ds-plan-reviewer

To review a data science plan for implementation readiness, validate its task decomposition, spec alignment, and verification steps against the original specification. This process ensures all data profiling, cleaning, and analysis tasks are atomic and well-ordered before coding begins.

The Iron Law of Plan Review is a principle enforcing strict validation of data science plans to prevent costly implementation rework. It ensures plan documents are thoroughly checked for spec alignment, completeness, and proper task granularity before development starts.

Check data science plan alignment by comparing the plan document directly against the specification to verify all requirements are addressed. This validates that intermediate outputs and task decomposition correctly map to the defined data profiling and analysis steps.

Yes, you can audit task decomposition and intermediate outputs by verifying that each task is atomic, well-ordered, and includes clear verification steps. This ensures the plan defines necessary data cleaning, analysis, and visualization steps completely.

Yes, data science plan review works directly with markdown documents by reviewing the PLAN.md file against the SPEC.md file. This validates task granularity and specification alignment to prevent rework before implementation begins.

Manual data science plan reviews often miss critical gaps in task granularity and spec alignment, leading to costly implementation rework. Automated plan auditing catches these completeness issues by strictly validating intermediate outputs and verification steps.