feature-recovery

Root cause diagnosis extracts the underlying issue from error outputs by parsing failure messages, classifying error types, and identifying what went wrong. This Skill analyzes test failures, timeouts, runtime errors, Git conflicts, authentication issues, and network failures to pinpoint the specific problem so you understand exactly what broke.

Automatic recovery depends on recoverability assessment. This Skill determines whether an error can be fixed automatically—such as adding a missing import or correcting a syntax error—or if it requires human intervention, then generates concrete recovery steps or escalation guidance accordingly.

Recoverable errors include missing imports, syntax mistakes, configuration issues, and common runtime problems that can be systematically fixed. Non-recoverable errors—such as infrastructure unavailability or design flaws—are flagged with escalation reasons so you know when to involve human review instead of attempting automation.

By automating failure diagnosis and generating actionable recovery plans, this Skill eliminates manual debugging overhead. Developers get immediate root cause identification and step-by-step recovery instructions, allowing faster retries and reducing time spent analyzing logs and error messages.

The recovery plan outputs root cause, recoverability status, concrete recovery steps, human-readable instructions, suggested actions for automation, and escalation reasons. This structured JSON payload gives both AI agents and developers everything needed to resolve or escalate the failure.