Requireganizer is an AI-assisted requirements and test-design workspace. It turns product intent into a traceable chain of product overview, user stories, requirements, acceptance criteria, test scenarios, test cases, executable tests, and a deterministic project scaffold.
The AI harness is contract-first: each operation exposes one scoped OpenAI-compatible function, validates its arguments and engineering relationships, and applies successful proposals atomically.
Read the project documentation for the complete workflow, architecture, harness contracts, prompt design, state invalidation, testing, environment, and deployment model.
The published site is available at sassanh.github.io/requireganizer.
Requirements:
- Node.js 22.13 or newer
- pnpm 11
pnpm install
cp .env.example .env.local
pnpm run devOpen http://localhost:3000.
The default configuration uses an OpenCode Zen model. Configure credentials or another OpenAI-compatible function-calling endpoint and model through .env.local; see the environment reference.
pnpm run checkThis runs linting, strict TypeScript checks, unit and harness contract tests, the production application build, and the production VitePress build. GitHub Actions runs the same gate for pushes to main and pull requests.
To work on the docs locally:
pnpm run docs:devRequireganizer remains under active development. Generated artifacts require human review, and the final application-code stage is an architectural destination rather than a complete autonomous implementation pipeline.
Read CONTRIBUTING.md before opening a pull request.
Requireganizer is released under the MIT License.