# ZhikunCode — Self-Hosted Multi-Agent Coding Suite

> ZhikunCode is a self-hosted AI coding suite with browser-first control and Docker deployment; README notes the first build can take 15–30 minutes.

## Install

Copy the content below into your project:

## Quick Use

```bash
git clone https://github.com/zhikunqingtao/zhikuncode.git
cd zhikuncode
cp .env.example .env
# edit .env and set your LLM API key
docker compose up -d
# open http://localhost:8080
```

## Intro

ZhikunCode is a self-hosted AI coding suite with browser-first control and Docker deployment; README notes the first build can take 15–30 minutes.

**Best for:** teams who want a private, browser-driven coding agent stack

**Works with:** Docker 20.10+, Docker Compose v2, browser UI (localhost:8080), multiple LLM providers (per README)

**Setup time:** 15-30 minutes

### Key facts (verified)

- GitHub: 229 stars · 68 forks · pushed 2026-05-13.
- License: MIT · owner avatar + repo URL verified via GitHub API.
- README-verified entrypoint: `pip install -r requirements.txt`.

## Main

- Use the README’s 3-step Docker path: clone, set `.env`, then `docker compose up -d` and access the UI in the browser.

- Plan for the first build: README calls out a 15–30 minute initial image build (later starts are much faster).

- When enabling optional MCP services, follow the README’s guidance to toggle capability registry entries rather than guessing tool IDs.

### Source-backed notes

- README provides the exact Docker quick start commands, including `cp .env.example .env` and `docker compose up -d`.
- README states the first build may take 15–30 minutes depending on network speed, and the UI is served at `http://localhost:8080`.
- README describes multiple collaboration modes (Team/Swarm/SubAgent) and a browser-first workflow.

### FAQ

- **Do I need Docker to start?**: README recommends Docker Compose as the easiest path; local dev is also described.
- **Where do I use it?**: In a browser at `http://localhost:8080` after startup per README.
- **Why does first start take longer?**: README notes the first run builds Docker images (15–30 minutes).

## Source & Thanks

> Source: https://github.com/zhikunqingtao/zhikuncode
> License: MIT
> GitHub stars: 229 · forks: 68

---

<!-- ZH -->

## Quick Use

```bash
git clone https://github.com/zhikunqingtao/zhikuncode.git
cd zhikuncode
cp .env.example .env
# edit .env and set your LLM API key
docker compose up -d
# open http://localhost:8080
```

## Intro

ZhikunCode 是自托管 AI 编程套件，主打浏览器全流程操控与 Docker 一键部署，并支持多 Agent 协作；README 提示首次构建约需 15–30 分钟，启动后访问 http://localhost:8080。

**Best for:** 想要私有化部署、用浏览器统一管控的团队

**Works with:** Docker 20.10+、Docker Compose v2；浏览器 UI（localhost:8080）；多模型提供方（README 说明）

**Setup time:** 15-30 minutes

### Key facts (verified)

- GitHub：229 stars · 68 forks；最近更新 2026-05-13。
- 许可证：MIT；作者头像与仓库链接均已通过 GitHub API 复核。
- README 中核对过的入口命令：`pip install -r requirements.txt`。

## Main

- 按 README 的 3 步 Docker 路径：clone → 配置 `.env` → `docker compose up -d`，然后用浏览器访问 UI。

- 给首次构建留时间：README 说明第一次需要 15–30 分钟构建镜像，后续启动会快很多。

- 如需启用可选 MCP 服务，按 README 指引在能力注册表里切换 enabled，而不是凭空猜工具 ID。

### Source-backed notes

- README 给出了 Docker 快速开始命令，包括 `cp .env.example .env` 与 `docker compose up -d`。
- README 提示首次构建可能需要 15–30 分钟，启动后可访问 `http://localhost:8080`。
- README 描述了 Team/Swarm/SubAgent 多种协作模式与浏览器优先的工作流。

### FAQ

- **一定要用 Docker 吗？**：README 推荐 Docker Compose 最省事，同时也提供本地开发方式。
- **启动后从哪里使用？**：按 README 访问 `http://localhost:8080` 的浏览器界面。
- **为什么第一次启动很慢？**：README 说明首次运行会构建镜像，约需 15–30 分钟。

## Source & Thanks

> Source: https://github.com/zhikunqingtao/zhikuncode
> License: MIT
> GitHub stars: 229 · forks: 68


---
Source: https://tokrepo.com/en/workflows/zhikuncode-self-hosted-multi-agent-coding-suite
Author: AI Open Source