Wuwei banner
wuwei-io wuwei-io

Wuwei

AI community

Description

Free, open-source, local-first AI agent desktop app — a Claude Code / Cursor / Copilot alternative. Reads & writes files, edits code, runs commands. No setup, no API key needed. Windows, macOS, Linux.

Installation

This entry records only its repository, not the path inside it, so there is no exact command to give. Open the source below and copy the folder into ~/.claude/skills/, or the file into ~/.claude/agents/.

README

Wuwei AI

**A free, open-source, local-first AI agent client**

Tell an AI what you want in one sentence: read/write files, precise edits, run commands, search the web — every step behind a permission prompt. Switch between Claude / OpenAI / Chinese LLMs in one click, bring your own key or use Wuwei's hosted credits.

[Website](https://wuweiai.io) · [Download](https://wuweiai.io) · [简体中文](./README.zh-CN.md)

![license](https://img.shields.io/badge/license-MIT-green) ![platform](https://img.shields.io/badge/platform-Windows%20%7C%20macOS%20%7C%20Linux-blue)


Wuwei AI demo: one prompt → read file → permission prompt → edit code → done

Live demo · click to watch the HD mp4


What is this

Wuwei is an **AI agent** — not autocomplete, not a chat box. At its core: a **large language model + a tool-execution loop + an interface** (terminal TUI / desktop GUI).

You state a goal and it reads files, edits code, runs commands, and searches the web to finish the task from start to end — while the harness is fully self-built, open source and auditable, with your code and data staying on your own machine.

  • 🆓 Genuinely free — works out of the box, no subscription, no credit card
  • 💻 Local-first — your project and code stay on your machine, never uploaded somewhere you can't see
  • 🔍 Open source (MIT) — auditable line by line, no hidden telemetry, no region tagging
  • 🔌 Bring any model — Claude / any OpenAI-compatible endpoint / local models (vLLM, Ollama) / Chinese LLMs, one-click switch
  • 🌏 Direct access in China — no VPN needed for Chinese models
  • 🛡️ Permission prompts — confirm before writing files or running commands (y / N / a = always allow this tool)