requirements-as-code 0.7.13

RAC — lint and diff product requirements written in Markdown.

Requirements as Code

Treat product knowledge like source code.

RAC is a command-line tool for managing requirements, decisions, roadmaps, prompts, and design artifacts as plain Markdown, right inside your Git repository.

pip install requirements-as-code

What is RAC?

The code is structured, the tests are automated, the infrastructure is versioned — but the reasoning behind what you build is scattered across documents, tickets, chats, and AI conversations. RAC brings that knowledge back into the repository.

You write product thinking in Markdown; RAC validates it, inspects it, and connects it — so it stays durable, reviewable, and usable as context for both humans and AI. No proprietary formats, no hosted platform, no lock-in.

Who is it for?

• Software and product teams who want the why behind their software versioned alongside the code.
• AI-native teams who need structured, durable context instead of more scattered chat history.

Install

Requires Python 3.11+.

pip install requirements-as-code
# or
uv tool install requirements-as-code

Quick Start

rac validate requirement.md   # check one artifact
rac validate rac/             # check every artifact in a directory
rac inspect requirement.md    # see its type and completeness
rac stats rac/                # summarize a directory of artifacts
rac review rac/               # full repository review, worst problems first

New to RAC? Walk through your first artifact in five minutes: docs/quickstart.md.

Supported Artifact Types

• Requirements — what needs to exist
• Decisions — why choices were made (ADRs)
• Roadmaps — where the product is heading
• Prompts — reusable AI collaboration patterns
• Designs — product experience thinking

Everything stays plain Markdown — see docs/artifacts.md.

Documentation

• Quickstart — install and try RAC in five minutes
• CLI reference — every command, flag, and exit code
• Artifact types — the five types and their sections
• Relationships — link artifacts and validate the links
• Repository workflow — organize a repo with RAC
• Testing & contributing — local setup and verification
• Examples — small, realistic artifacts

How RAC earns trust

RAC asks you to trust it with your product knowledge, so it holds itself to the same standard it applies to your repository:

• It dogfoods itself. RAC's own planning corpus under rac/ is validated by RAC in CI (rac validate rac/, rac relationships rac/ --validate, rac review rac/) — if the tool's rules break the tool's own artifacts, the build fails.
• Output is a contract. Golden tests pin the CLI's human and JSON output byte-for-byte; any change to what RAC prints is reviewed as a product change.
• JSON is versioned. Machine-readable output carries a schema_version and only changes additively.

Project Status

RAC is early and evolving quickly. A terminal Explorer for browsing your knowledge base is planned. Contributions, ideas, and experiments are welcome — see CONTRIBUTING.md.

License

MIT