MCP Servers · Kochava-Studios

witsy

Witsy is a desktop AI assistant (Electron app) that acts as a universal MCP (Model Context Protocol) client, enabling you to use virtually any LLM provider (OpenAI, Anthropic, Gemini, Ollama, etc.) with a unified interface. It supports chat, image/video generation, speech-to-text, RAG over local documents, and system-wide AI commands via keyboard shortcuts.

Talk to DEV.co Discuss Implementation

Source: GitHub — github.com/Kochava-Studios/witsy

GitHub stars

163

Forks

TypeScript

Primary language

AGPL-3.0

License (OSI-approved)

Key facts

Objective fields from the source. Values we can't verify are shown as “Unknown” rather than guessed.

Field	Value
Repository	Kochava-Studios/witsy
Owner	Kochava-Studios
Primary language	TypeScript
License	AGPL-3.0 — OSI-approved
Stars	2k
Forks	163
Open issues	54
Latest release	v3.5.2 (2026-03-04)
Last updated	2026-04-23
Source	https://github.com/Kochava-Studios/witsy

What witsy is

TypeScript/Vue.js 3 Electron application providing a local HTTP API (port 8090) and MCP server support. Supports 15+ LLM providers via API keys or local Ollama; includes vision models, embeddings, document RAG via OpenAI/Ollama embeddings, and global prompt/command shortcuts. Runs on localhost by default; requires Node.js for building.

Quickstart

Get the witsy source

Clone the repository and explore it locally.

terminalbash

1git clone https://github.com/Kochava-Studios/witsy.git2cd witsy3# follow the project's README for install & configuration

Need it deployed, integrated, or customized instead? DEV.co ships production installs.

Best use cases

Multi-provider LLM experimentation

Quickly test and compare responses from OpenAI, Anthropic, Gemini, Groq, DeepSeek, and others in one interface without switching apps. Ideal for teams evaluating models or building prompts across providers.

Local-first AI workflows with Ollama

Run open-source models (Llama, Mistral) locally on your machine with zero API costs. Useful for privacy-sensitive work or organizations with restricted cloud API usage.

System-wide productivity automation

Use global shortcuts (Shift+Ctrl+Space, Alt+Ctrl+Space) to generate content, run commands, or transcribe directly within any desktop application without copy-paste friction.

Implementation considerations

AGPL-3.0 license requires source code disclosure if you modify and redistribute; commercial internal use is permissible but consult legal if bundling or redistributing.
API key storage is local; audit local file permissions and consider encrypting keys at rest for sensitive environments.
Ollama integration requires separate installation and model downloads (~GB per model); plan disk and bandwidth for initial setup.
HTTP API (localhost:8090) is localhost-only by default; reverse proxy with auth required for remote access—no built-in authentication.
MCP server support is vendor-specific (Anthropic MCP); not all third-party MCP implementations are guaranteed compatible; test early.

When to avoid it — and what to weigh

Web-based collaborative workflows required — Witsy is desktop-only (Electron). If your team needs browser-based, shared access or real-time collaboration, this is not a fit.
Regulated data governance — Local conversation history and per-provider API keys create audit complexity. Regulated environments (healthcare, finance) need formal security/compliance review before use.
No BYOK budget or API key management at scale — Witsy requires external API keys from multiple providers. If your organization cannot manage LLM spend or prefers unified billing, this adds operational overhead.
Requires native installers only — Building from source requires npm/Node.js. If your deployment model strictly forbids dev toolchain, pre-built binaries mitigate this, but custom builds are not simple.

License & commercial use

Licensed under AGPL-3.0 (GNU Affero General Public License v3.0). This is a copyleft license: internal use is permitted; if you modify and deploy it as a service or redistribute, you must disclose source code and license derivative works under AGPL-3.0. Commercial use is permissible under these terms.

Commercial use is permitted under AGPL-3.0 for internal deployment. If you modify Witsy and offer it as a service or resell it, you must provide source code and comply with AGPL-3.0 requirements. Requires legal review if bundling or modifying for external redistribution. Standard commercial licensing terms are not published; contact maintainers for clarification.

DEV.co evaluation signals

Editorial assessment — not user reviews. Directional, with an explicit confidence level.

Signal	Assessment
Maintenance	Active
Documentation	Adequate
License clarity	Clear
Deployment complexity	Low
DEV.co fit	Good
Assessment confidence	High

Security considerations

Local HTTP API runs on localhost only (mitigates network exposure); remote access requires reverse proxy—no built-in auth. API keys stored locally; no encryption mentioned (audit local file permissions). Electron app surface includes global hotkeys and file access (RAG); OS permissions should be reviewed. No formal security audit, CVE history, or security policy documented. Data is sent to external providers (OpenAI, Anthropic, etc.); review provider data retention policies if handling sensitive content.

Alternatives to consider

OpenAI ChatGPT Plus / Web Interface

Browser-based, zero setup, but locked to OpenAI models; no MCP, no local models, no multi-provider parity.

Claude.ai (Anthropic)

Browser-based, polished UX, but Anthropic-only; no multi-provider support, no MCP, no local-first option.

LM Studio (local-first GUI)

Desktop app for local Ollama/GGUF models; no cloud provider support, no MCP, no system-wide commands, simpler but more limited.

Software development agency

Build on witsy with DEV.co software developers

Download Witsy from GitHub releases or brew install. Bring your own API keys or run models locally with Ollama. Evaluate deployment fit with your legal and security teams given AGPL-3.0 terms.

Talk to DEV.co

Related open-source tools

Surfaced by semantic similarity across the DEV.co open-source index.

deepchat scira-mcp-chat python-utcp

Related on DEV.co

Explore the category and the services that help you build with it.

MCP Servers AI Development Custom Software Development Talk to DEV.co

witsy FAQ

Can I run Witsy without an API key?

Yes. Install Ollama locally, download open-source models (Llama, Mistral), and use them in Witsy for free. Cloud providers (OpenAI, Anthropic) require API keys and incur costs.

Is my data private?

Conversations are stored locally on your machine. Prompts sent to cloud providers (OpenAI, Anthropic, etc.) are subject to their data retention policies; review them for sensitive use cases. Local Ollama models do not phone home.

Can I use Witsy in a company environment?

For internal use, yes. AGPL-3.0 permits commercial use. If you modify it or offer it as a service, you must disclose source code. Consult legal before deployment in regulated industries.

What is MCP and why does it matter?

Model Context Protocol (Anthropic standard) allows LLMs to call external tools/APIs. Witsy is rare in supporting MCP with any LLM, not just Anthropic. This enables richer automation (e.g., execute code, access databases).

Software development & web development with DEV.co

Adopting witsy is usually one piece of a larger software development effort. As a software development agency, DEV.co provides software development services and web development expertise — pairing senior software developers and web developers with your team to design, build, and operate mcp servers software in production.

Ready to unify your LLM workflow?

Download Witsy from GitHub releases or brew install. Bring your own API keys or run models locally with Ollama. Evaluate deployment fit with your legal and security teams given AGPL-3.0 terms.

Talk to DEV.co Schedule a Consultation