GitHub Actions Retention Changes on October 1, 2026: What to Audit Now
Prepare for GitHub Actions retention changes affecting checks, workflow runs, statuses, artifacts, and logs with a practical audit and archive plan.
Starting October 1, 2026, the retention setting in GitHub Actions will control more than artifacts and logs. Checks, workflow runs, and commit statuses will also be removed after the configured retention period.
That makes the setting an operational and compliance decision, not just a storage-cost control.
What Is Changing
GitHub currently keeps checks, workflow runs, and commit statuses for more than 400 days even when a shorter Actions retention period is configured. From October 1, those records follow the same policy as artifacts and logs.
The default is 90 days. Organization and enterprise caps still limit how high a repository can set its retention. Public repositories cannot retain these records beyond 90 days.
The policy also covers checks and statuses created by third-party integrations, not only records created by GitHub Actions.
Why This Can Surprise DevOps Teams
A green check on an old commit may be part of release evidence. A workflow run can explain which commit, environment, approver, and artifact created a production release. If your audit process assumes that evidence remains visible indefinitely, the new cleanup behavior creates a gap.
Changing the setting later will not restore records that GitHub has already deleted. Treat September as a migration window.
Audit the Current Retention Setting
Review the setting at enterprise, organization, and repository levels. In an organization, open Settings → Actions → General and locate Check, workflow run, status, artifact and log retention.
Build an inventory with these columns:
| Repository | Current days | Required days | Reason | External archive |
|---|---|---|---|---|
| application | 90 | 90 | normal CI | no |
| production-infra | 30 | 365 | change evidence | yes |
| regulated-service | 90 | 730 | audit requirement | yes |
Do not blindly maximize every repository. Longer artifact and log retention can increase billable storage. Checks, workflow-run metadata, and statuses are not billed as artifact storage, but the logs and artifacts attached to those runs are.
Separate CI Convenience from Audit Evidence
GitHub Actions is an execution system, not necessarily your permanent evidence store. If evidence must outlive the allowed retention period, export the minimum required record to an approved archive.
A release evidence bundle might contain:
- commit SHA and repository
- workflow name and run ID
- actor and approval record
- deployment environment
- artifact checksum rather than a second artifact copy
- infrastructure plan or security scan summary
- timestamps and final conclusion
Protect the archive with retention locks, access controls, encryption, and an owner. Copying every log forever creates cost and privacy problems without improving audit quality.
Create a Safe Migration Plan
- Identify repositories that produce production releases.
- Map regulatory and customer requirements to an explicit number of days.
- Compare the requirement with enterprise and public-repository caps.
- Export evidence that must remain longer than GitHub allows.
- Test retrieval from the archive before relying on it.
- Record the owner and review date for each exception.
- Recheck settings after October 1 and confirm old records expire as expected.
Common Mistakes
Do not assume raising retention restores deleted data. Do not use artifact retention alone as a compliance program. Do not archive secrets printed in logs. Finally, do not forget third-party commit statuses: the new policy applies to them too.
Bottom Line
The October 1 change makes your existing Actions retention value more powerful. Review it before cleanup begins, retain ordinary CI data only as long as it remains useful, and move long-lived release evidence into a system designed for records management.
For a broader runner strategy, compare GitHub-hosted, larger, and self-hosted runners.
Sources
Today I Fixed
Short real fixes from production — posted daily
Stay ahead of the curve
Get the latest DevOps, Kubernetes, AWS, and AI/ML guides delivered straight to your inbox. No spam — just practical engineering content.
Related Articles
5 DevOps Portfolio Projects That Actually Get You Hired in 2026
Not just another list of project ideas. These are the specific projects that hiring managers at top companies are looking for — with exactly what to build and how to present them.
AI Agents Will Replace DevOps Bash Scripts — And That's a Good Thing
The future of DevOps automation is not more bash scripts. AI agents that can reason, adapt, and self-correct are quietly making traditional scripting obsolete. Here is what that means for DevOps engineers in 2026 and beyond.
AWS CodePipeline vs GitHub Actions — Which CI/CD Tool to Use? (2026)
AWS CodePipeline and GitHub Actions both automate deployments. But they have very different strengths. Here's an honest comparison with real examples.