GitHub Copilot Claude and Codex Agent Sessions Dashboard Outage

What happened

On April 10, GitHub Copilot experienced a service disruption where third-party Claude and Codex agent sessions failed to appear in the agents tab dashboard. The incident lasted approximately 20 minutes, from 13:08 to 13:28 UTC, affecting teams using AI-powered code assistance and testing workflows. GitHub resolved the issue but has not yet released a detailed root cause analysis. This follows a separate April 1 incident where Copilot users experienced elevated 5xx errors and latency issues with

Business impact

Teams relying on AI-assisted testing and code review workflows faced visibility gaps that could delay release cycles and code quality checks. Enterprise development teams using Copilot for automated test generation or debugging lost access to session history and agent interactions, potentially disru

Background

GitHub Copilot has become integral to many enterprise development workflows, with teams increasingly using AI agents like Claude and Codex for test case generation, code review, and debugging. The recent pattern of Copilot service disruptions highlights infrastructure risks as organizations integrate AI tools into critical testing and development pipelines. Enterprise teams have reported using Cla

What this means for your team

Implement fallback testing processes that do not depend on AI agent availability. Document manual alternatives for critical test case generation and code review workflows currently handled through Copilot agents. Monitor GitHub's status page proactively and establish alerts for Copilot service disru

What to watch

GitHub has promised a detailed root cause analysis for the dashboard incident. Monitor for patterns in Copilot reliability, particularly around agent session management, as repeated outages suggest underlying infrastructure issues that could affect enterprise adoption.