robocorp/robocorp

What it solves

It provides a comprehensive framework for building software automations in Python, enabling developers to automate browser actions, Windows desktop applications, and data handling (like Excel) while integrating with a centralized control system for secrets, logging, and work items.

How it works

The project offers a suite of specialized Python libraries that handle different automation domains. It uses a meta-package for core framework features and separate libraries for specific tasks: robocorp.browser (via Playwright) for web automation, robocorp.windows for desktop apps, and robocorp.excel for spreadsheet management. It also includes an "AI coding agent skill" that allows AI agents like Claude Code or Cursor to bootstrap and run these automation packages.

Who it’s for

Python developers and automation engineers who need to build scalable, maintainable automation tasks that can be managed via a control room.

Highlights

• AI Agent Integration: Includes a specific skill to teach AI coding agents how to build and validate Robocorp tasks.
• Broad Automation Scope: Supports web, Windows desktop, and Excel automation.
• Control Room Integration: Built-in libraries for managing secrets (Vault), assets (Storage), and data flow (Work Items).
• AI Actions: Provides sema4ai-actions to create "AI Actions" that give AI agents the ability to interact with software.