MistralProto

UI prototype generation from requirements descriptions using Mistral AI.
Developed as part of the Sber X PROSTO X ITMO Hackathon

Quick Start

1. Install dependencies

npm install

2. Configuration

Create a .env file in the project root:

MISTRAL_KEY=QWERTYEXAMPLE

Optional:

MISTRAL_MODEL=mistral-small-latest  
API_PORT=3001                        
SERVER_PORT=3000                   
DATABASE_URL=postgresql://user:pass@localhost:5432/mistralproto 

PostgreSQL is used to store sessions and dialogues for the API. The database schema is created automatically when the API starts, or can be created manually:

npm run db:migrate

3. Run

CLI:

npm start

Enter your requests in the console. Files from the input/ directory are supported (.txt, .pdf, .docx).

• exit — exit the application
• save — save the latest result to output/
• Filename (e.g. requirements.txt) — load text from the input/ directory

---

API Server:

npm run api

API: http://localhost:3001 (or the port specified by API_PORT).

Sessions and dialogues (when DATABASE_URL is configured):

• POST /api/sessions — create a session
• POST /api/sessions/:id/dialogues — create a dialogue
• GET /api/sessions/:id/dialogues — get a list of dialogues
• GET /api/dialogues/:id?session_id= — get a dialogue with its messages
• POST /api/dialogues/:id/send — send a message, run the pipeline, and save the result to the database

WEB:

src/frontend/index.html