screen-reading 开源项目推荐（12 个）

FrostyIsBored /FH6-Auction-House-Sniper

Forza Horizon 6（FH6）的自动拍卖行狙击机器人。屏幕阅读自动收购。Windows，开源。

peakmojo /macos-visual-agent

On-device screen understanding for macOS

xlly-s /FORZA-6-AUCTION-HOUSE-SNIPER

一个致力于极限极限竞速地平线6拍卖行的狙击机器人，使用蟒蛇和图像重建。

yhm0619 /dsh-pet-read

基于 dsh-pet 的互动碎碎念：每 30 秒识屏，用视觉模型在宠物头顶随口吐槽（不需要回复、不带表情包）。

wenhefu /forza-horizon-6-helper

地平线6 自动助手 / Forza Horizon 6 helper — 挂机刷车刷钱、EventLab 刷图、拍卖场抢车、清理重复车、一键买光与统计未拥有车辆。仅虚拟 Xbox 手柄 + 只读截屏，不注入、不改游戏文件。Forza Horizon 6 (FH6) AFK money/car farming, auto-buy, auction sniping, duplicate cleanup, unowned-car survey — virtual gamepad + screen reading, no game injection.

Remade-With-Rust /carmenta

OCR in pure Rust: documents, screens, and a change-gated LIVE frame reader. Detection, recognition and reading order computed from geometry, no Python, no C/C++ by default, nothing GPL. Memory safe. The ffai-carmenta crate.

Peter-Ufens /vision-public

Oeil PC : lire l'ecran (grille, OCR, veille active). Labo perso, autodeveloppement IA. Apache-2.0.

Hirkadawg /FullyLocal_AI_DesktopCompanion

A fully local AI desktop companion for Windows: it reads what is on your screen and talks about it, typed or spoken, in English or Turkish, and remarks on its own. Runs entirely offline through Ollama — no account, no telemetry, no cloud.

bochen2029-pixel /glance

The window, as text, as it changes: reads a window through its own accessibility tree, or through pixels and the OS's OCR where there is no tree (Citrix, RDP), and turns every change into one timestamped line, a fact for a clock, or JSON for an agent. Never types, never clicks, never sends a byte. One Windows exe, no network stack, no model. MIT.

DirectionStudios /Desire

Meet Desire: A more capable and advanced version of you as a personal assistant to you. With local-offline priority, the actual ai agent that's been promised has been finally delivered.

devcell-sh /go-textshot

Extract text and UI elements from screenshots in Go, via OCR and icon-detection models run through ONNX Runtime

Parth-Gupta-github /FloatNote

Real-time meeting intelligence — transcription, screen reading, AI summaries, and a chatbot that knows your meeting.