Browse, test, and automate websites from what’s visible.
browser automation
Automates browsing, website interaction, data extraction, screenshots, and visual UI testing.
When to use it
Use for browsing, extracting website data, interacting with pages, taking screenshots, testing UI behavior, or automating web workflows.
Give it a browser task; it performs the requested workflow and reports results, findings, and generated file paths.
This skill
websites
Fills forms and uploads files
user's Chrome browser
Navigates your active Chrome tab
npx -y @midscene/web@1
Midscene through npx
background Chrome process
Starts a persistent Chrome process
Requires a configured model with strong visual grounding capabilities.
Reads the model provider API key from system environment variables or a .env file in the current working directory.