Library/midscene
Library/midsceneForked

web-infra-dev/midscene

midscene

AI-powered, vision-driven UI automation for every platform.

Builder

web-infra-dev

web-infra-dev

web-infra-dev • individual

Stars

12,492

Using upstream star count

Forks

925

Using upstream fork count

Open Issues

0

Activity Score

0/100

0 commits in 30d

Created

Jul 23, 2024

Project creation date

README Summary

Midscene is an AI-powered UI automation framework that uses computer vision to interact with web applications across different platforms. It provides intuitive APIs for performing actions like clicking, typing, and extracting data from UI elements using natural language descriptions instead of traditional selectors.

AI Dev Skills

Unmapped

Computer VisionMultimodal AIVision-Language ModelsUI UnderstandingAutomated TestingCross-platform AutomationNatural Language Interface Design

Tags

Computer VisionMultimodal AIVision-Language ModelsUI UnderstandingAutomated TestingCross-platform AutomationNatural Language Interface DesignSoftware TestingEnd-to-end TestingMultimodal ReasoningMobile App DevelopmentSelf-hostedTextAI-powered Developer ToolsMobile App TestingLocal DevelopmentCross-platform UI AutomationVisual Regression TestingDesktop Application TestingWeb DevelopmentAutomated UI TestingAgentic AIWeb ScrapingMultimodalQuality AssuranceCI/CD PipelinesDeveloper ToolsImageBrowser/WASMCI/CD Pipeline IntegrationTypeScript

Taxonomy

Recent Activity

Updated 23 days ago

7 Days

0

30 Days

0

90 Days

0

Quality

beta
Quality
high
Maturity
beta

Categories

MLOps & InfrastructurePrimaryDev Tools & AutomationNLP & TextML Platform & InfrastructureMultimodal AIEdge & Mobile AIOther AI / MLFoundation ModelsAI AgentsComputer VisionRobotics

PM Skills

Scale & ReliabilityDeveloper Platform

Languages

TypeScript100.0%

Timeline

Project created
Jul 23, 2024
Forked
Mar 23, 2026
Your last push
23 days ago
Upstream last push
6 days ago
Tracked since
Mar 21, 2026

Similar Repos

pgvector cosine similarity · $0

Loading…