pot-app/pot-desktop

★ 17,961JavaScriptAudience · generalComplexity · 2/5Setup · easy

Mindmap

mindmap
  root((pot-desktop))
    What it does
      Text translation
      Screen OCR
      Keyboard shortcuts
    Providers
      OpenAI
      DeepL
      Google Translate
      Local Ollama
    Platforms
      Windows
      macOS
      Linux Wayland
    Use cases
      Multilingual work
      Image text extraction
      Language learning

mindmap root((pot-desktop)) What it does Text translation Screen OCR Keyboard shortcuts Providers OpenAI DeepL Google Translate Local Ollama Platforms Windows macOS Linux Wayland Use cases Multilingual work Image text extraction Language learning

Click or tap to explore — scroll the page freely

Things people build with this

USE CASE 1

Instantly translate any selected text on your screen without switching apps, using a keyboard shortcut.

USE CASE 2

Extract text from a screenshot or on-screen image using OCR, then translate or copy it.

USE CASE 3

Use your preferred translation service, DeepL, OpenAI, Google, or a local Ollama model, rather than being locked to one provider.

USE CASE 4

Export translated words to an Anki deck for language learning.

Tech stack

JavaScriptTypeScriptRustTauri

Getting it running

Difficulty · easy Time to first run · 30min

Some translation services require API keys from third-party providers such as DeepL or OpenAI before they can be used.

In plain English

Pot is a cross-platform desktop app for translating text and reading text out of images (OCR, optical character recognition, the process of turning a picture of words into editable text). You install it on your computer and use keyboard shortcuts to translate whatever you have selected, whatever is on your clipboard, or whatever you draw a box around on the screen. It works on Windows, macOS, and Linux, including Linux setups using Wayland (a modern display system used by KDE, Gnome, and Hyprland). Rather than locking you to one translation provider, Pot lets you plug in a long list of online services like OpenAI, Gemini Pro, DeepL, Google, Bing, Yandex, and a range of Chinese services such as Baidu, Tencent, Youdao, and Volcano. It can also run offline against a locally installed Ollama instance for translation, and against system OCR on each platform (Windows Media OCR, Apple's Vision framework, and Tesseract on Linux) plus Tesseract.js. A plugin system using files with a .potext extension lets the community add new translation services, OCR engines, text-to-speech backends, and vocabulary-book exporters such as Anki. You would reach for Pot if you read or write in multiple languages on a daily basis, for example, working across English and Chinese documentation, captioning screenshots, or copying text out of images. It is also useful when you want to pick your own translation provider rather than be tied to one company's cloud. The app is built with Tauri (a framework for building lightweight desktop apps with web technologies), with a JavaScript user interface and a Rust core. The full README is longer than what was provided.

Copy-paste prompts

Prompt 1

How do I set up Pot on macOS to translate selected text using DeepL whenever I press a keyboard shortcut?

Prompt 2

Walk me through configuring Pot to use a locally running Ollama model for offline translation instead of a cloud service.

Prompt 3

How do I set up Pot on Linux with a Wayland desktop so OCR and translation keyboard shortcuts work correctly?

Prompt 4

How do I connect Pot to my OpenAI API key and use GPT for translations instead of the default providers?

Prompt 5

Show me how to export words I have translated in Pot directly into an Anki deck so I can study them later.

Open on GitHub → Explain another repo

← pot-app on gitmyhub — every repo by this author, as a profile.

Verify against the repo before relying on details.