Interactive Feedback MCP
About
An MCP server for AI-assisted development tools like Cursor and Claude, supporting interactive feedback workflows with AI.
Details
- Author
- zengxiaolou
- Categories
- Developer Tools, Automation, AI
Jump to
Setup
Install Interactive Feedback MCP in your MCP client (Claude Desktop, Cursor, Windsurf, and others).
Repository: https://github.com/zengxiaolou/Interactive-Feedback-MCP
Follow the installation instructions in the repository README, then restart your MCP client.
Interactive Feedback MCP 是一个高性能的 MCP Server,专为Cursor、Claude Desktop和Windsurf等AI开发工具设计。采用现代化三栏布局和毛玻璃效果UI,支持实时交互反馈、多媒体处理和智能项目分析。
- 🔄 实时双向对话- AI助手可暂停并请求用户澄清,避免猜测性开发
- 🎯 预定义选项- 快速选择常用操作,提升开发效率
- 📊 智能分析- 自动分析用户意图、紧急程度和项目上下文
- ⚡ 性能优化- 启动时间<2秒,UI响应<100ms,内存占用<100MB
- 🖼️ 三栏布局- 消息内容(40%) + 智能推荐(40%) + 项目信息(20%)
- ✨ 毛玻璃效果- 深色主题,强制模式,不受系统主题影响
- 🌏 中文优化- 完美支持中文字体和UTF-8编码
- 📱 响应式设计- 适配不同屏幕尺寸和DPI设置
- 🏗️ MCP协议- 基于FastMCP框架的标准化工具调用
- 🎯 PySide6 UI- 现代化Qt界面框架
- ⚡ 性能监控- 内置性能跟踪和优化系统
- ⚙️ 配置管理- 统一配置系统,支持主题切换和个性化设置
- 每次提示消耗API额度,成本高昂
- 基于猜测的开发导致错误代码
- 单向交互,无法及时澄清需求
- 迭代效率低,调试时间长
- 🔄 AI可在单次请求内多轮交互
- 💰 工具调用不计入API使用量
- ✅ 确认后再执行,减少错误
- 🚀 效率提升5倍以上
- Python: 3.10+
- 系统: Windows 10+, macOS 12+, Ubuntu 20.04+
- 内存: 建议4GB+
- 存储: 500MB可用空间
# macOS/Linux curl -LsSf https://astral.sh/uv/install.sh | sh # Windows (PowerShell) powershell -c "irm https://astral.sh/uv/install.ps1 | iex" # 或者使用 pip pip install uv
git clone https://github.com/your-username/interactive-feedback-mcp.git cd interactive-feedback-mcp
# 测试MCP服务器 uv run server.py # 测试UI界面 uv run enhanced_feedback_ui.py --prompt "测试消息" --output-file test.json
{ "mcpServers": { "interactive-feedback": { "command": "uv", "args": [ "--directory", "/absolute/path/to/interactive-feedback-mcp", "run", "server.py" ], "timeout": 600, "autoApprove": ["interactive_feedback"] } } }
{ "mcpServers": { "interactive-feedback": { "command": "uv", "args": [ "--directory", "/absolute/path/to/interactive-feedback-mcp", "run", "server.py" ], "timeout": 600 } } }
- 使用绝对路径确保正确找到项目
- Windows用户请使用正斜杠/或双反斜杠\\
- 配置后需要重启AI客户端
在Cursor Settings > Rules for AI中添加:
# Interactive Feedback MCP 使用规则 ## 强制交互协议 - 收到用户消息后,必须先调用 interactive_feedback 工具进行智能分析 - 提供预定义选项供用户快速选择 - 执行操作前必须获得用户确认 - 完成任务后询问是否需要进一步操作 ## 使用场景 - 需求不明确时:询问澄清 - 有多种实现方案时:提供选项 - 重要操作前:请求确认 - 任务完成后:询问后续需求 ## 格式示例
创建~/.interactive_feedback_mcp/config.json:
{ "ui": { "theme": "enhanced_glassmorphism", "language": "zh_CN", "font_family": "PingFang SC", "font_size": 14, "window_width": 1400, "window_height": 1200, "panel_ratios": [40, 40, 20] }, "performance": { "max_startup_time": 2.0, "max_response_time": 100.0, "max_memory_usage": 100.0, "enable_monitoring": true } }
- 📊 多级别日志记录(DEBUG, INFO, WARNING, ERROR, CRITICAL)
- 🔄 文件轮转和大小控制(默认10MB轮转,保留5个备份)
- ⚡ 性能监控(记录操作耗时,识别慢操作)
- 📋 项目上下文记录(自动记录项目信息和Git状态)
- 🐛 错误详情追踪(包含堆栈信息和上下文)
- 👁️ 实时日志监控
logs/ ├── interactive_feedback_mcp.log # 主日志文件 ├── errors.log # 错误日志 ├── performance.log # 性能日志 └── project_context.log # 项目上下文日志
# 查看日志摘要 python manage_logs.py summary # 查看最近50行主日志 python manage_logs.py view --type main --lines 50 # 查看错误日志 python manage_logs.py view --type error # 搜索日志内容 python manage_logs.py search "错误关键词" --type all # 分析错误统计 python manage_logs.py analyze # 实时监控日志 python manage_logs.py monitor # 清理30天前的日志 python manage_logs.py cleanup --days 30 # 导出日志文件 python manage_logs.py export --output logs_backup.zip # 配置日志级别 python manage_logs.py config --level DEBUG
- 📝 日志级别和格式
- 📦 文件大小和轮转设置
- 🖥️ 控制台输出控制
- ⏱️ 性能监控阈值
- 🧹 自动清理策略
python manage_logs.py view --type performance
python manage_logs.py view --type context
python manage_logs.py search "UI启动失败" --type all
请使用 interactive_feedback 询问我想要什么类型的API设计
- 左栏(40%):消息内容和用户输入
- 中栏(40%):AI智能推荐和选项
- 右栏(20%):项目信息和Git状态
- Ctrl+Enter:提交反馈
- Escape:取消操作
- Ctrl+1-5:快速选择预定义选项
- Ctrl+/:显示帮助信息
interactive-feedback-mcp/ ├── server.py # MCP服务器入口 ├── enhanced_feedback_ui.py # UI主程序 ├── rules.md # 开发规范文档 ├── pyproject.toml # 项目配置 ├── ui/ # UI组件模块 │ ├── components/ # 核心组件 │ │ ├── three_column_layout.py # 三栏布局 │ │ ├── enhanced_markdown_renderer.py # 渲染引擎 │ │ └── main_window.py # 主窗口 │ ├── styles/ # 样式主题 │ │ └── enhanced_glassmorphism.py # 毛玻璃主题 │ ├── utils/ # 工具模块 │ │ ├── performance.py # 性能监控 │ │ └── config_manager.py # 配置管理 │ └── widgets/ # 自定义控件 └── tests/ # 测试文件
# 安装开发依赖 uv sync --dev # 运行测试 uv run python -m pytest tests/ # 代码格式化 uv run python -m black . # 类型检查 uv run python -m mypy .
- 启动时间: < 2秒
- UI响应: < 100毫秒
- 内存使用: < 100MB (空闲状态)
- CPU占用: < 5% (空闲状态)
# 检查编码设置 python -c "import locale; print(locale.getpreferredencoding())" # 强制UTF-8 export PYTHONIOENCODING=utf-8
# 重置配置 rm ~/.interactive_feedback_mcp/config.json # 重新启动应用 uv run enhanced_feedback_ui.py --prompt "测试" --output-file test.json
# 检查性能指标 uv run python -c " from ui.utils.performance import global_performance_monitor monitor = global_performance_monitor monitor.start_monitoring() print(monitor.get_current_metrics()) "
- 检查路径配置是否正确
- 确认uv命令可用
- 查看客户端日志错误信息
- 验证server.py可正常运行
- Fork 项目
- 创建功能分支:git checkout -b feature/new-feature
- 遵循开发规范
- 编写测试用例
- 提交PR
- 遵循rules.md中的开发规则
- 使用Black进行代码格式化
- 保持测试覆盖率 > 80%
- 编写清晰的提交消息
- FastMCP- MCP框架支持
- PySide6- UI框架
- Cursor- AI开发工具
- Claude- AI助手
- 问题反馈:GitHub Issues
- 功能建议:GitHub Discussions
This is a web browser that enables your coding agent, such as Claude Code, to visit websites on your behalf and assist you in identifying bugs or creating UI test cases.
Agent-native developer Q&A API with MCP + A2A endpoints for citations, job pickup, and answer submission.
MCP bridge that lets Claude Code delegate heavy tasks to the Antigravity CLI (agy) — purpose-built tools, model routing with fallback, session continuity, and output truncation to save Claude's context and tokens.
Embeds intelligent guidance into AI workflows to organize development and ensure quality.
Open-source AI coding stack — bundled MCP servers, agent runtime, and developer tooling for shipping AI-native dev tools.
Local agent workbench bundling OpenHands, Goose, Aider, and ashlrcode against one local LLM, with ashlr-plugin MCP servers pre-wired.
Async Parallel Antigravity for Codex & Claude Code
Run parallel, resumable, human-operable Antigravity CLI sessions from Codex, Claude Code, or any MCP-capable agent harness.
knowledge network for AI coding agents. Developers connect their agents to a shared pool of verified solutions — saving tokens, reducing debugging time, and getting better results. Solution authors earn when their work helps others.
Orchestrates multiple Claude Code agents across iTerm2 sessions, providing centralized management and inter-agent communication.
Multi-LLM Design and Build Team. Confer and create with a team of LLMs.
An MCP server for initial app planning and creating a good starting point for an app.
Sign in to leave a review
Use Google, GitHub, or an email account so ratings stay tied to real people.
No reviews posted yet.


