OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
Agent Frameworks

Cradle

Cradle is an open-source Python framework for foundation-model agents to interact with computer tasks through screenshots as input and keyboard and mouse operations as output.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Cradle — OpenViST project profile in Agent Frameworks.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use Cradle when developing a foundation-model agent that must perceive a computer screen and interact with a game or software through ordinary keyboard and mouse input.

Who it is for

  • Developers building agents that operate visible computer environments.
  • Researchers and engineers adapting a computer-control framework to games or desktop software.

Documented uses

  • Build agents that interpret screenshots and act through keyboard and mouse operations.
  • Configure computer-control agents for supported games and software examples.
  • Organize reusable skills for computer tasks within a standardized environment.
Before you choose

Requirements & limits

The public repository is licensed under MIT. Its setup instructions call for Python 3.10, dependencies, OCR tooling and an API key for at least one supported model provider: OpenAI or Claude.

  • Cradle is a development framework rather than a ready-to-use desktop agent.
  • Each game or software environment has its own settings and linked setup guide.
  • The documented model-provider setup covers OpenAI and Claude.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Prepare Cradle’s Python and OCR environment

Set up the repository’s documented Python environment and install the OCR model.

Before you start

  • Python 3.10
  • Conda
  • The Cradle GitHub repository

Steps

  1. Clone the Cradle repository and change into its directory.

  2. Create and activate a Conda environment with Python 3.10.

  3. Install the repository dependencies with pip install -r requirements.txt.

  4. Install spaCy’s en_core_web_lg model using one of the README’s listed methods.

Expected outcome

The documented Python dependencies and OCR model are installed; environment-specific configuration is the next stage.

Guide limitations

  • This covers environment setup only, not running an agent or configuring a particular game or application.
  • The README directs users to separate environment guides because target settings vary.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
13 October 2026
Original website title
BAAI-Agents/Cradle · Official homepage title · retained in its source language
How we review sourcesSuggest a correction