rowser-based AI coding assistant — zero install, zero deploy, zero config. Agentic loop with 10 tools, 18 skills, and a force-directed Cosmos visualization. Runs entirely client-side.
# JetBot
**Browser-based AI Coding Assistant — Zero Install. Zero Deploy. Zero Config.**
[中文文档](README.zh-CN.md)
JetBot is a fully browser-based AI coding assistant. No backend server, no deployment pipeline, no installation required — just open the page and start working. All code executes client-side via JavaScript; LLM calls go directly from the browser's `fetch()` to your chosen API provider.
```
┌─────────────────────────────────────────────┐
│ Browser │
│ │
│ ┌──────────┐ ┌───────────┐ ┌───────────┐ │
│ │ React │ │ Agentic │ │ Virtual │ │
│ │ UI │←→│ Loop │←→│ FS │ │
│ └──────────┘ └─────┬─────┘ │ (IndexedDB)│ │
│ │ └───────────┘ │
│ │ fetch() │
└──────────────────────┼──────────────────────┘
↓
┌────────────────┐
│ LLM Provider │
│ OpenAI/DeepSeek│
│ Ollama/Custom │
└────────────────┘
```
## Quick Start
```bash
git clone https://github.com/hongxin/JetBot.git
cd JetBot/jetbot
npm install
npm run dev
```
Open `http://localhost:5173`, select an LLM provider, enter your API key, and start chatting.
**Local Ollama (fully offline):**
```bash
ollama serve
ollama pull qwen3.5:27b
# Select the Ollama provider in JetBot — no API key needed
```
**Production build:**
```bash
npm run build
# Outputs dist/ — pure static files deployable to any static host
```
## Features
### Agentic Loop
An autonomous tool-calling loop where the LLM selects tools, executes actions, and analyzes results until the task is complete. Supports up to 100 iterations with circuit breaker protection (3 consecutive failures) and automatic duplicate-error detection.
### 10 Built-in Tools
| Tool | Permission | Description |
|------|-----------|-------------|
| `read_file` | safe | Read files from the virtual file HAL 分层混合模型工作流 — 强模型(Claude)负责理解/拆解/验收,低成本模型(DeepSeek)负责检索/提取/清洗。Hermes Agent skill。
An LLM agent fine-tuned on DeepSeek for spaced repetition, dynamically integrating knowledge points based on the Ebbinghaus forgetting curve.
基于 STM32F103 构建的端到端 AI 智能手表生态。自研“零重定位”原生机器码动态加载引擎与页面栈式 UI 框架;集成生产级 OTA 回滚保护机制与高带宽(921600 baud)串口协议栈。通过 Node.js 中继实现 DeepSeek AI 语义控制及 ASRPRO 语音全双工交互,是一个集成了分布式计算、现代存储管理与 AI Agent 的嵌入式全栈工程。
A Meta-Agent-Driven Self-Evolving Multi-Agent System for UAV Detection and Tracking
One command to run Hermes AI Agent with a browser UI. Zero prerequisites. 一行命令,AI 就位。
网页应用Agent,接入DeepSeek、Mimo等模型