executor-workflow

Multi-layer development plans can be executed incrementally by orchestrating wave-based execution. This approach validates baseline tests, manages branch patterns, and applies annotated checkpoint tags to preserve history and enable targeted rollback safely.

Wave-based plan execution is a structured method for delivering complex features by breaking work into sequential units. Each wave includes validation, TDD, review, merge, and checkpoint tagging to ensure safe, incremental delivery across logical layers.

Git branch checkpoints are managed by enforcing per-wave branch patterns and annotated checkpoint tags. This preserves commit history and enables targeted rollback if a wave fails or requires forensic inspection.

Wave-based execution is needed when a plan spans three or more logical layers, such as data, API, and UI. It is required when partial delivery, validation, and rollback granularity are necessary for safe feature implementation.

Blockers during automated plan execution are handled by pausing the workflow exclusively for user-declared blockers. The executor logs autonomous decisions and surfaces these blockers for user input, keeping failed waves available for review.

TDD checkpoints can be integrated into multi-layer feature delivery by enforcing per-wave baseline tests. Each wave unit applies test-driven development validation before proceeding to review, merge, and checkpoint tagging.