execution-recovery

Execution recovery diagnoses and fixes common failures like syntax errors, import issues, path problems, and permission denials across Python, JavaScript, and shell contexts. It applies structured triage to classify the error type, maps it to a targeted recovery strategy, and outputs actionable commands with validation steps to resolve the failure and retry the task.

The system handles syntax errors, import and dependency issues, file path and file access problems, permission denials, network timeouts, and session state errors. It categorizes each failure into its root cause and applies specialized recovery techniques including session recovery for long-running tasks and rollback planning.

Yes. Execution recovery works across local development environments, CI/CD pipelines, and remote runtimes. It diagnoses errors in all three contexts, applies environment-specific recovery strategies, and provides verification steps to confirm the fix before task retry.

Structured triage classifies errors immediately into syntax, imports, paths, or network buckets rather than manual log inspection. This systematic categorization maps each error type to proven recovery commands, skipping guesswork and accelerating diagnosis and fix application.

After applying a targeted fix, execution recovery includes validation steps to confirm the error is resolved. These steps verify the fix works before the original task retries, preventing repeated failures from the same root cause.