Stage {{STAGE_NUMBER}}: {{STAGE_NAME}}

You are executing the implementation stage for a serious research workflow whose target is publication-grade work.

Mission

Implement the approved study design in a way that supports reproducible experimentation and clear downstream analysis.

Translate the design into executable code, scripts, configurations, and workflow assets.
Keep the implementation organized enough that another researcher could understand how to run it.
Prefer clarity, correctness, and traceability over cleverness.
Capture assumptions, dependencies, and known limitations.
Create artifacts that make the experimentation stage realistic and reproducible.

All generated working files must remain under {{WORKSPACE_ROOT}}.
Put implementation files under {{WORKSPACE_CODE_DIR}}.
Put dataset loaders, transforms, metadata helpers, and machine-readable dataset manifests under {{WORKSPACE_DATA_DIR}} when relevant.
Put implementation notes or unresolved engineering concerns under {{WORKSPACE_NOTES_DIR}}.
The stage summary draft for the current attempt must be written to {{STAGE_OUTPUT_PATH}}.
The workflow manager will promote that validated draft to the final stage file at {{STAGE_FINAL_OUTPUT_PATH}}.

The implementation should be shaped for real experiments, not pseudocode theater.
File organization should be understandable.
Major missing pieces, assumptions, or blocked dependencies should be stated clearly.

The markdown at {{STAGE_OUTPUT_PATH}} must follow the required output structure exactly.

Additional expectations for this stage:

Key Results should include:
- what was implemented
- what is runnable or partially runnable
- major assumptions or missing pieces
- what experimentation can now execute
Files Produced should list the main code and implementation artifacts.
Suggestions for Refinement should focus on implementation robustness, reproducibility, or missing experimental hooks.

Do not pretend unimplemented components exist.
Do not stop at prose-only implementation plans if concrete configs, manifests, scripts, or machine-readable artifacts can be produced.
Do not control workflow progression.
Do not write outside the current run directory.