🔍 Read the full analysis: GitHub’s AI Agent Surge Moves The Pull Request Bottleneck To Review on ThorstenMeyerAI.com
Get the latest gadgets delivered free with Prime
- Fast, free delivery on millions of items
- Prime Video, Amazon Music and more included
- Member-only deals all year
TL;DR
GitHub’s 2026 Octoverse report says more than 40% of pull requests on the platform in September began with a coding agent, up from 17% in March. The company introduced tools for AI-assisted review and model routing as review capacity, approval rules and costs become more pressing; a merge does not establish that code is correct.
More than 40% of GitHub pull requests in September 2026 started with a coding agent’s first draft, up from 17% in March, according to the platform’s new Octoverse report. GitHub says that amounted to 110 million agent-started pull requests in one month, a shift that puts more pressure on developers to review and approve code rather than produce its initial draft.
The report, released October 28 at GitHub Universe, says 90% of pull requests attributed to agents were merged within two weeks in September. The comparable rate was 83% in February. Pull requests without a detected agent remained at about 86%. These figures describe whether code was merged, not whether it was correct, secure or free of later rework.
GitHub defines an agent-started pull request as one whose first draft came from a coding agent the company can detect. The company describes its count as a floor because it may miss some agent use. The report includes public and private repositories unless otherwise specified, but does not provide a public-versus-private breakdown for this metric. A pull request can still count as agent-started even if a person later substantially edits it.
GitHub’s announcements this week focus on the review stage. Copilot code review Max draws on multiple AI reviewers and combines their findings; GitHub also made Copilot approvals eligible to count toward required approvals and released Copilot Fusion, which routes work to models for different steps. GitHub estimates average review costs of $0.25 for Lite, $1.50 for Balanced and $10 for Max, with GitHub Actions minutes charged separately.
Infographic 01 · The agent share
Four in ten pull requests now start with an agent.
GitHub’s own telemetry, March to September 2026. The first draft comes from an agent; review and merge stay with people.
Merged within two weeks
Share of pull requests merged within 14 days
Merged fast is not the same as correct. GitHub publishes no revert or defect rate.
Bigger changes to review
Share of pull requests that change more than one file
More files per change means more for a reviewer to check.
What GitHub counts: an “agent-started pull request” is one “whose first draft came from a coding agent that we can detect.” Undetected agents are missing, so GitHub calls the figure a floor. Metrics cover public and private repositories unless noted.
Source: GitHub Octoverse 2026, 28 Oct 2026 (text, chart captions, glossary and methodology). All figures are GitHub’s own.
Powered by Thorsten Meyer AI
Review Capacity Becomes the Constraint
When agents produce more proposed changes, the workload does not disappear; it shifts toward checking diffs, validating behavior and deciding what can ship. The reported increase in merged agent pull requests suggests that this work is already moving through GitHub’s pipeline at scale. But merge rates alone cannot show whether teams are saving time overall or taking on more defects and rework.
For organizations, AI review introduces costs and governance choices alongside coding capacity. Teams may need to budget for review credits and Actions usage, decide which repositories can count an AI approval toward a requirement, and track outcomes such as reverts, incidents and follow-up fixes. For individual developers, writing a clear task and critically reading an agent’s changes become more central parts of the job.
The shift may also affect early-career training. Small, bounded coding tasks can provide opportunities to learn codebases and review practices, while those same tasks are among the work agents can take on. The report does not quantify any employment or training effect, so the longer-term impact on junior roles remains an open question.
Products Worth Considering
Bose QuietComfort Wireless Bluetooth Headphones with Active Noise Cancellation, Mic, USB‑C Charging, Deep Bass, 24‑Hour Battery – White Smok
Bose QuietComfort Wireless Noise-Cancelling Over-Ear Headphones, Black
LG UltraGear 27-Inch QHD OLED Gaming Monitor, 240Hz, 0.03ms, G-SYNC Compatible, FreeSync Premium Pro, HDMI 2.1
As an affiliate, we earn on qualifying purchases.
How GitHub Defines Agent Work
The Octoverse is GitHub’s annual report on activity across its platform. This edition covers October 1, 2025, through September 30, 2026. GitHub says developers pushed code 17.9 billion times during that period, while merged pull requests rose from 538 million in 2025 to 1.3 billion in 2026. The report’s headline agent measure should be read using its specific wording: “agent-started”, rather than assuming every part of a pull request was written by an AI.
GitHub’s report uses related labels in different places, including “agent-generated,” “agent-authored” and “agent-opened.” Its methodology also says it recalculated 2024 and 2025 figures under updated definitions and reporting windows, meaning older Octoverse totals are not directly comparable. The company reports that 55% of monthly active Copilot users now use agent features, up from 35% a year earlier, and that use of its cloud agent grew 230% year over year. These are GitHub’s platform metrics, not an independent measurement of developer productivity.
“A pull request whose first draft came from a coding agent that we can detect.”
— GitHub, in the Octoverse report methodology
What Merge Rates Cannot Show
The report does not establish whether agent-written changes are more or less reliable than human-written changes. A merge is a workflow outcome, not a quality measure, and the supplied figures do not report comparative defect rates, security issues, reverts or reviewer hours. GitHub’s detection-based definition also means the actual proportion of agent-assisted pull requests is unknown.
It is also unclear how many teams will use Copilot approvals to satisfy required review rules, how often AI review catches issues missed by people, and how much review time the new tools save after their costs are included. GitHub provides estimated prices for its review tiers, but not a complete estimate of total spending across different usage patterns.
Teams Will Test Review Workflows
GitHub’s new review and routing features give teams more ways to handle agent-generated changes, but the report does not specify how quickly adoption will spread or how the company will measure their results. The next useful evidence will be whether organizations publish data on review time, rework, defects and incidents, alongside raw pull-request and merge counts.
Companies adopting these tools will need to set approval policies and monitor costs as usage grows. GitHub’s next Octoverse reporting may offer another comparison, though changes to definitions and reporting methods will matter when evaluating the trend. For now, the confirmed development is a rapid rise in detected agent-started pull requests—and a parallel effort by GitHub to address the review work they create.
Key Questions
What does GitHub mean by an agent-started pull request?
GitHub counts a pull request when its first draft came from a coding agent the company can detect. A person may edit the draft later and it can still count. GitHub says the figure is a floor because it may not detect all agent use.
How many agent-started pull requests did GitHub report?
GitHub says agents started 110 million pull requests in September 2026, representing more than 40% of pull requests on the platform that month. The share was 17% in March, according to the report.
Does a high merge rate mean the agent code was correct?
No. GitHub reported that 90% of detected agent pull requests were merged within two weeks in September, but a merge does not establish code quality. The report figures provided do not show comparative defect, rework or incident rates.
How much does Copilot code review cost?
GitHub estimates an average review at $0.25 for Lite, $1.50 for Balanced and $10 for Max. GitHub Actions minutes are additional, and actual spending depends on usage.
Can GitHub’s cloud agent approve its own pull request?
According to GitHub, its cloud agent cannot approve or merge its own pull request. The person who requested the change also cannot approve it.
Source: ThorstenMeyerAI.com
