Server facts
uindow MCP Server
Project
- Author
- uindow
- Repository
- github.com/uindow/uindow
- License
- Not specified
- Category
- Browser Automation
Tech
- Runtime / language
- Not specified
- Transports
- stdio
- Remote available
- No
- Authentication
- Not specified
Activity
- GitHub stars
- 8
- Last commit
- Data as of
Tags
- Browser Automation
- Playwright
- Web Scraping
- ai browser
- atlas
- browser
- cypress
- no code
- puppeteer
- selenium
- testim
- trusted events
- web
- web automation
Overview
Uindow centers browser automation on a local Chromium/Electron application, using operating-system input for cursor movement, typing, and native file dialogs rather than simulated page events. Users can record browser interactions into JavaScript, develop modules in the integrated editor, or have an MCP-compatible assistant author and operate automations. Its stdio connection exposes application controls alongside tools for discovering, creating, renaming, and deleting agents. Separate execution controls start, resume, or stop an agent, while status checks and logs provide visibility into its progress.
The toolset also supports inspecting pages through PNG screenshots or structured descriptions of visible elements. For immediate actions, an assistant can execute JavaScript through Uindow's SDK; for saved workflows, it can edit module states and functions, organize their keys, and export the source as YAML. Input tools define settings such as text, file selections, and tables, while output tools declare results that can be retrieved after execution. SDK documentation and a sample module are available through dedicated tools. Using the application requires a free Uindow account.
uindow setup notes
- uindow records stdio only; confirm a network transport upstream before planning remote access.
- uindow has no recorded license; confirm reuse terms in its repository.