GitHub is experiencing a widespread outage affecting numerous services, including its website, API, Actions, and Pull Requests, which began on August 17, 2026. The company confirmed the incident at 9:40 AM EDT, stating it was investigating performance issues.
The outage has led to significant error rates across the platform. GitHub's status page indicates approximately 20% error rates for its general web experience and API traffic. More severely, archive downloads and raw repository content downloads are experiencing error rates around 50%.
Affected services include API Requests, Actions, Webhooks, Issues, and Pull Requests. Authentication services such as SAML, OIDC, SCIM, and Team Sync are also impacted. Users have reported server errors when attempting to access the site, and difficulties loading commits, repositories, and Pull Request pages.
The degraded performance of GitHub Actions means that automated builds, tests, deployments, and other continuous integration/continuous delivery (CI/CD) workflows are also being affected. At 10:31 AM EDT, GitHub further confirmed that its AI coding service, Copilot, was experiencing degraded availability.
While Git Operations, Packages, Pages, and Codespaces are currently listed as operational, many critical components of the platform remain degraded. GitHub has not yet disclosed the root cause of the outage, stating that its investigation is ongoing.
As of 11:42 AM EDT, GitHub reported it was performing mitigations, though error rates for web experiences, API traffic, and repository content downloads remained at their previously observed levels of 20% and 50% respectively.
