Skip to content

Requra/frontend

Repository files navigation

Requra.AI robot logo

Requra.AI

From unstructured project context to traceable, review-ready requirements.

Requra.AI turns documents, meetings, and notes into requirements, user stories, acceptance criteria, evidence, stakeholder feedback, and delivery exports.

Open interactive demo  ·  Workflow  ·  Architecture  ·  Run locally

Documents, meetings, and notes transformed by Requra.AI into requirements, user stories, and evidence

Main screens

Requra.AI recruiter dashboard with project portfolio and workspace pulse Requra.AI projects workspace with searchable project cards

Requra.AI AI results workspace with requirements and quality metrics Requra.AI evidence view linking source documents to requirement coverage

Product workflow

  1. Create a project and add source documents or meeting context.
  2. Start asynchronous analysis and follow explicit progress states.
  3. Review summaries, requirements, stories, criteria, risks, and evidence.
  4. Invite stakeholders to comment through focused public review links.
  5. Resolve feedback and export delivery-ready artifacts.

Core experience

Area What is implemented
Projects Create, edit, search, filter, sort, inspect, and delete project workspaces.
Analysis Queued, processing, completed, partial, rejected, failed, and retry states.
Results Summaries, requirements, user stories, acceptance criteria, quality signals, and source evidence.
Collaboration Viewer/commenter invitations, public review, feedback submission, and team inboxes.
Meetings Scheduling, invitations, join flow, live controls, consent, recording orchestration, and summaries.
Export Browser-generated CSV and Jira-ready JSON data.

Architecture

Requra.AI frontend architecture from React Router and feature modules through state boundaries, typed APIs, backend, and AI pipeline

The normal production path is React frontend → .NET backend → external AI pipeline → normalized results.

  • TanStack Query owns server state, mutations, cache invalidation, and polling.
  • Zustand owns focused session and client state.
  • React Router keeps project and results views linkable.
  • React Hook Form + Zod keep forms local, typed, and validated.
  • Axios adapters isolate transport contracts from feature UI.
  • MSW fixtures power the explicit, backend-free demo variant without silently replacing failed production requests.

Stack

Responsibility Technology
Application React 19, TypeScript 5.9, Vite 7, React Router 7
Data TanStack Query 5, Zustand 5, Axios
UI Tailwind CSS 4, Radix UI, CVA, Lucide, Sonner
Forms React Hook Form, Zod
Motion GSAP, Lenis, Framer Motion
Testing Vitest, Testing Library, MSW, JSDOM

Run locally

Requirements: a current Node.js LTS release and npm.

git clone <repository-url>
cd frontend
npm ci
cp .env.example .env
npm run dev

PowerShell:

Copy-Item .env.example .env
npm run dev

Vite serves http://localhost:5173. Environment options and safety gates are documented inline in .env.example.

Command Purpose
npm run dev Start the development server.
npm run build Type-check and create the production bundle.
npm run test:run Run the test suite once.
npm run lint Run repository-wide ESLint.
npm run preview Preview the production bundle.

Demo and quality

The /demo entry opens a deterministic, resettable workspace with synthetic data. Its dedicated runtime uses the normal UI and API contracts with same-origin MSW handlers, browser-local versioned fixtures, and production-service isolation.

  • Production build: passing.
  • Test suite: 15 files, 97 tests passing.
  • Responsive verification: 320×568 through 1920×1080.
  • Accessibility checks include keyboard activation, visible focus, exact accessible metric values, touch-target sizing, text scaling, and reduced motion.

Current scope

The backend and AI pipeline are external services. Demo meeting media and transcription are simulated, Jira-ready export does not create remote issues, and direct browser-to-AI processing remains a guarded opt-in fallback rather than the production default.


Capture context. Expose evidence. Align stakeholders. Deliver clearly.

About

✨ Modern React frontend for Requra.AI — an AI-powered requirements workspace that turns meetings and documents into review-ready requirements, user stories, and backlog outputs.

Topics

Resources

Stars

Watchers

Forks

Releases

Packages

Contributors

Languages