Intellacc

Prediction-market social platform with encrypted messaging and community-driven market creation.

What you get

• Social feed: posts, comments, attachments, profiles, and leaderboards.
• Markets: Rust LMSR prediction engine with events, outcomes, and trading.
• Persuasive Alpha: users submit prediction questions, validators review and approve, and rewards are auto-distributed by rule.
• E2EE Messaging: one-to-one and group chats with OpenMLS (WASM), plus safety-number and device trust flows.
• Security: passkey + PRF unlock support, device linking, and tiered verification.
• Moderation: AI-assisted content flagging and admin report/review paths.

Stack

• Frontend: frontend (VanJS + Vite) on :5173
• Backend: backend (Express + Socket.IO) on :3000
• Prediction engine: prediction-engine (Rust/Axum) on :3001
• Database: PostgreSQL
• Reverse proxy: Caddy on 80/443

Quick start

1. docker network create intellacc-network
2. Configure env files: backend/.env, prediction-engine/.env
3. docker compose up -d
4. Open:
   • App: http://localhost:5173
   • API: http://localhost:3000
   • Engine: http://localhost:3001

Useful commands

• docker compose up -d
• docker compose down
• docker compose logs -f
• docker compose ps
• docker exec intellacc_backend npm test
• docker exec intellacc_frontend npm test
• ./scripts/test_prediction_engine.sh (standard quick prediction-engine suite)
• ./scripts/test_prediction_engine.sh --full (full prediction-engine suite)
• docker exec -it intellacc_db psql -U intellacc_user -d intellaccdb

Docs

• docs/unified-backlog.md
• docs/persuasive-alpha-v1-implementation-plan.md
• docs/mls-status.md
• docs/verification-implementation-plan.md

Test users

• user1@example.com / password123
• user2@example.com / password123