Skip to content

Commit c4e7394

Browse files
ceilf6ceilf6
committed
docs: README 长尾关键词
1 parent 53b9112 commit c4e7394

2 files changed

Lines changed: 22 additions & 4 deletions

File tree

‎README.md‎

Lines changed: 11 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -8,14 +8,23 @@
88
[![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](https://opensource.org/licenses/MIT)
99
[![Node.js Version](https://img.shields.io/badge/node-%3E%3D20.0.0-brightgreen)](https://nodejs.org)
1010

11-
> Enterprise-grade AI Agent System - Constrained by SDD, Powered by MCP for Controlled Perception and Execution
11+
> Enterprise-grade frontend AI coding agent and MCP-powered automation system - constrained by Specification Driven Development (SDD) for controlled planning, code generation, browser-aware execution, and repository workflows
1212
1313
[中文文档](docs/README-CN.md) | [Quick Start](docs/QUICKSTART.md) | [Architecture](docs/architecture.md) | [Design Doc](docs/design.md)
1414

15-
FrontAgent is an AI Agent system designed specifically for frontend engineering, addressing core challenges faced when deploying agents in real-world engineering scenarios:
15+
FrontAgent is an open-source AI coding agent for frontend engineering. It helps teams build, modify, validate, and ship web applications through an agentic CLI, VS Code extension, desktop app, local MCP server, RAG planning, browser-aware automation, and SDD guardrails.
1616

1717
> **Distilled Planner Models**: FrontAgent's Planner stage has been distilled into Hugging Face planner assets collected under [FrontAgent: Frontend Engineering Agent](https://hf.co/collections/ceilf6/frontagent-frontend-engineering-agent). Load the published adapters on their supported Qwen Coder base models to generate frontend execution plans directly, without calling large LLM APIs. The training workflow, prompts, evaluation scripts, and Hugging Face release metadata live in [models/frontagent-planner](models/frontagent-planner).
1818
19+
Use FrontAgent when you need a frontend AI agent that can:
20+
21+
- Generate and refactor React, TypeScript, Vite, Tailwind CSS, and modern web UI code from structured execution plans.
22+
- Run as an AI agent CLI, VS Code AI extension, desktop agent app, or stdio MCP server for Claude Desktop, Cursor, Codex, and other MCP hosts.
23+
- Combine repository-aware RAG, Filesense navigation, facts memory, and module dependency tracking to reduce path hallucinations.
24+
- Execute browser-aware validation, page inspection, shell commands, and git/gh repository workflows behind explicit safety controls.
25+
- Enforce Specification Driven Development (SDD), minimal patches, self-healing error recovery, and quality gates for production frontend teams.
26+
- Use distilled Qwen Coder planner models from Hugging Face for local or lower-cost frontend execution planning.
27+
1928
- ✅ **Two-Stage Architecture** - Separate planning and execution to avoid JSON parsing errors and enable dynamic code generation
2029
- ✅ **Phase-Based Execution** - Steps grouped by phases with error recovery within each phase
2130
- ✅ **Self-Healing** - Tool Error Feedback Loop automatically analyzes errors and generates fix steps

‎docs/README-CN.md‎

Lines changed: 11 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -8,14 +8,23 @@
88
[![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](https://opensource.org/licenses/MIT)
99
[![Node.js Version](https://img.shields.io/badge/node-%3E%3D20.0.0-brightgreen)](https://nodejs.org)
1010

11-
> 面向前端工程的企业级 AI Agent 系统 — 以 SDD 为约束、通过 MCP 控制感知与执行
11+
> 面向前端工程的企业级 AI 编程 Agent 与 MCP 自动化系统 — 以 Specification Driven Development (SDD) 约束规划、代码生成、浏览器感知执行与仓库工作流
1212
1313
[English README](../README.md) | [快速开始](QUICKSTART.md) | [架构](architecture.md) | [设计文档](design.md)
1414

15-
FrontAgent 是一个专为前端工程设计的 AI Agent 系统，解决了在真实工程场景中部署 agent 时遇到的核心问题：
15+
FrontAgent 是一个开源前端 AI 编程 Agent，面向真实前端工程场景，支持通过 Agent CLI、VS Code AI 插件、桌面客户端、本地 MCP Server、RAG 规划、浏览器感知自动化与 SDD 护栏来构建、修改、验证和交付 Web 应用。
1616

1717
> **蒸馏 Planner 模型**：FrontAgent 的 Planner 阶段已蒸馏为 Hugging Face Planner 资产，统一收录在 [FrontAgent: Frontend Engineering Agent](https://hf.co/collections/ceilf6/frontagent-frontend-engineering-agent)。在支持的 Qwen Coder 基座模型上加载已发布 adapter，即可直接生成前端执行计划，无需调用大型 LLM API。训练流程、提示词、评估脚本和 Hugging Face 发布元数据位于 [models/frontagent-planner](../models/frontagent-planner)。
1818
19+
如果你正在寻找前端 AI Agent，FrontAgent 适合这些场景：
20+
21+
- 基于结构化执行计划生成和重构 React、TypeScript、Vite、Tailwind CSS 与现代 Web UI 代码。
22+
- 作为 AI Agent CLI、VS Code AI 插件、桌面 Agent 应用，或 Claude Desktop、Cursor、Codex 等 MCP Host 可接入的 stdio MCP Server。
23+
- 结合仓库级 RAG、Filesense 导航、事实记忆与模块依赖跟踪，减少路径幻觉和上下文遗漏。
24+
- 在显式安全控制下执行浏览器感知验证、页面检查、Shell 命令和 git/gh 仓库工作流。
25+
- 用 Specification Driven Development (SDD)、最小补丁、自愈式错误恢复和质量门禁约束生产级前端团队的 AI 编程流程。
26+
- 使用 Hugging Face 上的 Qwen Coder 蒸馏 Planner 模型，在本地或低成本场景中生成前端执行计划。
27+
1928
- ✅ **两阶段架构** — 规划与执行分离，避免 JSON 解析错误并支持动态代码生成
2029
- ✅ **分阶段执行** — 步骤按阶段分组，支持阶段内错误恢复
2130
- ✅ **自愈能力** — 工具错误反馈循环会自动分析错误并生成修复步骤

0 commit comments

Comments
 (0)