create .env file

# replace by your gpt-4o own
OPENAI_API_KEY="YOUR_TOKEN"

Refer to this document if your want to use other models like Qwen: https://midscenejs.com/choose-a-model

Ensure that Node.js is installed. Install the @midscene/cli globally

npm i -g @midscene/cli

Run all scripts

For windows, you need to replace ./ with .\, like midscene .\midscene-scripts\.

midscene ./midscene-scripts/

Extract status info from github status page

midscene ./midscene-scripts/extract-github-status.yaml

Perform a testing case on sauce demo

midscene ./midscene-scripts/sauce-demo.yaml

Perform a search on ebay.com

midscene ./midscene-scripts/search-headphone-on-ebay.yaml

Serve the server_root folder as a static server and test the index.html file

midscene ./midscene-scripts/local-static-server.yml

Run a script with headed mode (i.e. you can see the browser window when running)

midscene --headed ./midscene-scripts/sauce-demo.yaml

Keep the browser window open after the script finishes

midscene --keep-window ./midscene-scripts/sauce-demo.yaml

https://midscenejs.com/automate-with-scripts-in-yaml.html