# GenericAgent

**Repository Path**: cfh008/GenericAgent

## Basic Information

- **Project Name**: GenericAgent
- **Description**: No description available
- **Primary Language**: Unknown
- **License**: MIT
- **Default Branch**: main
- **Homepage**: None
- **GVP Project**: No

## Statistics

- **Stars**: 0
- **Forks**: 1
- **Created**: 2026-04-23
- **Last Updated**: 2026-06-22

## Categories & Tags

**Categories**: Uncategorized

**Tags**: None

## README

<div align="center">

<img src="assets/images/bar.jpg" width="880" alt="GenericAgent Banner"/>

# GenericAgent

**A Minimal, Self-Evolving Autonomous Agent Framework**

*~3K lines of seed code · 9 atomic tools · ~100-line Agent Loop*

<p>

  <a href="https://gaagent.ai"><img src="https://img.shields.io/badge/Official_Website-gaagent.ai-00A67E?style=flat-square" alt="Official Website"/></a>
  <a href="https://arxiv.org/abs/2604.17091"><img src="https://img.shields.io/badge/Technical_Report-PDF-EA4335?style=flat-square&logo=adobeacrobatreader&logoColor=white" alt="Technical Report"/></a>
  <a href="https://github.com/JinyiHan99/GA-Technical-Report"><img src="https://img.shields.io/badge/Code_%26_Data-Reproduction-181717?style=flat-square&logo=github" alt="Reproduction Repo"/></a>
  <a href="https://datawhalechina.github.io/hello-generic-agent/"><img src="https://img.shields.io/badge/Tutorial-Datawhale-blue?style=flat-square" alt="Tutorial"/></a>
  <a href="https://fudankw.cn/sophub"><img src="https://img.shields.io/badge/Skill_Hub-Sophub-purple?style=flat-square" alt="Sophub"/></a>
</p>

<p>
  <a href="https://trendshift.io/repositories/25944" target="_blank"><img src="https://trendshift.io/api/badge/repositories/25944" alt="Trendshift" style="width: 250px; height: 55px;" width="250" height="55"/></a>
</p>

**[English](#-english) · [中文](#-中文)**

</div>

> 📌 **Official:** GitHub + https://gaagent.ai only. DintalClaw is the sole authorized commercial partner; others are not affiliated.

---

<a id="-english"></a>

## 🌟 Overview

**GenericAgent** is a minimal, self-evolving autonomous agent framework. Its core is just **~3K lines of code**. Through **9 atomic tools + a ~100-line Agent Loop**, it grants any LLM system-level control over a local computer — covering browser, terminal, filesystem, keyboard/mouse input, screen vision, and mobile devices (ADB).

> Design philosophy — **don't preload skills, evolve them.**

Every time GenericAgent solves a new task, it automatically crystallizes the execution path into a reusable **Skill**. The longer you use it, the more skills accumulate — forming a personal skill tree grown entirely from 3K lines of seed code.

> 🤖 **Self-Bootstrap Proof** — Everything in this repository, from installing Git and running `git init` to every commit message, was completed autonomously by GenericAgent. The author never opened a terminal once.

### 📑 Table of Contents

- [Key Features](#-key-features)
- [Demo Showcase](#-demo-showcase)
- [Quick Start](#-quick-start)
- [Usage](#-usage)
- [Unlocking Advanced Capabilities](#-unlocking-advanced-capabilities)
- [Architecture](#-architecture)
- [Self-Evolution Mechanism](#-self-evolution-mechanism)
- [Comparison](#-comparison)
- [Evaluation](#-evaluation)
- [Roadmap & News](#-roadmap--news)
- [Community & Support](#-community--support)
- [License](#-license)

---

## 📋 Key Features

| Feature | Description |
| :--- | :--- |
| 🧬 **Self-Evolving** | Automatically crystallizes each task into a Skill. Capabilities grow with every use, forming your personal skill tree. |
| 🪶 **Minimal Architecture** | ~3K lines of core code. Agent Loop is ~100 lines. No complex dependencies, zero deployment overhead. |
| ⚡ **Strong Execution** | **TMWebdriver** injects into a real browser (preserving login sessions). 9 atomic tools take direct control of the system. |
| 🔌 **High Compatibility** | Supports Claude / Gemini / Kimi / MiniMax and other major models. Cross-platform. |
| 💰 **Token Efficient** | <30K context window — a fraction of the 200K–1M other agents consume. Less noise, fewer hallucinations, higher success rate, lower cost. |

---

## 🎯 Demo Showcase

<table>
  <tr>
    <td align="center" width="50%"><b>🛡️ Real-Browser CAPTCHA Survival</b></td>
    <td align="center" width="50%"><b>🌐 Autonomous Web Exploration</b></td>
  </tr>
  <tr>
    <td><img src="assets/demo/discord_hcaptcha_real_browser.gif" width="100%" alt="Discord hCaptcha passed in real browser"></td>
    <td><img src="assets/demo/autonomous_explore.png" width="100%" alt="Web Exploration"></td>
  </tr>
  <tr>
    <td><sub>While configuring a Discord bot, an hCaptcha <i>"Are you human?"</i> challenge pops up mid-task — GA's real browser session passes it and the task continues. See <a href="#browser-realness-of-ga-web-tools">Browser Realness</a>.</sub></td>
    <td><sub>Autonomously browses and periodically summarizes web content.</sub></td>
  </tr>
  <tr>
    <td align="center"><b>🧋 Food Delivery Order</b></td>
    <td align="center"><b>📈 Quantitative Stock Screening</b></td>
  </tr>
  <tr>
    <td><img src="assets/demo/order_tea.gif" width="100%" alt="Order Tea"></td>
    <td><img src="assets/demo/selectstock.gif" width="100%" alt="Stock Selection"></td>
  </tr>
  <tr>
    <td><sub><i>"Order me a milk tea"</i> — navigates the delivery app, selects items, completes checkout.</sub></td>
    <td><sub><i>"Find GEM stocks with EXPMA golden cross, turnover &gt; 5%"</i> — quantitative screening.</sub></td>
  </tr>
  <tr>
    <td align="center"><b>💰 Expense Tracking</b></td>
    <td align="center"><b>💬 Batch Messaging</b></td>
  </tr>
  <tr>
    <td><img src="assets/demo/alipay_expense.png" width="100%" alt="Alipay Expense"></td>
    <td align="center"><img src="assets/demo/wechat_batch.png" width="65%" alt="WeChat Batch"></td>
  </tr>
  <tr>
    <td><sub><i>"Find expenses over ¥2K in the last 3 months"</i> — drives Alipay via ADB.</sub></td>
    <td><sub>Sends bulk WeChat messages, fully driving the WeChat client.</sub></td>
  </tr>
</table>

---

## 🚀 Quick Start

> ⚠️ **Python version**: use **Python 3.11 or 3.12**. **Do not** use Python 3.14 — it is incompatible with `pywebview` and a few other GA dependencies.
>
> 📖 Detailed installation guide: **[installation.md](docs/installation.md)** · **[installation_zh.md（中文）](docs/installation_zh.md)**

### For LLM Agents

Fetch the installation guide and follow it:

```bash
curl -fsSL https://raw.githubusercontent.com/lsdefine/GenericAgent/refs/heads/main/docs/installation.md
```

### For Humans

#### Method 1 — Clone & install *(recommended)*

```bash
git clone https://github.com/lsdefine/GenericAgent.git && cd GenericAgent
uv venv && uv pip install -e ".[ui]"
cp mykey_template_en.py mykey.py   # fill in your LLM API key
```

Dependencies are deliberately tiered: the agent core needs only `requests`, plus four lightweight packages (`beautifulsoup4`, `bottle`, `simple-websocket-server`, `aiohttp`) for TMWebdriver's local server. The `[ui]` extra pulls in frontend libraries (Streamlit, `prompt_toolkit`/`rich` for the TUI, …) — install it for the bundled UIs, or skip it entirely and drive the agent headless. No Playwright, no LangChain, no browser binaries to download.

Then launch:

```bash
python frontends/tui_v3.py   # Terminal UI (recommended)
python launch.pyw            # Streamlit web UI
```

#### Method 2 — One-line installer *(convenience)*

Sets up a self-contained directory with an isolated Python environment, Git, and a ready-to-run package. The script is in [`assets/`](assets/) if you'd like to read it first.

**Windows PowerShell**

```powershell
powershell -ExecutionPolicy Bypass -c "$env:GLOBAL=1; irm https://raw.githubusercontent.com/lsdefine/GenericAgent/main/assets/ga_install.ps1 | iex"
```

**Linux / macOS**

```bash
GLOBAL=1 bash -c "$(curl -fsSL https://raw.githubusercontent.com/lsdefine/GenericAgent/main/assets/ga_install.sh)"
```

> 💡 GenericAgent grows its environment **through the Agent itself** — don't pre-install everything. See [Unlocking Advanced Capabilities](#-unlocking-advanced-capabilities) below.

---

## 💻 Usage

### Frontends

#### Terminal UI *(recommended)*

A lightweight, scrollback-first terminal interface built on `prompt_toolkit` + `rich`. Supports multiple concurrent sessions and real-time streaming.

```bash
python frontends/tui_v3.py
```

<details>
<summary><b>⚠️ Windows TUI Troubleshooting</b></summary>

TUI rendering on Windows can be flaky depending on terminal + font. Common causes:

1. `prompt_toolkit` / `rich` are not on the latest version — `pip install -U prompt_toolkit rich` first.
2. PowerShell / cmd ship with terminals that have rough Unicode + key-binding support. **Prefer Git Bash on Windows**, which is much better behaved.
3. If it still looks broken, ask GA itself to fix it:
   > *"My experience using `frontends/tui_v3.py` in PowerShell / cmd / Git Bash on Windows is very poor — lots of incompatibility. Please refer to Claude Code's best practices for the Windows terminal and fix all font and rendering incompatibilities."*

</details>

#### Streamlit UI

```bash
python launch.pyw
```

### Bot Interface (IM)

GenericAgent also supports IM frontends such as Telegram, Discord, and Lark.

| Platform | Command |
| :--- | :--- |
| Telegram | `python frontends/tgapp.py` |
| Discord | `python frontends/dcapp.py` |
| Lark / Feishu | `python frontends/fsapp.py` |

> WeChat, QQ, WeCom and DingTalk are also supported — see the Chinese section below.
> For detailed setup, ask GenericAgent itself.

---

## 🔓 Unlocking Advanced Capabilities

In GA, advanced capabilities are unlocked by **instructing the agent**, not by reading
docs or installing extras. Each instruction below makes GA read its pre-installed SOPs
(battle-tested playbooks in its memory), install whatever is missing, adapt to your OS,
and persist the result into its own memory.

| Capability | Just tell GA |
| :--- | :--- |
| 🌐 Web automation | *"Set up your web automation capability."* — GA guides you through the one manual step: dragging the bundled Chrome extension into `chrome://extensions`. |
| 🔤 OCR | *"Set up your OCR capability with rapidocr and save it to memory."* |
| 👁️ Vision | *"Set up your vision capability from the template in memory/."* — GA copies the template, wires it to your existing LLM keys, and self-tests. |
| 🖱️ Computer use | *"Probe this system and set up your computer-use capability."* |

> 💡 **About language**: the pre-installed SOPs are written in Chinese — GA reads them
> natively, so this never blocks you. If you prefer an English knowledge base, just say:
> *"Read your pre-installed SOPs and rewrite them in English (keep code, paths and error
> strings verbatim)."*
>
> 🌍 **About platforms**: the SOPs were honed on Windows, but cross-platform adaptation is
> itself a GA task — on macOS/Linux, GA swaps in the platform equivalents (window
> enumeration, input control, screenshots) on its own. Same self-evolution principle.

---

## 🧠 Architecture

GenericAgent accomplishes complex tasks through **Layered Memory × Minimal Toolset × Autonomous Execution Loop**, continuously accumulating experience during execution.

### 1️⃣ Layered Memory System

> *Memory crystallizes throughout task execution, letting the agent build stable, efficient working patterns over time.*

| Layer | Name | Description |
| :---: | :--- | :--- |
| **L0** | Meta Rules | Core behavioral rules and system constraints |
| **L1** | Insight Index | Minimal memory index for fast routing and recall |
| **L2** | Global Facts | Stable knowledge accumulated over long-term operation |
| **L3** | Task Skills / SOPs | Reusable workflows for completing specific task types |
| **L4** | Session Archive | Archived task records distilled from finished sessions for long-horizon recall |

### 2️⃣ Autonomous Execution Loop

> *Perceive environment state → Task reasoning → Execute tools → Write experience to memory → Loop*

The entire core loop is just **~100 lines of code** ([`agent_loop.py`](agent_loop.py)).

### 3️⃣ Minimal Toolset

> *GenericAgent provides only **9 atomic tools**, forming the foundational capabilities for interacting with the outside world.*

| Tool | Function |
| :--- | :--- |
| `code_run` | Execute arbitrary code (Python / PowerShell) |
| `file_read` | Read files |
| `file_write` | Write / create / overwrite files |
| `file_patch` | Patch / modify files |
| `web_scan` | Perceive web content |
| `web_execute_js` | Control browser behavior |
| `ask_user` | Human-in-the-loop confirmation |
| `update_working_checkpoint` | *(memory)* Short-term working notepad |
| `start_long_term_update` | *(memory)* Distill long-term memory |

### 4️⃣ Capability Extension

> *Capable of dynamically creating new tools.*

Via `code_run`, GenericAgent can dynamically install Python packages, write new scripts, call external APIs, or control hardware at runtime — crystallizing temporary abilities into permanent tools.

<div align="center">
  <img src="assets/images/workflow.jpg" alt="GenericAgent Workflow" width="420"/>
  <br/><em>GenericAgent Workflow Diagram</em>
</div>

---

## 🧬 Self-Evolution Mechanism

This is what fundamentally distinguishes GenericAgent from every other agent framework.

```text
[New Task]
   │
   ▼
[Autonomous Exploration]   ─►  install deps · write scripts · debug · verify
   │
   ▼
[Crystallize into Skill]   ─►  write to memory layer
   │
   ▼
[Direct Recall on Next Similar Task]
```

| What you say | First time | Every time after |
| :--- | :--- | :--- |
| *"Read my WeChat messages"* | Install deps → reverse DB → write read script → save Skill | **one-line invoke** |
| *"Give me a morning digest of Hacker News"* | Write scraper → build digest → schedule daily run → save Skill | **one-line invoke** |
| *"Monitor stocks and alert me"* | Install `mootdx` → build selection flow → configure cron → save Skill | **one-line start** |
| *"Send this file via Gmail"* | Configure OAuth → write send script → save Skill | **ready to use** |

After a few weeks, your agent instance will have a skill tree no one else in the world has — all grown from 3K lines of seed code.

---

## 📊 Comparison

| Feature | **GenericAgent** | OpenClaw | Claude Code |
| :--- | :---: | :---: | :---: |
| **Codebase** | ~3K lines | ~530,000 lines | Open-sourced (large) |
| **Deployment** | `pip install` + API Key | Multi-service orchestration | CLI + subscription |
| **Browser Control** | Real browser (session preserved) | Sandbox / headless browser | Via MCP plugin |
| **OS Control** | Mouse/kbd, vision, ADB | Multi-agent delegation | File + terminal |
| **Self-Evolution** | Autonomous skill growth | Plugin ecosystem | Stateless between sessions |
| **Out of the Box** | Few core files + starter skills | Hundreds of modules | Rich CLI toolset |

---

## 📈 Evaluation

> 📂 Full evaluation datasets and results: [**JinyiHan99/GA-Technical-Report**](https://github.com/JinyiHan99/GA-Technical-Report/tree/main)

We evaluate GenericAgent across **five dimensions**:

| # | Dimension | Question | Benchmarks |
| :---: | :--- | :--- | :--- |
| 1 | **Task Completion & Token Efficiency** | Can GA complete hard tasks more cheaply than leading agents? | SOP-Bench, Lifelong AgentBench, RealFin-Benchmark |
| 2 | **Tool-Use Efficiency** | Can a minimal atomic toolset solve what specialized toolsets solve, with less overhead? | Tool Efficiency Benchmark (11 simple + 5 long-horizon) |
| 3 | **Memory System Effectiveness** | Does condensed hierarchical memory beat full/redundant memory and embedding-based retrievers? | SOP-Bench (dangerous goods), LoCoMo, 20-skill stress test |
| 4 | **Self-Evolution Capability** | Can the agent distill experience into reusable SOPs and code, without intervention? | 9-round LangChain longitudinal study, 8-task cross-task web benchmark |
| 5 | **Web Browsing Capability** | Does density-driven design survive the open web? | WebCanvas, BrowseComp-ZH, Custom Tasks (22) |

Baselines across these dimensions include **Claude Code**, **OpenAI CodeX**, and **OpenClaw**, evaluated under *Claude Sonnet 4.6*, *Claude Opus 4.6*, *GPT-5.4*, and *MiniMax M2.7* backbones.

<table>
  <tr>
    <td align="center" width="50%">
      <img src="assets/images/result_radar.png" width="100%" alt="Tool-use efficiency radar"/><br/>
      <sub><b>Tool-use efficiency radar.</b> GA dominates token, request, and tool-call axes while preserving quality across four task dimensions.</sub>
    </td>
    <td align="center" width="50%">
      <img src="assets/images/result_convergence.png" width="100%" alt="Cross-task self-evolution convergence"/><br/>
      <sub><b>Cross-task self-evolution.</b> Second- and third-run GA executions converge to a stable low-cost regime across eight web tasks, while OpenClaw shows no such convergence.</sub>
    </td>
  </tr>
</table>

### Browser Realness of GA Web Tools (TMWebdriver)

GA web tools are powered by **TMWebdriver** — a local WebSocket server plus a Chrome extension — running through a **real, persistent Chrome/Chromium session** rather than a disposable headless sandbox, preserving cookies, login state, extensions, GPU/WebGL behavior, and normal browser-session fingerprints.

| Detection Service / Signal | Vanilla Headless Automation | GA Web Tools | Notes |
| :--- | :---: | :---: | :--- |
| SannySoft headless test | Often detected | ✅ 56/56 passed | `bot.sannysoft.com` |
| bot.incolumitas.com | Commonly fails webdriver / CDP checks | ✅ 36/36 passed | `WEBDRIVER`, `SELENIUM_DRIVER`, `webDriverAdvanced` all OK |
| BrowserScan bot detection | Often abnormal | ✅ Normal | `browserscan.net` |
| Device & Browser Info bot test | Multiple bot flags | ✅ Human / `isBot=false` | `deviceandbrowserinfo.com` |
| FingerprintJS bot detection demo | Often detected | ✅ Passed | Demo flow completed without bot verdict |
| reCAPTCHA v3 demo | Low bot-like score | ✅ 0.9 human-like score | Score-based risk signal; 0.9 is above typical production thresholds |

For reCAPTCHA v3, `0.9` is not a "checkbox solved" result; it is the high-confidence human-like score returned by the risk model, typically sufficient to avoid extra challenges in production flows.

---

## 📅 Roadmap & News

- **2026-05-23** — 🆕 **TUI v3 released** (`frontends/tui_v3.py`). Block-based scrollback with proper resize reflow, per-terminal color profile for cross-terminal parity, and feature parity with v2.
- **2026-05-18** — 🆕 **Morphling mode**. Project-level skill absorption — extract goal + tests from any external repo, then decide per component: call, rewrite, or discard. See `memory/morphling_sop.md`.
- **2026-05-17** — 🆕 **Goal Hive mode**. Multi-worker cooperative Goal mode — BBS-coordinated master/workers running long-horizon objectives in parallel. See `memory/goal_hive_sop.md`.
- **2026-05-15** — 🖥️ **Desktop GUI released**. One-line installs ship a ready-to-run desktop app (`frontends/GenericAgent.exe`). Developers launch via `python launch.pyw`.
- **2026-05-14** — 🆕 **Conductor sub-agent orchestration**. Spawn, supervise, and auto-clean parallel sub-agents; first-class delegation primitives complementing `/btw` side-questions.
- **2026-05-12** — 🆕 **TUI v2 released** (`frontends/tuiapp_v2.py`). Refined Textual frontend with image-paste folding, file paste, block-delete, Ctrl+C copy, history navigation, and `/llm` / `/export` / `/continue` pickers.
- **2026-05-08** — 🆕 **Goal mode** (`reflect/goal_mode.py`). Time-budget-driven self-driven loop — "keep optimizing X for N hours" with no premature delivery.
- **2026-04-21** — 📄 [**Technical Report on arXiv**](https://arxiv.org/abs/2604.17091) — *GenericAgent: A Token-Efficient Self-Evolving LLM Agent via Contextual Information Density Maximization*.
- **2026-04-11** — Introduced **L4 session archive memory** and scheduler cron integration.
- **2026-03-23** — Personal WeChat supported as a bot frontend.
- **2026-03-10** — [Released million-scale Skill Library](https://mp.weixin.qq.com/s/q2gQ7YvWoiAcwxzaiwpuiQ?scene=1&click_id=7) *(Chinese)*.
- **2026-03-08** — [Released "Dintal Claw" — a GenericAgent-powered government-affairs bot](https://mp.weixin.qq.com/s/eiEhwo-j6S-WpLxgBnNxBg) *(Chinese)*.
- **2026-03-01** — [Featured by Jiqizhixin (机器之心)](https://mp.weixin.qq.com/s/uVWpTTF5I1yzAENV_qm7yg) *(Chinese)*.
- **2026-01-16** — GenericAgent **V1.0** public release.

---

## ⭐ Community & Support

If this project helped you, please consider leaving a **Star!** 🙏

### 🚩 Friendly Links

Thanks to the **LinuxDo** community for the support!

[![LinuxDo](https://img.shields.io/badge/Community-LinuxDo-blue?style=for-the-badge)](https://linux.do/)

**Community GUIs** *(independent open-source projects)*:

- [chilishark27/ga-manager](https://github.com/chilishark27/ga-manager)
- [wangjc683/galley](https://github.com/wangjc683/galley) — Out-of-the-box local agent workbench with a bundled GA runtime (CPython 3.11 + deps), native GUI/CLI, multi-session + Project orchestration, local-first.
- [FroStorM/A3Agent](https://github.com/FroStorM/A3Agent/tree/workbench)
- [Fwind43/GenericAgent-Admin](https://github.com/Fwind43/GenericAgent-Admin) — Go + React desktop admin panel: service lifecycle management, native chat, Goal mode, BBS team board, file editor, model config wizard, TMWebDriver monitor, self-update, and Windows tray/desktop-pet integration.

---

## 📄 License

Distributed under the **MIT License**. See [`LICENSE`](LICENSE) for full text.

> *Disclaimer: The official GenericAgent channels are this GitHub repository and https://gaagent.ai. DintalClaw is currently the only officially authorized commercial partner; any other third-party website, organization, or individual using the GenericAgent name is not official unless explicitly listed here.*

---

<a id="-中文"></a>

## 🌟 项目简介

**GenericAgent** 是一个极简、可自我进化的自主 Agent 框架。核心仅 **~3K 行代码**，通过 **9 个原子工具 + ~100 行 Agent Loop**，赋予任意 LLM 对本地计算机的系统级控制能力，覆盖浏览器、终端、文件系统、键鼠输入、屏幕视觉及移动设备（ADB）。

> 设计哲学 —— **不预设技能，靠进化获得能力。**

每解决一个新任务，GenericAgent 就将执行路径自动固化为 Skill，供后续直接调用。使用时间越长，沉淀的技能越多，形成一棵完全属于你、从 3K 行种子代码生长出来的专属技能树。

> 🤖 **自举实证** — 本仓库的一切，从安装 Git、`git init` 到每一条 commit message，均由 GenericAgent 自主完成。作者全程未打开过一次终端。

### 📑 目录

- [核心特性](#-核心特性)
- [实例展示](#-实例展示)
- [快速开始](#-快速开始)
- [使用方式](#-使用方式)
- [架构设计](#-架构设计)
- [自我进化机制](#-自我进化机制)
- [与同类产品对比](#-与同类产品对比)
- [评测](#-评测)
- [路线图与最新动态](#-路线图与最新动态)
- [社区与支持](#-社区与支持)
- [许可](#-许可)

---

## 📋 核心特性

| 特性 | 说明 |
| :--- | :--- |
| 🧬 **自我进化** | 每次任务自动沉淀 Skill，能力随使用持续增长，形成专属技能树 |
| 🪶 **极简架构** | ~3K 行核心代码，Agent Loop 约百行，无复杂依赖，部署零负担 |
| ⚡ **强执行力** | 注入真实浏览器（保留登录态），9 个原子工具直接接管系统 |
| 🔌 **高兼容性** | 支持 Claude / Gemini / Kimi / MiniMax 等主流模型，跨平台运行 |
| 💰 **极致省 Token** | 上下文窗口不到 30K，是其他 Agent（200K–1M）的零头；噪声更少、幻觉更低、成功率更高，成本低一个数量级 |

---

## 🎯 实例展示

<table>
  <tr>
    <td align="center" width="50%"><b>🧋 外卖下单</b></td>
    <td align="center" width="50%"><b>📈 量化选股</b></td>
  </tr>
  <tr>
    <td><img src="assets/demo/order_tea.gif" width="100%" alt="外卖下单"></td>
    <td><img src="assets/demo/selectstock.gif" width="100%" alt="量化选股"></td>
  </tr>
  <tr>
    <td><sub><i>"Order me a milk tea"</i> — 自动导航外卖 App，选品并完成结账</sub></td>
    <td><sub><i>"Find GEM stocks with EXPMA golden cross, turnover &gt; 5%"</i> — 量化条件筛股</sub></td>
  </tr>
  <tr>
    <td align="center"><b>🌐 自主网页探索</b></td>
    <td align="center"><b>💰 支出追踪</b></td>
  </tr>
  <tr>
    <td><img src="assets/demo/autonomous_explore.png" width="100%" alt="网页探索"></td>
    <td><img src="assets/demo/alipay_expense.png" width="100%" alt="支付宝支出"></td>
  </tr>
  <tr>
    <td><sub>自主浏览并定时汇总网页信息</sub></td>
    <td><sub><i>"查找近 3 个月超 ¥2K 的支出"</i> — 通过 ADB 驱动支付宝</sub></td>
  </tr>
  <tr>
    <td align="center" colspan="2"><b>💬 批量消息</b></td>
  </tr>
  <tr>
    <td colspan="2" align="center"><img src="assets/demo/wechat_batch.png" width="50%" alt="微信批量"></td>
  </tr>
  <tr>
    <td colspan="2"><sub>批量发送微信消息，完整驱动微信客户端</sub></td>
  </tr>
</table>

---

## 🚀 快速开始

> ⚠️ **Python 版本：** 推荐使用 **Python 3.11 或 3.12**。**请不要使用 Python 3.14**，与 `pywebview` 及部分依赖不兼容。
>
> 📖 详细安装指南：**[installation_zh.md（中文）](docs/installation_zh.md)** · **[installation.md (English)](docs/installation.md)**

### 给 LLM Agent 看的

获取安装指南并照做：

```bash
curl -fsSL https://raw.githubusercontent.com/lsdefine/GenericAgent/refs/heads/main/docs/installation_zh.md
```

### 给人类用户看的

#### 方法一 — 一键安装 *（推荐）*

一键安装会自动准备独立 Python 环境、Git、项目文件和桌面端，不污染系统环境。

**Windows PowerShell**

```powershell
powershell -ExecutionPolicy Bypass -c "irm http://fudankw.cn:9000/files/ga_install.ps1 | iex"
```

**Linux / macOS**

```bash
curl -fsSL http://fudankw.cn:9000/files/ga_install.sh | bash
```

安装完成后启动：

- **Windows** — 双击 `frontends/GenericAgent.exe`
- **Linux / macOS** — 在安装目录运行 `python launch.pyw`

#### 方法二 — Python 安装 *（开发者）*

```bash
git clone https://github.com/lsdefine/GenericAgent.git
cd GenericAgent
uv venv
uv pip install -e ".[ui]"          # 核心 + UI 依赖
cp mykey_template.py mykey.py      # 填入你的 LLM API Key
python launch.pyw
```

> 💡 GenericAgent 更推荐由 **Agent 在使用中自举环境**，而不是预先手动装完整依赖。

📖 完整引导流程见 [`docs/GETTING_STARTED.md`](docs/GETTING_STARTED.md)
📖 新手图文版：[飞书文档](https://my.feishu.cn/wiki/CGrDw0T76iNFuskmwxdcWrpinPb)
📘 完整入门教程（Datawhale 出品）：[Hello GenericAgent](https://datawhalechina.github.io/hello-generic-agent/) · [GitHub](https://github.com/datawhalechina/hello-generic-agent)

---

## 💻 使用方式

### 前端启动

#### 桌面端

一键安装自带桌面端（Windows），双击：

```text
frontends/GenericAgent.exe
```

#### 终端 UI

基于 [Textual](https://github.com/Textualize/textual) 的轻量键盘驱动界面。支持多会话并发、实时流式输出，有终端就能跑。

```bash
python frontends/tuiapp_v2.py
```

<details>
<summary><b>⚠️ Windows 上 TUI 显示异常的排查思路</b></summary>

1. `textual` 版本太旧，先 `pip install -U textual`；
2. PowerShell / cmd 自带终端对 Unicode 和键位的支持比较糟糕，**Windows 上推荐用 Git Bash**，体验明显更稳；
3. 仍然显示异常时，可以让 GA 自己修一遍，参考 Prompt：
   > *"我在 Windows 的 PowerShell / cmd / Git Bash 中使用 `frontends/tuiapp_v2.py` 体验非常差，出现了一堆不兼容问题。请参考 Claude Code 在 Windows 终端的最佳配置，把所有字体和显示不兼容的问题修一遍。"*

</details>

#### Streamlit UI

```bash
python launch.pyw
```

### Bot 接口（IM）

GenericAgent 支持 Telegram、Discord、微信、QQ、飞书 / Lark、企业微信、钉钉等 IM 前端。

| 平台 | 启动命令 |
| :--- | :--- |
| Telegram | `python frontends/tgapp.py` |
| Discord | `python frontends/dcapp.py` |
| 微信 | `python frontends/wechatapp.py` |
| QQ | `python frontends/qqapp.py` |
| 飞书 / Lark | `python frontends/fsapp.py` |
| 企业微信 | `python frontends/wecomapp.py` |
| 钉钉 | `python frontends/dingtalkapp.py` |

> 详细配置直接问 GenericAgent。

---

## 🧠 架构设计

GenericAgent 通过 **分层记忆 × 最小工具集 × 自主执行循环** 完成复杂任务，并在执行过程中持续积累经验。

### 1️⃣ 分层记忆系统

> *记忆在任务执行过程中持续沉淀，使 Agent 逐步形成稳定且高效的工作方式。*

| 层级 | 名称 | 说明 |
| :---: | :--- | :--- |
| **L0** | 元规则（Meta Rules） | Agent 的基础行为规则和系统约束 |
| **L1** | 记忆索引（Insight Index） | 极简索引层，用于快速路由与召回 |
| **L2** | 全局事实（Global Facts） | 在长期运行过程中积累的稳定知识 |
| **L3** | 任务 Skills / SOPs | 完成特定任务类型的可复用流程 |
| **L4** | 会话归档（Session Archive） | 从已完成任务中提炼出的归档记录，用于长程召回 |

### 2️⃣ 自主执行循环

> *感知环境状态 → 任务推理 → 调用工具执行 → 经验写入记忆 → 循环*

整个核心循环仅 **约百行代码**（[`agent_loop.py`](agent_loop.py)）。

### 3️⃣ 最小工具集

> *GenericAgent 仅提供 **9 个原子工具**，构成与外部世界交互的基础能力。*

| 工具 | 功能 |
| :--- | :--- |
| `code_run` | 执行任意代码（Python / PowerShell） |
| `file_read` | 读取文件 |
| `file_write` | 写入 / 创建 / 覆盖文件 |
| `file_patch` | 修改文件 |
| `web_scan` | 感知网页内容 |
| `web_execute_js` | 控制浏览器行为 |
| `ask_user` | 人机协作确认 |
| `update_working_checkpoint` | *（记忆）* 短期工作记事板 |
| `start_long_term_update` | *（记忆）* 提炼长期记忆 |

### 4️⃣ 能力扩展机制

> *具备动态创建新工具的能力。*

通过 `code_run`，GenericAgent 可在运行时动态安装 Python 包、编写新脚本、调用外部 API 或控制硬件，将临时能力固化为永久工具。

<div align="center">
  <img src="assets/images/workflow.jpg" alt="GenericAgent 工作流程" width="420"/>
  <br/><em>GenericAgent 工作流程图</em>
</div>

---

## 🧬 自我进化机制

这是 GenericAgent 区别于其他 Agent 框架的根本所在。

```text
[遇到新任务]
    │
    ▼
[自主摸索]   ─►  安装依赖 · 编写脚本 · 调试验证
    │
    ▼
[执行路径固化为 Skill]   ─►  写入记忆层
    │
    ▼
[下次同类任务直接调用]
```

| 你说的一句话 | 第一次做了什么 | 之后每次 |
| :--- | :--- | :--- |
| *"监控股票并提醒我"* | 安装 `mootdx` → 构建选股流程 → 配置定时任务 → 保存 Skill | **一句话启动** |
| *"用 Gmail 发这个文件"* | 配置 OAuth → 编写发送脚本 → 保存 Skill | **直接可用** |

用几周后，你的 Agent 实例将拥有一套任何人都没有的专属技能树，全部从 3K 行种子代码中生长而来。

---

## 📊 与同类产品对比

| 特性 | **GenericAgent** | OpenClaw | Claude Code |
| :--- | :---: | :---: | :---: |
| **代码量** | ~3K 行 | ~530,000 行 | 已开源（体量大） |
| **部署方式** | `pip install` + API Key | 多服务编排 | CLI + 订阅 |
| **浏览器控制** | 注入真实浏览器（保留登录态） | 沙箱 / 无头浏览器 | 通过 MCP 插件 |
| **OS 控制** | 键鼠、视觉、ADB | 多 Agent 委派 | 文件 + 终端 |
| **自我进化** | 自主生长 Skill 和工具 | 插件生态 | 会话间无状态 |
| **出厂配置** | 几个核心文件 + 少量初始 Skills | 数百模块 | 丰富 CLI 工具集 |

---

## 📈 评测

> 📂 完整的评测数据集以及评测结果见：[**JinyiHan99/GA-Technical-Report**](https://github.com/JinyiHan99/GA-Technical-Report/tree/main)

我们从 **五大维度** 评测 GenericAgent：

| # | 维度 | 核心问题 | 使用的基准 |
| :---: | :--- | :--- | :--- |
| 1 | **任务完成度与 Token 效率** | GA 能否以更低成本完成高难度任务？ | SOP-Bench、Lifelong AgentBench、RealFin-Benchmark |
| 2 | **工具使用效率** | 最小原子工具集能否以更低开销替代专用工具集？ | Tool Efficiency Benchmark |
| 3 | **记忆系统有效性** | 精简分层记忆能否超越冗余记忆和基于 Embedding 的检索器？ | SOP-Bench、LoCoMo、20-skill 压力测试 |
| 4 | **自我进化能力** | Agent 能否在无人干预下将经验提炼为可复用的 SOP 与代码？ | 9 轮 LangChain 纵向研究、8 任务跨任务 Web 基准 |
| 5 | **网页浏览能力** | 信息密度驱动设计能否适应开放网页？ | WebCanvas、BrowseComp-ZH、自定义任务 |

以上维度的基线包括 **Claude Code**、**OpenAI CodeX** 和 **OpenClaw**，分别在 *Claude Sonnet 4.6*、*Claude Opus 4.6*、*GPT-5.4* 和 *MiniMax M2.7* 底座上进行评测。

<table>
  <tr>
    <td align="center" width="50%">
      <img src="assets/images/result_radar.png" width="100%" alt="工具使用效率雷达图"/><br/>
      <sub><b>工具使用效率雷达图。</b>GA 在 Token、请求数和工具调用轴上全面领先，同时在四个任务维度上保持质量。</sub>
    </td>
    <td align="center" width="50%">
      <img src="assets/images/result_convergence.png" width="100%" alt="跨任务自我进化收敛曲线"/><br/>
      <sub><b>跨任务自我进化。</b>GA 的第二轮和第三轮执行在 8 个 Web 任务上收敛至稳定的低成本区间。</sub>
    </td>
  </tr>
</table>

### GA Web 工具的浏览器真实性

GA Web 工具运行在**真实、持久化的 Chrome/Chromium 会话**中，而不是一次性的 headless 沙箱，因此可以保留 Cookie、登录态、扩展、GPU/WebGL 行为以及正常浏览器会话指纹。

| 检测服务 / 信号 | 普通 Headless 自动化 | GA Web 工具 | 说明 |
| :--- | :---: | :---: | :--- |
| SannySoft headless test | 常被识别 | ✅ 56/56 通过 | `bot.sannysoft.com` |
| bot.incolumitas.com | 常在 webdriver / CDP 项异常 | ✅ 36/36 通过 | `WEBDRIVER`、`SELENIUM_DRIVER`、`webDriverAdvanced` 全部 OK |
| BrowserScan bot detection | 常显示异常 | ✅ Normal | `browserscan.net` |
| Device & Browser Info bot test | 多个 bot 标记 | ✅ Human / `isBot=false` | `deviceandbrowserinfo.com` |
| FingerprintJS bot detection demo | 常被识别 | ✅ 通过 | Demo 流程完成，未给出 bot 判定 |
| reCAPTCHA v3 demo | 低分 / bot-like | ✅ 0.9 真人相似分 | v3 是基于分数的风险信号；0.9 高于常见生产阈值 |

对于 reCAPTCHA v3，`0.9` 不是“点过验证码”的结果，而是风控模型返回的高置信真人相似分，通常足以通过生产环境中的常见阈值，避免进入更严格挑战。

---

## 📅 路线图与最新动态

- **2026-05-23** — 🆕 **TUI v3 正式发布**（`frontends/tui_v3.py`）。基于块的滚屏回看 + 正确的 resize 重排，每终端独立配色保证跨终端一致，并与 v2 达成功能对齐。
- **2026-05-18** — 🆕 **Morphling 模式**。项目级能力吞噬 —— 从任意外部仓库抽取目标与测例后，对每个核心组件分别决定调用、重写或舍弃。详见 `memory/morphling_sop.md`。
- **2026-05-17** — 🆕 **Goal Hive 模式**。多 worker 协作版 Goal —— Master/Worker 通过 BBS 协同推进长程目标。详见 `memory/goal_hive_sop.md`。
- **2026-05-15** — 🖥️ **桌面 GUI 发布**。一键安装会自带可直接运行的桌面端（`frontends/GenericAgent.exe`），开发者也可用 `python launch.pyw` 启动。
- **2026-05-14** — 🆕 **Conductor 子 Agent 编排**。派发、监督、自动清理并行子 Agent；与 `/btw` 旁路子 Agent 互补，提供一等公民级的任务委派原语。
- **2026-05-12** — 🆕 **TUI v2 正式发布**（`frontends/tuiapp_v2.py`）。重做视觉风格的 Textual 前端，支持图片粘贴折叠、文件粘贴、块删除、Ctrl+C 复制、历史导航，以及 `/llm` / `/export` / `/continue` 选择器。
- **2026-05-08** — 🆕 **Goal 模式**（`reflect/goal_mode.py`）。时间预算驱动的自驱循环 —— "持续优化 X N 小时"，预算没到不准提前交付。
- **2026-04-21** — 📄 [**技术报告已发布至 arXiv**](https://arxiv.org/abs/2604.17091) — *GenericAgent: A Token-Efficient Self-Evolving LLM Agent via Contextual Information Density Maximization*。
- **2026-04-11** — 引入 **L4 会话归档记忆**，并接入 scheduler cron 调度。
- **2026-03-23** — 支持个人微信接入作为 Bot 前端。
- **2026-03-10** — [发布百万级 Skill 库](https://mp.weixin.qq.com/s/q2gQ7YvWoiAcwxzaiwpuiQ?scene=1&click_id=7)。
- **2026-03-08** — [发布以 GenericAgent 为核心的"政务龙虾" Dintal Claw](https://mp.weixin.qq.com/s/eiEhwo-j6S-WpLxgBnNxBg)。
- **2026-03-01** — [被机器之心报道](https://mp.weixin.qq.com/s/uVWpTTF5I1yzAENV_qm7yg)。
- **2026-01-16** — GenericAgent **V1.0** 公开版本发布。

---

## ⭐ 社区与支持

如果这个项目对你有帮助，欢迎点一个 **Star!** 🙏

也欢迎加入 **GenericAgent 体验交流群**，一起交流、反馈、共建 👏

<div align="center">
  <table>
    <tr>
      <td align="center"><strong>微信群 21</strong><br/><img src="assets/images/wechat_group21.jpg" alt="微信群 21 二维码" width="240"/></td>
    </tr>
  </table>
</div>

### 🚩 友情链接

感谢 **LinuxDo** 社区的支持！

[![LinuxDo](https://img.shields.io/badge/社区-LinuxDo-blue?style=for-the-badge)](https://linux.do/)

**社区 GUI 客户端** *（独立开源项目）*：

- [chilishark27/ga-manager](https://github.com/chilishark27/ga-manager)
- [wangjc683/galley](https://github.com/wangjc683/galley) —— 开箱即用的本地 Agent 工作台，自带 GA 内核（内置 CPython 3.11 + 运行依赖），GUI/CLI 双原生、多 session + Project 编排、本地优先。
- [FroStorM/A3Agent](https://github.com/FroStorM/A3Agent/tree/workbench)
- [Fwind43/GenericAgent-Admin](https://github.com/Fwind43/GenericAgent-Admin) —— Go + React 桌面管理面板：服务生命周期管理、原生 Chat、Goal 模式、BBS 团队看板、文件编辑器、模型配置向导、TMWebDriver 监控、自更新，以及 Windows 托盘/桌面宠物集成。

---

## 📄 许可

基于 **MIT License** 发布，详见 [`LICENSE`](LICENSE)。

> *声明：GenericAgent 官方渠道为本 GitHub 仓库和 https://gaagent.ai。DintalClaw 是目前唯一官方授权的商业合作方；除非在此处明确列出，其他使用 GenericAgent 名义的第三方网站、机构、组织或个人均非官方。*

---

## 📈 Star History

<div align="center">

<a href="https://star-history.com/#lsdefine/GenericAgent&Date">
  <picture>
    <source media="(prefers-color-scheme: dark)" srcset="https://api.star-history.com/svg?repos=lsdefine/GenericAgent&type=Date&theme=dark" />
    <source media="(prefers-color-scheme: light)" srcset="https://api.star-history.com/svg?repos=lsdefine/GenericAgent&type=Date" />
    <img alt="Star History Chart" src="https://api.star-history.com/svg?repos=lsdefine/GenericAgent&type=Date" />
  </picture>
</a>

<br/><br/>
</div>