Witsy
A desktop AI assistant and universal MCP client that works with any LLM provider, offering chat, image/video generation, and system-wide productivity tools.
What is Witsy?
Witsy is a desktop AI assistant and universal Model Context Protocol (MCP) client that allows users to interact with a wide range of large language models and AI services from a single application. It solves the problem of AI provider fragmentation by offering a unified interface for chat, image/video generation, speech processing, and document analysis while letting users bring their own API keys.
Developers, power users, and professionals who regularly use multiple AI services and want a customizable, privacy-focused desktop assistant with system-wide integration and automation capabilities.
Witsy stands out as one of the first universal MCP clients, offering unparalleled provider flexibility, deep system integration with features like Prompt Anywhere and AI Commands, and a strong focus on user control through its BYOK model and local operation.
Overview
Witsy is a desktop AI assistant that functions as a universal Model Context Protocol (MCP) client, allowing users to connect to virtually any large language model provider. It provides a centralized interface for AI-powered tasks while keeping users in control of their API keys and data.
Key Features
- Universal MCP Client — Connect and use any LLM provider that supports the MCP standard or OpenAI API compatibility.
- Multi-Modal AI — Supports chat, image generation, video creation, text-to-speech, and speech-to-text across numerous providers.
- Prompt Anywhere — Generate content directly in any application using a global keyboard shortcut.
- AI Commands — Apply LLM-powered transformations to selected text in any app via customizable commands.
- Document Chat (RAG) — Chat with local files and documents using retrieval-augmented generation.
- HTTP API & CLI — Programmatic access via a local HTTP server and command-line interface for automation.
- Agent Webhooks — Trigger AI agents via HTTP requests for integration with external systems and workflows.
Philosophy
Witsy embraces a BYOK (Bring Your Own Keys) philosophy, prioritizing user control over API keys and data while providing a unified, extensible interface to the fragmented AI provider ecosystem.
Use Cases
Best For
- Using multiple LLM providers (OpenAI, Anthropic, Google, Ollama, etc.) from one interface
- Generating and editing content directly within any desktop application
- Building automated workflows that trigger AI agents via HTTP webhooks
- Chatting with local documents and files using retrieval-augmented generation
- Transcribing audio and performing speech-to-text with various cloud and local models
- Creating custom AI commands for text transformation across all applications
Not Ideal For
- Teams needing centralized billing and shared API key management
- Users who require mobile or web-based AI access on the go
- Beginners seeking a simple, all-in-one AI tool without configuration hassle
- Projects demanding real-time multi-user collaboration features
Pros & Cons
Pros
Connects to over a dozen LLM providers including OpenAI, Anthropic, Google, and local options via Ollama, as listed in the README, making it a one-stop shop for diverse AI tasks.
Features like Prompt Anywhere and AI Commands allow generating and transforming text directly in any application using global shortcuts, enabling seamless workflow integration.
Provides a local HTTP server, CLI, and agent webhooks for programmatic access, as detailed in the README, allowing automation and integration with external systems.
Users retain full control of their API keys and data, with no data sent to Witsy's servers, ensuring privacy and cost transparency as emphasized in the philosophy.
Cons
Requires obtaining and configuring API keys from multiple providers and setting up local tools like Ollama, which the README admits involves significant prerequisites and can be time-consuming.
As a desktop application, it lacks mobile or web access, restricting usage to devices where it's installed and running, unlike cloud-based alternatives.
Missing features for team workspaces or shared conversations, as indicated by the TODO list mentioning 'Workspaces / Projects' as pending, limiting its use for collaborative environments.
Most advanced features depend on cloud API keys, which incur costs and require internet connectivity, with limited offline functionality beyond basic local models.
Frequently Asked Questions
Related Projects
Atomic ChatLocal AI app and inference engine for agents. Run open-weight LLMs locally — private, 100% offline on your computer. Join our Discord: https://discord.com/invite/8wGSsvmg4V
Apple On-Device OpenAIOpenAI-compatible API server for Apple on-device models
Usage4ClaudeMonitor all your Claude/Codex usage limits in real-time from your macOS menu bar - supports 5-hour, 7-day, extra usage, 7-day Opus and 7-day Sonnet quotas.
Claudoscopenative macOS app that gives you a real-time dashboard for your Claude Code and Cowork sessions, with analytics, conversation history, security hardening, real time secrets detection and project insights.
Found a gem we're missing?
Open-Awesome is built by the community, for the community. Submit a project, suggest an awesome list, or help improve the catalog on GitHub.