On April 22, 2026 from 18:49 to 19:32 UTC , the Copilot Cloud Agent service began failing during session execution for users running the Agent HQ Codex agent. Codex agent sessions failed to start for all entry points (issue assignment, @copilot comment mentions). 0.5% of total Copilot Cloud Agent jobs were impacted (~2,000 failed jobs). Copilot and other agent sessions were unaffected.
This was caused by a model resolution mismatch in Codex agent sessions, resulting in an incompatible model being used at runtime. A mitigation was deployed to select a stable default model for Codex agent sessions.
We are working to harden the underlying model-resolution path so it correctly scopes to the requesting agent's supported models to prevent similar failure mode in the future.
Posted Apr 27, 2026 - 19:02 UTC
Monitoring
The degradation has been mitigated. We are monitoring to ensure stability.
Posted Apr 27, 2026 - 19:01 UTC
Update
We've found the issue and are working on deploying a solution to get Codex agent runs working again.
Posted Apr 27, 2026 - 17:26 UTC
Update
Copilot Cloud Agent (CCA) jobs using the Codex agent are failing after starting. To avoid this issue, please choose a different agent. We are investigating the cause and working towards remediation
Posted Apr 27, 2026 - 17:01 UTC
Investigating
We are investigating reports of impacted performance for some GitHub services.