Fair& is a small prototype project for detecting and rebalancing "invisible labor" requests. This repository contains:

• client/ — React UI (Equity console) used to visualize task distribution and team policies.
• server/ — Local API used by the client for demo data.
• src/ — Slack bot / tooling code (minimal, no-database Slack bot MVP).

This README covers local development and how to run the pieces.

• Node.js 18+ and npm
• (Optional) A Slack workspace and app if you intend to run the Slack bot

Install dependencies for each part of the project:

# root (Slack bot / shared libs)
npm install

# client (React app)
cd client && npm install

# server (local API)
cd ../server && npm install

Open three terminals or use a process manager.

• Start the demo API server (default port 3001):

cd server
npm run server

• Start the React client (default port 3000):

cd client
npm start

• Start the Slack bot (if configured):

# from repository root
npm run bot:always

• Copy any example env files (for the bot or server) to .env and fill values. The client reads REACT_APP_API_BASE_URL to override the default http://localhost:3001 API base.

• The client exposes a time-window selector (7/14/30/90/All time). All time now shows the full task history. Other windows (7/30/90) filter the complete task dataset so they reflect true last-N-days results, independent of policy.lookbackDays which continues to control policy-driven team views.
• The server mounts demo APIs under /api (e.g. /api/team-members, /api/tasks/with-details).
• The Slack bot is a minimal, in-memory demo — restarting the process resets state.

• If the client can't reach the API, set REACT_APP_API_BASE_URL and restart the client.
• If ports conflict, change PORT in the server environment or run the client on a different port.

If you'd like this README to include CI steps, deployment guidance, or a short architecture diagram, tell me what to add and I will update it.