MCP ExplorerExplorer

Everywhere

@DearVaon 10 months ago
8 Apache-2.0
FreeCommunity
AI Systems
#agent#ai#avalonia#chat#deepseek#llm#mcp#ollama#openai#rag#ui-automation
[WIP] Seamless AI Assistance that brings your Favorite LLM in Every app, Every time, Every where.

Overview

What is Everywhere

Everywhere is a cross-platform AI assistant designed to provide contextual, multimodal AI support within any desktop application, enhancing productivity in tasks such as browsing, writing, designing, or coding.

Use cases

Use cases include generating text in documents, translating phrases in web browsers, summarizing information from applications, and providing code explanations or visual Q&A based on screenshots.

How to use

Users can invoke the assistant in any text field to generate, rewrite, translate, or summarize content without disrupting their workflow. It also offers screen awareness to provide context-specific suggestions.

Key features

Key features include universal input assistance, screen awareness for contextual suggestions, MCP integration for automation, multimodal understanding for processing various types of content, and a privacy-first approach allowing users to choose between cloud and local models.

Where to use

Everywhere can be utilized in various fields including software development, content creation, design, and any other area where desktop applications are used.

Content

前往中文版本 »

Contributors
Forks
Stargazers
Issues
MIT License


Banner


🚧 NOTICE: This project is under active development and is not yet ready for use 🚧


(WIP) Explore the docs »

View Demo · Report Bug · Request Feature

About Everywhere

Everywhere is a cross‑platform AI assistant built with .NET and Avalonia. It delivers contextual, multimodal AI support directly inside any desktop application—whether you’re browsing, writing, designing, or coding.

✨ Key Features

  • Universal Input Assistance – Call the assistant in any text field to generate, rewrite, translate, or summarize without leaving your workflow.
  • Screen Awareness – Capture and analyze the current window or selected region to offer context‑specific suggestions and summaries.
  • MCP Integration – Leverage the Model Context Protocol to wire up custom tools and data sources for deeper automation.
  • Multimodal Understanding – Process text, images, and screenshots for code explanations, UI analysis, or visual Q&A.
  • LLMs under control – Choose between cloud LLMs (ChatGPT, DeepSeek, etc.) and fully local models (Ollama, LM Studio).

🛠 Built With

.NET CoreAvalonia


📋 System Requirements

Platform Minimum Version
Windows 10.0.19041.0
macOS Planned

🚀 Getting Started

Installation

Note: Pre‑built installers are not yet available; build from source for now.

✈️ Roadmap

  • [ ] Edit assistant (WIP)
  • [ ] Screen-based assistant
  • [ ] Web searching & thinking
  • [ ] MCP Tools
  • [ ] Voice input
  • [ ] RAG & Knowledge base

🤝 Contributing

We welcome issues, feature ideas, and PRs! See CONTRIBUTING.md for guidelines.

Top contributors:

contrib.rocks image

📄 License

Distributed under the Apache 2.0 License. See LICENSE for more information.

🙏 Acknowledgments

(back to top)

Tools

No tools

Comments

Recommend MCP Servers

View All MCP Servers