pull requests, and said use of AI-assisted code review grew more than 90% year over year.

So what

A merge rate measures what reviewers accepted, not what held up in production. If agents now write the first draft, track reverts and follow-up fixes on their changes, and spend review time where a mistake costs most.

SourceGitHub Bloggithub.blog/news-insights/octoverse/octoverse-commits-pushes-rea Go deeper · EssayYour review queue is a budget

Drafted with AI from the original source, then checked and edited by Sridhar Mukkandi. Spotted a mistake? Write to hello@tarkika.com and we'll correct it here.

More from The Signal

Models Google announces Gemini 4 Carbon, starting with cyber defenders onlyA model that is better at finding vulnerabilities is better at exploiting them too, which is why access is staged. The figures are Google's own: test any new model on your own tasks before trusting a leaderboard. Google Agents IBM makes Agent Identity generally available: an owner and short-lived access for every agentYou don't need a product to start: a list of every agent, its owner, the tools it may call and when its access ends is most of the value. Any agent without an owner should be read-only. IBM Research Microsoft’s ThinkingBox turns database checks into rewards for training agentsChecking the state an agent leaves behind is a more honest test than reading its transcript, and the same check doubles as a training reward. Start with the three actions your agent takes most often. Microsoft and Hugging FaceAll stories