https://github.com/SmetDenis/raycast-gemai

GemAI is a Raycast extension designed to supercharge your productivity by integrating Google's Gemini AI models directly into your workflow.
https://github.com/SmetDenis/raycast-gemai

ai gemini gemini-pro productivity raycast raycast-extension

Last synced: 9 months ago
JSON representation

GemAI is a Raycast extension designed to supercharge your productivity by integrating Google's Gemini AI models directly into your workflow.

Host: GitHub
URL: https://github.com/SmetDenis/raycast-gemai
Owner: SmetDenis
License: mit
Created: 2025-05-07T22:56:33.000Z (about 1 year ago)
Default Branch: main
Last Pushed: 2025-07-23T09:13:51.000Z (11 months ago)
Last Synced: 2025-07-23T11:16:07.198Z (11 months ago)
Topics: ai, gemini, gemini-pro, productivity, raycast, raycast-extension
Language: TypeScript
Homepage:
Size: 535 KB
Stars: 3
Watchers: 1
Forks: 1
Open Issues: 0
Metadata Files:
- Readme: README.md
- License: LICENSE

Awesome Lists containing this project

README

          # Raycast GemAI - Universal AI Assistant

**Raycast GemAI** is a powerful Raycast extension that brings both **Google Gemini** and **OpenAI GPT** models directly into your workflow. Perform text processing, complex reasoning, image analysis, and translation – all without leaving Raycast.

> 🇷🇺 **Русская версия:** [README_RUS.md](README_RUS.md)

>

> 🇨🇳 **中文版本:** [README_CN.md](README_CN.md)

## ✨ Key Features

- **🤖 Universal AI Support:** Seamlessly switch between Google Gemini and OpenAI models with automatic provider detection

- **🧠 Advanced Reasoning:** Full support for OpenAI's o1-preview and o1-mini with enhanced thinking capabilities

- **🖼️ Vision Analysis:** Image processing with GPT-4o and Gemini Vision, automatic model switching for optimal results

- **💬 Chat Room:** Interactive conversations with persistent context memory

- **📝 Text Tools:** Comprehensive suite for writing, editing, translation, and summarization

- **📊 Usage Analytics:** Real-time token tracking and accurate cost calculation

- **🎨 Custom Prompts:** Use your own Markdown files to customize AI behavior per command

## 🤖 Supported Models & Pricing

### OpenAI Models

| Model | Type | Input/Output Cost (per 1M tokens) | Best For |

|-------|------|-----------------------------------|----------|

| **GPT-4o** | Vision + Text | $2.50 / $10.00 | General tasks, image analysis |

| **GPT-4o-mini** | Vision + Text | $0.15 / $0.60 | Fast, cost-effective tasks |

| **o1-preview** | Reasoning | $15.00 / $60.00 | Complex problem solving |

| **o1-mini** | Reasoning | $3.00 / $12.00 | Efficient reasoning tasks |

### Google Gemini Models

| Model | Type | Input/Output Cost (per 1M tokens) | Best For |

|-------|------|-----------------------------------|----------|

| **Gemini 2.0 Flash-Lite** | Vision + Text | $0.075 / $0.30 | Fast, cost-effective tasks |

| **Gemini 2.0 Flash** | Vision + Text | $0.10 / $0.40 | Balanced performance |

| **Gemini 2.5 Flash** | Vision + Text | $0.15 / $0.60 | Enhanced performance |

| **Gemini 2.5 Flash Thinking** | Vision + Text | $0.15 / $0.60 + $3.50 thinking | Advanced reasoning |

| **Gemini 2.5 Pro** | Vision + Text | $1.25 / $10.00 | Complex reasoning and analysis |

## 📋 Available Commands

| Command | Description | Input Type | AI Provider |

|---------|-------------|------------|-------------|

| **Ask AI** | Ask questions using any available model | Text/Selection | Universal |

| **Chat Room** | Interactive chat with persistent context | Text Input | Universal |

| **Explain It** | Detailed explanations with context | Selection | Universal |

| **Summarize It** | Intelligent text summarization | Selection | Universal |

| **Rephrase It** | Rewrite while preserving meaning | Selection | Universal |

| **Fix Grammar** | Grammar and style correction | Selection | Universal |

| **Professional/Friendly Tone** | Transform text tone | Selection | Universal |

| **Make Longer/Shorter** | Expand or condense content | Selection | Universal |

| **Translate** | Multi-language translation | Selection | Universal |

| **Count Tokens** | Estimate costs and optimize usage | Text/Selection | Universal |

| **Screenshot → Markdown** | Convert images to formatted Markdown | Screenshot | Vision Models |

| **Screenshot → Explain** | Analyze and describe images | Screenshot | Vision Models |

| **Screenshot → Translate** | Extract and translate image text | Screenshot | Vision Models |

| **Usage Statistics** | Detailed analytics and costs | - | - |

## 🚀 Quick Start

### Installation

1. **Raycast Store (Recommended):** Search for "GemAI" in Raycast Store

2. **Local Development:**

   ```bash

   git clone https://github.com/smetdenis/raycast-gemai.git

   cd raycast-gemai

   npm install && npm run build

   ```

### Configuration

Access settings via **Raycast Preferences → Extensions → GemAI**:

**Required:**

- **Gemini API Key:** Get from [Google AI Studio](https://aistudio.google.com/app/apikey)

- **OpenAI API Key:** Get from [OpenAI Platform](https://platform.openai.com/api-keys)

**Optional:**

- **Default Model:** Choose preferred model for all commands

- **Custom Models:** Add OpenAI-compatible models (Azure, local deployments)

- **Temperature:** Control AI creativity (0.0 = focused, 1.0 = creative)

- **Custom Prompts:** Point to directory with custom system prompts

- **Languages:** Set primary/secondary languages for translation

## 🎨 Custom System Prompts

Create custom Markdown files to override built-in prompts:

1. **Create prompts directory:** `mkdir ~/Documents/Prompts/Raycast`

2. **Create prompt files** (e.g., `AskQuestion.md`):

   ```markdown

   # Custom Ask AI Prompt

   

   You are a specialized technical consultant with expertise in software development.

   Always provide practical, actionable advice with code examples when relevant.

   ```

3. **Configure in Raycast:** Set **Prompt Directory** and specify custom file names per command

**Available prompt files:** `AskQuestion.md`, `ChatRoom.md`, `Explainer.md`, `Grammar.md`, `Professional.md`, `Friend.md`, `Translator.md`, `Screenshot-Explain.md`, `Screenshot-Markdown.md`

## 💡 Usage Examples

```bash

# Text Processing

Select text → Fix Grammar: "there dog is running" → "Their dog is running"

Select text → Professional: "hey, can u help?" → "Could you please assist me?"

# Advanced Reasoning (o1-series)

Ask AI → "Solve this differential equation step by step..."

# Vision & Screenshots

Take screenshot → Screenshot → Markdown: Converts UI to formatted Markdown

Take screenshot → Screenshot → Explain: "What's happening in this diagram?"

# Smart Model Switching

o1-mini + Screenshot → Auto-switches to GPT-4o for image processing

```

## 🔧 Development

### Key Features

- **Universal Architecture:** Single codebase supporting multiple AI providers

- **Reasoning Model Support:** Full o1-series compatibility

- **Vision Auto-switching:** Intelligent model selection for multimodal tasks

- **Real-time Analytics:** Accurate token counting and cost calculation

### Building

```bash

npm install     # Install dependencies

npm run dev     # Development with watch

npm run build   # Production build

```

## 🤝 Contributing

1. Fork the repository

2. Clone locally: `git clone https://github.com/smetdenis/raycast-gemai.git`

3. Make changes following TypeScript strict mode and backward compatibility

4. Test with both Gemini and OpenAI models

5. Submit pull request with detailed description

## 🔒 Privacy & License

- **Privacy:** API keys stored securely in Raycast, no data stored or transmitted to third parties

- **License:** [MIT License](LICENSE)

- **Policies:** Subject to Google Gemini and OpenAI privacy policies

---

**Raycast GemAI** - Your universal AI assistant, powered by the best models from Google and OpenAI.

For support: [GitHub Issues](https://github.com/smetdenis/raycast-gemai/issues)

ecosyste.ms

Data

Tools

Indexes

Applications

Experiments

Awesome

https://github.com/SmetDenis/raycast-gemai

Awesome Lists containing this project

README