Skip to content

a7mddra/squigit

Folders and files

NameName
Last commit message
Last commit date

Latest commit

Β 

History

1,022 Commits
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 

logo

Squigit CI squigit License

Squigit demo

Squigit is an open-source "Circle to Search" desktop utility that brings the power of AI vision directly to your screen. It provides seamless screen capture, analysis, and visual search capabilities with a single hotkey.

Explore detailed guides and architecture in our documentation.

πŸš€ Why Squigit?

  • 🎯 Native Performance: Polyglot architecture with C++ capture, Rust orchestration, and Tauri UI.
  • πŸ”— Bring Your Own Key (BYOK): No middleman. Plug your own Gemini and ImgBB API keys for direct, cost-effective access.
  • ⚑ Instant Global Hotkey: Super/Cmd/Win + Shift + A freezes your desktop for instant capture from anywhere.
  • 🧠 Multi-Model Support: Switch between Gemini 3.1 Flash (efficiency) and Gemini 3.1 Pro (power).
  • πŸͺ„ Google Lens Integration: One-click visual search by automatically hosting images and launching Google Lens.
  • πŸ”’ Local-First Security: Your API keys, images, and conversations never leave your machine.

πŸ“¦ Download

Pre-requisites before installation

  • Windows 10+, macOS 12+, or Linux (X11/Wayland)
  • 64-bit system

Quick Install

Download the appropriate installer for your operating system:

Windows

Download Squigit_Installer.exe.

macOS

Download squigit-installer.dmg.

Linux

Download squigit-installer.

Installation

Grab the latest version of Squigit for your operating system from our Latest Releases page.

πŸͺŸ Windows

  1. Download the latest Windows setup executable.
  2. Run the installer.
  3. SmartScreen Warning: Because Squigit is an open-source project without a commercial EV certificate, Microsoft Defender SmartScreen may initially flag the installer as unrecognized.
    • Fix: If you see the "Windows protected your PC" screen, simply click More info, and then select Run anyway.

🍎 macOS (Apple Silicon / ARM only)

Note: Squigit's local AI engines are currently optimized strictly for Apple Silicon (M1/M2/M3/M4). Intel Macs are not supported.

  1. Download the latest .dmg file.
  2. Open it and drag Squigit into your Applications folder.
  3. Gatekeeper / "App is damaged" Warning: Because we use an ad-hoc open-source signature, macOS Gatekeeper may block the app from launching or claim the file is damaged.
    • Fix: Open your Terminal and run this command to clear the Apple quarantine flag:
      sudo xattr -rd com.apple.quarantine /Applications/Squigit.app
    • (Alternatively: Go to System Settings -> Privacy & Security, scroll to the bottom, and click "Open Anyway").

🐧 Linux (AppImage only)

Note: Squigit is distributed as an AppImage to ensure maximum compatibility across different Linux distributions.

  1. Download the latest .AppImage file.
  2. Make it Executable: By default, Linux prevents new downloads from running as programs. You must grant the file execution permissions before launching it.
    • Fix: Open your terminal in the directory where you downloaded the file and run:
      chmod +x Squigit-*.AppImage
    • You can now double-click the file to run it, or integrate it into your desktop environment using tools like AppImageLauncher.

⭐ Key Features

Instant Screen Capture & Analysis

  • Press Super/Cmd/Win + Shift + A anywhere to freeze your screen
  • Draw any shape around content with intelligent cropping
  • Get immediate AI analysis powered by Google Gemini models
  • Continue conversations with follow-up questions

Multi-Modal AI Integration

  • Direct Gemini API Access: Your prompts go straight to Google's models with zero hops
  • Visual Context Injection: Screenshots are automatically included in chat context
  • Model Hot-Swapping: Instantly switch between Gemini models for different tasks

Cross-Platform Visual Search

  • One-Click Google Lens: Upload screenshots to ImgBB and open in Google Lens
  • Secure Image Handling: Local encryption with AES-256 for API keys
  • Privacy-First: No central server, no data collection

Advanced Desktop Integration

  • Global Hotkey Registration: Works across all applications
  • Multi-Monitor Support: Captures and manages all displays simultaneously
  • Platform-Optimized Capture: Different strategies for X11, Wayland, Windows, and macOS

πŸ” Authentication: Bring Your Own Key (BYOK)

Squigit operates on a zero-trust, local-first model where you provide your own API keys.

Setup Process

  1. Launch Squigit and signup
  2. Generate API Keys on the official provider sites:
  3. Copy to Clipboard - Squigit automatically detects and securely stores your keys

Security Guarantees

  • Local Storage: Keys encrypted with AES-256 and stored in your user directory
  • Direct Communication: No proxy servers - requests go directly to generativelanguage.googleapis.com and api.imgbb.com
  • Clipboard Safety: Keys are never manually pasted; automatic detection prevents keylogging exposure
  • Revocable Access: You control and can revoke keys anytime via provider dashboards

πŸš€ Getting Started

Basic Usage

Instant Capture (Hotkey Method)

  1. Press Super/Cmd/Win + Shift + A
  2. Draw a shape around the content you want to analyze
  3. Release to capture and open Squigit with AI analysis

Direct Application Launch

  1. Open Squigit from your applications menu
  2. Drag & drop an image or use the file dialog
  3. Chat with the AI about your uploaded image

Google Lens Integration

  1. Capture or upload an image
  2. Click the "Google Lens" button in the chat interface
  3. Visual search opens in your default browser

Quick Examples

Analyze Code from Screen

# 1. Press Super+Shift+A
# 2. Draw around the code snippet
# 3. Ask: "Explain this function and suggest improvements"

Research Web Content

# 1. Press Super+Shift+A on an article
# 2. Draw around the text
# 3. Ask: "Summarize the key points and find related sources"

Visual Product Search

# 1. Press Super+Shift+A on a product image
# 2. Draw around the product
# 3. Click "Google Lens" to find it online

πŸ“š Documentation

Getting Started

Architecture & Development

Development & Contribution

Deployment & Operations

API Reference

Policies

🀝 Contributing

We welcome contributions! Squigit is fully open source, and we encourage the community to:

  • Report bugs and suggest features
  • Improve documentation
  • Submit code improvements
  • Share platform-specific optimizations

See our Contributing Guide for development setup, coding standards, and how to submit pull requests.

Check our TODO List for planned features and priorities.

πŸ“– Resources

πŸ“„ Legal


Built with ❀️ by the open source community