mwin

mwin: Track OpenAI, Claude, Gemini and OpenAI-compatible models then give solutions to improve your agent system.
Our goal is to make llm application more valuable and effortlessly improve llm capabilities.

Quickstart

Mwin is very easy to start. You need Docker and Docker Compose before deployment. After starting the containers, you can see track contents on web http://localhost:5173/.

Deployment

Start all services with Docker Compose:

docker compose up --build -d

This starts:

• aitrace-postgres on localhost:16432
• aitrace-backend on localhost:8080
• aitrace-web on http://localhost:5173/

The backend stores tracked Base64 images in /data/mwin/media. Docker Compose keeps database data and media files in persistent volumes:

• aitrace_pgdata
• aitrace_media

Stop the deployment with:

docker compose down

If you need to build and run images manually, use these commands:

docker build -t aitrace-backend .
docker build --target web-runtime -t aitrace-web .
docker build --target postgres -t aitrace-postgres .

docker network create aitrace || true
docker run -d --name aitrace-postgres --network aitrace -p 16432:5432 -v aitrace_pgdata:/var/lib/postgresql/data aitrace-postgres
docker run -d --name aitrace-backend --network aitrace -e SPRING_DATASOURCE_URL=jdbc:postgresql://aitrace-postgres:5432/aitrace -v aitrace_media:/data/mwin/media -p 8080:8080 aitrace-backend
docker run -d --name aitrace-web --network aitrace -p 5173:80 -e BACKEND_HOST=aitrace-backend aitrace-web

Python-sdk

Mwin python sdk need an api key. So you have to be sure that you deploys successfully. Then enter http://localhost:5173/ to get your api key.
Click here to know how to use mwin python-sdk.

Development

Mwin project package manager is uv. If you are a beginner uver, please click uv link: uv official link

cd sdk
uv sync
uv .venv/Script/activate

You can watch more detailed debug information by using --log-level=DEBUG or set AT_LOG_LEVEL=DEBUG for Windows or export AT_LOG_LEVEL=DEBUG for Linux and Mac.