What problem does it solve? Turning an approved requirements and design specification into an executable, well-structured implementation task list is slow and error-prone, often leaving requirements uncovered or tasks with hidden dependencies. ## Core Features & Use Cases - Requirements-to-Tasks Mapping: Reads spec.json, requirements.md, and design.md to generate a tasks.md where every requirement ID is covered by at least one task. - Parallel Execution Analysis: Marks tasks with (P) when they have non-overlapping component boundaries and no data dependencies, enabling concurrent implementation. - Built-in Review Gates: Runs a task plan review and an independent task-graph sanity check before writing tasks.md, catching coverage gaps, sizing issues, and dependency mistakes. - Use Case: After approving requirements and design for a new feature, run the skill to produce a localized tasks.md with 1-3 hour sub-tasks annotated with Requirements:, Boundary:, and Depends: fields, then hand off to implementation. ## Quick Start Ask the AI to generate implementation tasks for your approved spec by running the kiro-spec-tasks skill with your feature name, optionally adding -y to auto-approve requirements and design.
