error-recovery

You can automate failure classification and recovery by applying a decision-tree to workflow errors, distinguishing blockers from fixable issues, and orchestrating up to three auto-fix attempts before escalating unrecoverable cases to manual intervention.

The best way to handle sprint or epic implementation failures involves inspecting state indicators and failure reports to determine recoverability, guiding auto-recovery attempts, and escalating unrecoverable blockers to manual intervention.

The decision-tree mechanism classifies automation workflow failures by evaluating failure reports and state indicators to distinguish fixable issues from blockers, determining whether the system can attempt auto-recovery or must escalate to manual intervention.

No specific dependencies are required to use automated issue classification for deployment issues, as the Skill operates independently to inspect failure reports and state indicators to guide auto-recovery or manual intervention.

When auto-recovery attempts fail to resolve CI/CD pipeline errors after up to three retries, the system escalates unrecoverable cases to manual intervention and provides clear reporting to guide the next steps.

Yes, you can use this error-recovery approach for feature development and deployment workflows, as it applies during sprint or epic implementations and CI/CD pipelines to classify failures and orchestrate automatic recovery or manual escalation.