A Cookiecutter template for creating standardised NHS England Reproducible Analytical Pipeline (RAP) projects.
This template provides a standardised project structure for developing Reproducible Analytical Pipelines within NHS England. It includes pre-configured tooling for testing, linting, documentation, and dependency management following modern Python development practices.
Reproducible Analytical Pipelines (RAP) is a set of tools, principles, and techniques to help you improve your analytical processes. With RAP, you leverage open-source tools to make your work more efficient, more reusable, and less error-prone.
The core RAP principles from the NHS RAP Community of Practice are:
| Principle | Description |
|---|---|
| Automation | Minimise manual, error-prone steps through automation |
| Modular, reusable code | Write code in independent, loosely-coupled functions |
| Transparency | Publish code openly under appropriate licences |
| Open-source tools | Use Python, R, and other freely available languages |
| Version control | Use Git to track changes to code |
| Good coding practices | Follow standards like PEP8, use clear documentation |
| Testing | Implement automated tests for reliability |
| Peer review | Review code collaboratively to ensure quality |
Learn more about RAP at the NHS RAP Community of Practice website.
This template requires Python 3.10 to 3.13. This package is not published to PyPI, so install directly from GitHub:
pipx (recommended for CLI tools):
pipx install git+https://github.com/nhsengland/nhse-rap-cookiecutter.gitSee the pipx documentation for installation instructions.
uv:
uv tool install git+https://github.com/nhsengland/nhse-rap-cookiecutter.gitSee the uv documentation for installation instructions.
pip:
pip install git+https://github.com/nhsengland/nhse-rap-cookiecutter.gitGenerate a new project using the CLI tool:
nhs-rap-templateOr use the official cookiecutter command:
cookiecutter gh:nhsengland/nhse-rap-cookiecutterThe template prompts for the following information:
| Category | Option | Description | Choices |
|---|---|---|---|
| Project | project_name | Human-readable project name | Text |
| repo_name | Repository name (default: lowercase project_name with underscores) | Text | |
| module_name | Python module name (default: repo_name with dashes converted to underscores) | Text | |
| description | Brief project description | Text | |
| author_name | Your full name | Text | |
| author_email | Your email address | Text | |
| organization_name | Your organisation name (default: NHS England) | Text | |
| team_name | Your team name | Text | |
| team_email | Team contact email | Text | |
| git_hosting_platform | Git hosting platform | github, gitlab, azure_devops, other | |
| repository_url | Repository URL (can override default) | Text | |
| Python | python_version_number | Minimum Python version | 3.10, 3.11, 3.12, 3.13 |
| environment_manager | Virtual environment tool | uv, venv, conda | |
| Options | include_code_scaffold | Include example code modules | Yes, No |
| layout | Package layout (flat = package at root, src = package under src/) |
flat, src | |
| linting_and_formatting | Code quality tools | ruff, flake8+black+isort | |
| open_source_license | Project licence | MIT, Apache-2.0, GPL-3.0, No license file | |
| docs | Documentation tool | mkdocs, none |
Note: All generated projects include core Python packages (pandas, numpy, matplotlib, seaborn, jupyter, etc.) and development tools (pytest, pre-commit, linting) by default. The dependency file format (pyproject.toml or environment.yml) is determined automatically based on your environment manager choice.
your-project/
├── .github/
│ └── workflows/ # GitHub Actions CI (tests, lint, gitleaks, docs)
├── data/
│ ├── external/ # Data from third-party sources
│ ├── interim/ # Intermediate transformed data
│ ├── processed/ # Final canonical datasets
│ └── raw/ # Original immutable data
├── docs/ # MkDocs documentation with NHS styling
├── models/ # Trained models and predictions
├── notebooks/ # Jupyter notebooks for exploration
├── references/ # Data dictionaries and documentation
├── reports/
│ └── figures/ # Generated graphics
├── scripts/ # Setup and utility scripts
│ └── setup_repository.py # Automated repository setup
├── tests/
│ ├── unittests/ # Unit tests (pytest)
│ └── e2e/ # End-to-end integration tests
├── your_module/ # Source code package (flat layout; or src/your_module/ with the src layout)
│ ├── __init__.py
│ ├── config.py # Configuration management
│ ├── dataset.py # Data loading and processing
│ ├── features.py # Feature engineering
│ ├── plots.py # Visualisation functions
│ └── modeling/
│ ├── train.py # Model training
│ └── predict.py # Model inference
├── .env # Environment variables (not tracked in git)
├── .pre-commit-config.yaml # Pre-commit hooks configuration
├── badges.toml # Optional project badges
├── CODE_OF_CONDUCT.md # Community guidelines
├── LICENSE # Project license
├── LICENSE-OGL # Open Government License for docs
├── Makefile # Convenience commands
├── mkdocs.yml # Documentation configuration
├── OPEN_CODE_CHECKLIST.md # NHS England standards for publishing code
├── README.md
├── pyproject.toml # Project configuration and dependencies
└── setup.cfg # Legacy tool configuration (flake8 only)
All generated projects include an Open Code Checklist (OPEN_CODE_CHECKLIST.md) to ensure compliance with NHS England standards before publishing code. The checklist covers:
- Licensing and documentation requirements
- Security checks for sensitive data, credentials, and git history
- Third-party tool compliance with NCSC guidelines
- Code quality and testing standards
- RAP assessment criteria
The checklist is also integrated into the project documentation for easy reference during development.
For projects hosted on GitHub, the template generates a small, KISS set of workflows in .github/workflows/:
tests.yml– runs thepytestsuite on every push and pull request tomainlint.yml– checks code style with your chosen linter (ruff, or flake8 + black + isort)gitleaks.yml– scans the repository for committed secretsdocs.yml– builds and deploys the MkDocs site to GitHub Pages (only when docs are enabled)
The workflows adapt to your chosen environment manager (uv, venv, or conda), and matching status badges are added to the project README. For non-GitHub hosting (GitLab, Azure DevOps, etc.) the .github/ directory is omitted.
The template supports two package layouts via the layout prompt:
flat(default) – the importable package sits at the project root (your_module/). This matches the Cookiecutter Data Science style.src– the importable package sits under a top-levelsrc/directory (src/your_module/), aligned with the NHS RAP community package template. The src layout helps avoid accidentally importing the package from the working directory instead of the installed version.
The import name (import your_module) is identical for both layouts, so usage, tests, and documentation are unaffected by the choice.
After generating a project, use the automated setup script:
cd your-project-name
make setupThe setup script will:
- Initialize git repository with default branch
- Configure git remote with your repository URL
- Set up your Python environment (uv, venv, or conda)
- Install all project dependencies
- Install pre-commit hooks
- Create an initial commit
Alternatively, you can set up manually:
cd your-project-name
uv sync # Set up environment
uv run pre-commit install # Install pre-commit hooks
make test # Run tests
make docs # Build documentationFull documentation: https://nhsengland.github.io/nhse-rap-cookiecutter
See CONTRIBUTING.md for development setup and contribution guidelines.
uv run pytest tests/ -v # Run tests
uv run pytest tests/ --cov=nhse_rap_cookiecutter --cov-report=term # With coverage
uv run ruff format . && uv run ruff check . # Format and lint
make docs-serve # Serve docs with live reloadUnless stated otherwise, the codebase is released under the MIT Licence. This covers both the codebase and any sample code in the documentation.
HTML and Markdown documentation is © Crown copyright and available under the terms of the Open Government 3.0 licence.
This template is based on Cookiecutter Data Science, adapted for NHS England RAP standards and modern Python tooling.