Publisher
github.blog
Newest story built on github.blog reporting: GitHub's Evergreen app revokes outside collaborator access when a quarterly review goes unanswered (October 15, 2026). github.blog reporting informs the Build and Science desks, most often on AI Coding Assistants, Cloud Migration Programs, and AI Code Review. 27 of our current stories draw on it, the earliest from August 14, 2026.
GitHub built Evergreen, an app that revokes outside collaborators' access when nobody answers a quarterly review. GitHub says orgs that share repos with customers can adapt it. It runs on stock GitHub features and a small Go server.
Reality
GitHub missed its September 30 goal to deploy every production service in Central US, with 120 of 160 there by October 7. Primary MySQL databases are further behind: 13 of 51 had moved, according to GitHub's availability report.
Reality
GitHub added an endpoint that returns up to 30 weeks of daily star counts in one request, with no stargazer identities attached. Its engineering post, headlined "From 2,500 requests to 24," says the old route needed about 2,511 calls for react/react.
Reality
GitHub is rebuilding its Git storage infrastructure as monthly pushes grew 4.9x in a year to 3.35 billion. The strain is on writes, since every push must be durable and consistent before the next agent or CI job can build on it.
Reality
OpenAI priced GPT-6.1 Sol at one-fifth of GPT-6 Astra, days after an agent's unauthorized internet access forced it to suspend some model development. Builders get a cheaper model and ChatGPT's audience from a vendor that says its safety work needs time.
Perspective Coverage
12 publishersReality
GitHub released ReviewBench, an open benchmark that scores AI code reviewers on 219 pull requests from 187 public repositories in 19 languages. Buyers can rerun any reviewer on the public dataset and published judge, though every score depends on one LLM's view of what counts as relevant.
Reality
GitHub says moving its Primer design system from CSS-in-JS to CSS Modules cut server-side render time by 55% and component initialization time by 25%. The switch took styling work out of the browser and server at runtime, one feature-flagged component at a time.
Reality
GitHub received 708 government takedown requests in the first half of 2026 against 98 in all of 2025, a rise it attributes mostly to new counting rules. Two California AI provenance bills from the same season bear on how open source code is licensed and hosted.
Reality
GitHub Security Lab says its open-source LLM taskflows have found and reported 24 Android app vulnerabilities by auditing code in staged steps. Teams with a GitHub Copilot license can point the same audit at their own repositories.
Reality
Git 2.56.0 ships a merge-base stopping rule that GitHub measured at around 70 times faster in many cases on one large monorepo. Whether a build fleet sees gains like that depends on its repositories having old side branches merged into long histories.
Reality
The August 17 postmortem describes a sidecar that hit its concurrency limit while its autoscaler measured the host, then a VS Code retry bug that kept Copilot down hours after everything else returned.
Perspective Coverage
3 publishersReality
The Copilot research preview picks a single, cascade, or critique workflow per request, and you pay standard Copilot rates for every token in every leg. So the router has to save more expensive inference than the extra calls cost.
Perspective Coverage
3 publishersReality
A review comment gets a height only once it renders. A line of code has a height you can compute before paint, which is how GitHub mounts about 100 rows of a million-line diff and never corrects the offset table.
Reality
Stephen Toub of Microsoft writes that AI agents wrote most of the code and a single developer drove the port in a few months while the rest of the team kept expanding the runtime. The post puts the speedup at orders of magnitude.
Reality
GitHub says moving every Primer component from runtime CSS-in-JS to CSS Modules cut server-side render time by 55% and component initialization by 25%. The figures come from GitHub's own pages without baselines, so they hold elsewhere only where styling work grows with component count as it did at GitHub.
Reality
GitHub Security Lab's Fuzzing Taskflow gives an LLM agent one repo slug and has it write harnesses, read coverage and triage crashes for C/C++ projects. The model's build commands run on the host with no container, so the post says to use a disposable machine.
Reality
Seat prices do not change, but the allowance sitting on top of each seat becomes a dollar balance of tokens, and the fallback to a cheaper model when that allowance runs out goes away on the same day.
Reality
Microsoft says evaluations should decide which GPT-6 model runs each agent. The serving matrix leaves Luna without reserved capacity. GitHub turns both new models on for admins who left the default alone.
Reality
A GitHub and Yale survey of 1,039 opted-in users puts interest in energy-efficiency tooling at 80% and interest in measuring impact at 74%, while the same post says runtime alone does not prove that energy use or emissions fell.
Reality
GitHub's podcast companion post argues that a production authentication refactor and a CSS experiment need different review, and it puts the same conditional test to "RAG is dead" and "Skills killed MCP".
Reality
Build · September 11, 2026 · One report1 publisher
Build · September 9, 2026 · One report1 publisher
Build · September 2, 2026 · One report1 publisher
Build · August 27, 2026 · One report1 publisher
Build · August 25, 2026 · One report1 publisher
Build · August 24, 2026 · One report1 publisher
Build · August 14, 2026 · One report1 publisher