### Start Agent Installation Wizard

Source: https://github.com/iflow-ai/iflow-cli/blob/main/docs_cn/examples/subagent.md

Initiate the guided installation process for new agents by running the '/agents install' command.

```bash
# Start the Agent installation wizard
/agents install
```

--------------------------------

### Project Initialization and Dependency Installation

Source: https://github.com/iflow-ai/iflow-cli/blob/main/docs_en/examples/basic-usage.md

Demonstrates initializing a project with `/init` and then installing dependencies using `!npm install`. The AI can then analyze the project based on the installed dependencies.

```bash
User: /init

AI: I'll analyze your project structure...

User: !npm install

AI: Dependencies installed successfully. Based on your package.json, this is a React application.
Would you like me to check for any configuration issues?
```

--------------------------------

### Browse and Install Commands from Marketplace

Source: https://context7.com/iflow-ai/iflow-cli/llms.txt

Discover and install new slash commands from the marketplace using `iflow commands` CLI commands. Use IDs or names to get details and install commands to your project or globally.

```bash
/commands online
```

```bash
/commands get 123
```

```bash
/commands add 123
```

```bash
/commands add commit --scope global
```

--------------------------------

### Full settings.json Example

Source: https://github.com/iflow-ai/iflow-cli/blob/main/docs_en/configuration/settings.md

A comprehensive example demonstrating various configuration options within the settings.json file.

```json
{
  "selectedAuthType": "iflow",
  "apiKey": "sk-xxx",
  "baseUrl": "https://apis.iflow.cn/v1",
  "modelName": "Qwen3-Coder",
  "theme": "GitHub",
  "sandbox": "docker",
  "toolDiscoveryCommand": "bin/get_tools",
  "toolCallCommand": "bin/call_tool",
  "mcpServers": {
    "mainServer": {
      "command": "bin/mcp_server.py"
    },
    "anotherServer": {
      "command": "node",
      "args": ["mcp_server.js", "--verbose"]
    }
  },
  "telemetry": {
    "enabled": true,
    "target": "local",
    "otlpEndpoint": "http://localhost:4317",
    "logPrompts": true
  },
  "usageStatisticsEnabled": true,
  "hideTips": false,
  "hideBanner": false,
  "maxSessionTurns": 10,
  "summarizeToolOutput": {
    "run_shell_command": {
      "tokenBudget": 100
    }
  }
}
```

--------------------------------

### View Available Agents

Source: https://github.com/iflow-ai/iflow-cli/blob/main/docs_cn/examples/subagent.md

Use these commands to list local agents, view their descriptions, browse the online market, or start the installation wizard.

```bash
/agents list
/agents list desc
/agents online
/agents install
```

--------------------------------

### Run Local Observability Setup

Source: https://github.com/iflow-ai/iflow-cli/blob/main/docs_en/features/telemetry.md

Execute this command in the repository root to automate the setup of local observability pipelines, including installing OpenTelemetry Collector and Jaeger, starting services, and configuring workspace settings.

```bash
npm run telemetry -- --target=local
```

--------------------------------

### Manual Agent Installation Steps

Source: https://github.com/iflow-ai/iflow-cli/blob/main/docs_cn/examples/subagent.md

Manually install a custom agent by creating its directory and configuration file, then restarting the CLI.

```bash
# Create Agent directory
mkdir -p ~/.iflow/agents

# Create a new Agent file
nano ~/.iflow/agents/my-agent.md

# Restart CLI to load new Agent
iflow
```

--------------------------------

### Image Processing Examples

Source: https://github.com/iflow-ai/iflow-cli/blob/main/docs_en/features/interactive.md

Examples demonstrating how to prompt for image analysis and code generation based on visual input.

```bash
> [Pasted image #1] Please analyze the design issues of this user interface
```

```bash
> [Pasted image #2] Which of these two interface layouts is better?
```

```bash
> Help me write the corresponding CSS code based on [Pasted image #1] this design draft
```

--------------------------------

### Install iFlow CLI on Windows

Source: https://github.com/iflow-ai/iflow-cli/blob/main/docs_en/quickstart.md

Install iFlow CLI on Windows by first installing Node.js, then running the npm install command in a restarted terminal.

```shell
npm install -g @iflow-ai/iflow-cli@latest
```

--------------------------------

### React Project Configuration Example

Source: https://github.com/iflow-ai/iflow-cli/blob/main/docs_en/configuration/iflow.md

An example IFLOW.md file for a React project, outlining technology stack, code style, and file structure conventions.

```markdown
# React Project Configuration

## Project Type
This is a modern React application using TypeScript and Vite.

## Development Standards
- Prioritize functional components and Hooks
- State management using Zustand
- Styling using Tailwind CSS
- Testing using Jest + Testing Library

## Code Style
- Component files use PascalCase naming
- Hook files start with use
- Constants use UPPER_SNAKE_CASE
- Interfaces start with I (e.g., IUser)

## File Structure Conventions
```
src/
├── components/     # Reusable components
├── pages/         # Page components
├── hooks/         # Custom hooks
├── stores/        # Zustand stores
├── services/      # API calls
├── types/         # TypeScript types
└── utils/         # Utility functions
```

## Special Requirements
- All components must have TypeScript type declarations
- Asynchronous operations use React Query
- Routing uses React Router v6
```

--------------------------------

### Install MCP Server from Marketplace

Source: https://context7.com/iflow-ai/iflow-cli/llms.txt

Use the `iflow mcp add-json` command to install MCP servers from the marketplace. Specify the server name and its JSON configuration, including the command and arguments.

```bash
iflow mcp add-json 'playwright' '{"command":"npx","args":["-y","@iflow-mcp/playwright-mcp@0.0.32"]}'
```

--------------------------------

### Simple File Import Example

Source: https://github.com/iflow-ai/iflow-cli/blob/main/docs_en/features/memory-import.md

Demonstrates importing multiple files into a main configuration document.

```markdown
# Main configuration file

Welcome to my project!

@./getting-started.md

## Features

@./features/overview.md
```

--------------------------------

### UI Design Analysis Example

Source: https://github.com/iflow-ai/iflow-cli/blob/main/docs_en/features/interactive.md

Example of providing an image for UI/UX analysis.

```bash
> [Pasted image #1] Please analyze the user experience issues of this login page
```

--------------------------------

### File Analysis Example

Source: https://github.com/iflow-ai/iflow-cli/blob/main/docs_en/features/interactive.md

Example of combining file references with specific analysis requests.

```bash
> Please analyze the performance issues of @src/components/Header.tsx and provide optimization suggestions
```

--------------------------------

### Install iFlow CLI

Source: https://context7.com/iflow-ai/iflow-cli/llms.txt

Choose an installation method based on your operating system and package manager. Node.js 22+ is required for npm installation. Verify installation with `iflow --version`.

```bash
bash -c "$(curl -fsSL https://cloud.iflow.cn/iflow-cli/install.sh)"
```

```bash
brew tap iflow-ai/iflow-cli
brew install iflow-cli
```

```bash
npm i -g @iflow-ai/iflow-cli@latest
```

```bash
iflow --version
```

```bash
npm uninstall -g @iflow-ai/iflow-cli
```

--------------------------------

### Install and Manage Workflows

Source: https://context7.com/iflow-ai/iflow-cli/llms.txt

Commands for adding workflows from the marketplace and initializing the iFlow environment.

```bash
# Install workflow from marketplace
iflow workflow add "ppt-generator-v3-OzctqA"

# Start iFlow in project directory
iflow

# Execute workflow command
/ppt-generator

# Flowchart drawing workflow
iflow workflow add "excalidraw-OzctqA"
/excalidraw your drawing topic

# Workflow directory structure after installation:
# project/
# ├── .iflow/
# │   ├── agents/          # Agent configurations
# │   ├── commands/        # Command implementations
# │   ├── IFLOW.md         # Workflow documentation
# │   └── settings.json    # MCP configurations
# └── IFLOW.md             # Root workflow config
```

--------------------------------

### Backend Developer Quick Start

Source: https://github.com/iflow-ai/iflow-cli/blob/main/docs_en/scenarios.md

Commands for initializing API project analysis and performing database-related operations.

```bash
# 1. API project analysis
cd my-api-project
iflow
> /init
> Analyze the architecture of this API project and generate interface documentation

# 2. Database-related operations
> Design user table structure based on business requirements
> Optimize the performance of this SQL query
> Generate database migration scripts
```

--------------------------------

### Launch iFlow CLI

Source: https://github.com/iflow-ai/iflow-cli/blob/main/README.md

Command to start the iFlow CLI after installation. Navigate to your workspace in the terminal and run this command.

```shell
iflow
```

--------------------------------

### Install iFlow CLI

Source: https://github.com/iflow-ai/iflow-cli/blob/main/IFLOW.md

Execute this command in your terminal to download and run the installation script.

```bash
bash -c "$(curl -fsSL https://cloud.iflow.cn/iflow-cli/install.sh)"
```

--------------------------------

### Install MCP tools via Marketplace

Source: https://github.com/iflow-ai/iflow-cli/blob/main/docs_en/examples/mcp.md

Use the add-json command to install MCP tools from the marketplace. Commands can be executed at the project level or with the ! prefix inside the iFlow CLI.

```bash
# Basic syntax
iflow mcp add-json 'server-name' '{JSON configuration}'

# Example: Install Playwright automation tool
iflow mcp add-json 'playwright' '{"command":"npx","args":["-y","@iflow-mcp/playwright-mcp@0.0.32"]}'

# When executing in iFlow CLI (add ! prefix), manually refresh mcp list
!iflow mcp add-json 'playwright' '{"command":"npx","args":["-y","@iflow-mcp/playwright-mcp@0.0.32"]}'
/mcp refresh
```

```bash
iflow mcp add-json --scope user 'server-name' '{JSON configuration}'
```

--------------------------------

### Install iFlow CLI via One-Click Script

Source: https://github.com/iflow-ai/iflow-cli/blob/main/docs_en/quickstart.md

Execute a one-click installation script to install iFlow CLI and all required dependencies.

```shell
bash -c "$(curl -fsSL https://gitee.com/iflow-ai/iflow-cli/raw/main/install.sh)"
```

--------------------------------

### Project-Specific Memory Examples

Source: https://github.com/iflow-ai/iflow-cli/blob/main/docs_cn/configuration/iflow.md

Examples of tailored IFLOW.md configurations for different project types and team workflows.

```markdown
# React 项目配置

## 项目类型
这是一个现代 React 应用，使用 TypeScript 和 Vite。

## 开发规范
- 优先使用函数组件和 Hooks
- 状态管理使用 Zustand
- 样式使用 Tailwind CSS
- 测试使用 Jest + Testing Library

## 代码风格
- 组件文件使用 PascalCase 命名
- Hook 文件以 use 开头
- 常量使用 UPPER_SNAKE_CASE
- 接口以 I 开头（如 IUser）

## 文件结构约定
```
src/
├── components/     # 可复用组件
├── pages/         # 页面组件
├── hooks/         # 自定义 hooks
├── stores/        # Zustand stores
├── services/      # API 调用
├── types/         # TypeScript 类型
└── utils/         # 工具函数
```

## 特殊要求
- 所有组件必须有 TypeScript 类型声明
- 异步操作使用 React Query
- 路由使用 React Router v6
```

```markdown
# Node.js 后端项目

## 技术栈
- Framework: Express.js
- Database: PostgreSQL + Prisma ORM
- Authentication: JWT
- Validation: Zod
- Testing: Jest + Supertest

## 架构模式
采用分层架构：Controller -> Service -> Repository -> Database

## API 规范
- RESTful API 设计
- 统一错误处理中间件
- 请求参数验证使用 Zod
- 响应格式：`{ success: boolean, data?: any, error?: string }`

## 安全要求
- 所有路由需要适当的认证和授权
- 敏感信息使用环境变量
- 输入数据必须验证和清理
- 错误信息不暴露内部实现

## 部署配置
- 使用 Docker 容器化
- 环境变量通过 .env 文件管理
- 生产环境使用 PM2 进程管理
```

```markdown
# 团队开发规范

## 代码提交规范
- 使用 Conventional Commits 格式
- 提交前必须通过 lint 和 test
- 每个 PR 必须有代码审查
- 重要变更需要技术方案讨论

## 分支策略
- main: 生产环境代码
- develop: 开发环境代码  
- feature/*: 功能分支
- hotfix/*: 紧急修复分支

## 代码审查要点
1. 代码逻辑正确性
2. 性能考虑
3. 安全性检查
4. 测试覆盖度
5. 文档完整性

## 沟通协作
- 重大架构变更先讨论再实现
- 遇到问题及时在群里沟通
- 定期进行技术分享
- 保持代码和文档同步更新
```

--------------------------------

### Code Debugging Example

Source: https://github.com/iflow-ai/iflow-cli/blob/main/docs_en/features/interactive.md

Example of providing an image screenshot for debugging purposes.

```bash
> [Pasted image #1] How to solve the problem shown in this error screenshot?
```

--------------------------------

### Generated Memory Content Example

Source: https://github.com/iflow-ai/iflow-cli/blob/main/docs_cn/configuration/iflow.md

An example of the content structure generated by the /init command.

```markdown
# 项目上下文

## 项目概述
这是一个基于 TypeScript + React + Node.js 的全栈应用。

## 技术栈
- 前端：React 18, TypeScript, Vite
- 后端：Node.js, Express, TypeScript
- 数据库：PostgreSQL
- 测试：Jest, React Testing Library

## 项目结构
```
src/
├── components/     # React 组件
├── services/      # API 服务
├── utils/         # 工具函数
└── types/         # TypeScript 类型定义
```

## 开发规范
- 使用 TypeScript 严格模式
- 组件采用函数式写法
- 使用 ESLint + Prettier 代码格式化
- 提交前运行 pre-commit hooks
```

--------------------------------

### Install iFlow CLI SDK

Source: https://github.com/iflow-ai/iflow-cli/blob/main/docs_en/sdk/sdk-python.md

Install the required Python package via pip.

```bash
pip install iflow-cli-sdk
```

--------------------------------

### Start New Project with iFlow CLI

Source: https://github.com/iflow-ai/iflow-cli/blob/main/README.md

Initiate a new project by describing your desired creation in the terminal after navigating to your project directory.

```shell
cd new-project/
iflow
> Create a web-based Minecraft game using HTML
```

--------------------------------

### Main IFLOW.md Configuration

Source: https://github.com/iflow-ai/iflow-cli/blob/main/docs_en/configuration/iflow.md

Example of a main IFLOW.md file that imports other documentation files and includes project-specific instructions.

```markdown
# Project Overall Configuration

@./.iflow/architecture.md
@./.iflow/coding-style.md
@./.iflow/deployment.md

## Project-Specific Instructions
Please follow the above architecture and coding standards when working on this project.
```

--------------------------------

### IFLOW.md Context File Example

Source: https://context7.com/iflow-ai/iflow-cli/llms.txt

Example of an IFLOW.md context file used for project-specific instructions to the AI. This markdown file defines general instructions, coding style, and project structure.

```markdown
# Project: E-Commerce Platform

## General Instructions

- Use TypeScript for all new code with strict mode enabled
- Follow existing coding patterns and conventions
- Ensure all functions have JSDoc documentation
- Write unit tests for new functionality

## Coding Style

- Use 2 spaces for indentation
- Interface names prefixed with `I` (e.g., `IUserService`)
- Private members prefixed with underscore
- Use strict equality (`===`)

## Project Structure

- `src/api/` - REST API endpoints
- `src/services/` - Business logic
- `src/models/` - Data models
- `src/utils/` - Helper functions

```

--------------------------------

### Conditional Import Configuration

Source: https://github.com/iflow-ai/iflow-cli/blob/main/docs_en/features/memory-import.md

Example of importing different configuration files based on the environment.

```markdown
# Development environment configuration
@./configs/development.md

# Production environment configuration
@./configs/production.md
```

--------------------------------

### Initialize and Configure Frontend Development Environment

Source: https://github.com/iflow-ai/iflow-cli/blob/main/docs_en/scenarios.md

Commands to initialize a project and install the Playwright MCP tool for frontend tasks.

```bash
# 1. Project initialization
cd my-react-project
iflow
> /init

# 2. Install frontend-related MCP tools
!iflow mcp add-json 'playwright' '{"command":"npx","args":["-y","@iflow-mcp/playwright-mcp"]}'

# 3. Common task examples
> Create a responsive user card component with dark mode support
> Help me optimize the rendering performance of this React component
> Analyze accessibility issues on the current page
```

--------------------------------

### Large Text Input Examples

Source: https://github.com/iflow-ai/iflow-cli/blob/main/docs_en/features/interactive.md

Examples of pasting large text blocks into the CLI for analysis or refactoring.

```bash
> Please help me refactor this code: [Pasted text #1 +45 lines]
> Analyze the errors in this log file: [Pasted text #2 +120 lines]
```

--------------------------------

### Learn React Hooks and Docker

Source: https://github.com/iflow-ai/iflow-cli/blob/main/docs_en/scenarios.md

Initiate learning plans for technologies like React Hooks and understand concepts like Docker with practical examples.

```bash
# Technology learning
iflow
> I want to learn React Hooks, please give me a complete learning plan
> Explain what Docker is and provide practical examples
> Compare the pros and cons of Vue and React
```

--------------------------------

### Install Global MCP Server

Source: https://context7.com/iflow-ai/iflow-cli/llms.txt

Install an MCP server globally using the `--scope user` flag, making it available to all projects. This is useful for shared development tools.

```bash
iflow mcp add-json --scope user 'server-name' '{"command":"node","args":["server.js"]}'
```

--------------------------------

### Frontend Project Ignore Rules

Source: https://github.com/iflow-ai/iflow-cli/blob/main/docs_en/configuration/iflowignore.md

Example configuration for a React or Vue project.

```bash
# .iflowignore for React/Vue project

# Build output
build/
dist/
.next/

# Dependencies
node_modules/

# Logs and cache
*.log
.cache/
.parcel-cache/

# Environment variable files
.env.local
.env.production

# Test coverage
coverage/

# IDE configuration (but keep .vscode)
.idea/
*.swp
*.swo
```

--------------------------------

### Node.js Backend Project Example

Source: https://github.com/iflow-ai/iflow-cli/blob/main/docs_en/configuration/iflow.md

An example IFLOW.md file for a Node.js backend project, detailing the technology stack, architecture, API standards, and security requirements.

```markdown
# Node.js Backend Project

## Technology Stack
- Framework: Express.js
- Database: PostgreSQL + Prisma ORM
- Authentication: JWT
- Validation: Zod
- Testing: Jest + Supertest

## Architecture Pattern
Uses layered architecture: Controller -> Service -> Repository -> Database

## API Standards
- RESTful API design
- Unified error handling middleware
- Request parameter validation using Zod
- Response format: `{ success: boolean, data?: any, error?: string }`

## Security Requirements
- All routes need appropriate authentication and authorization
- Sensitive information uses environment variables
- Input data must be validated and sanitized
- Error messages should not expose internal implementation

## Deployment Configuration
- Use Docker for containerization
- Environment variables managed through .env files
- Production environment uses PM2 for process management
```

--------------------------------

### Absolute Path Import

Source: https://github.com/iflow-ai/iflow-cli/blob/main/docs_en/configuration/iflow.md

Shows how to reference files using an absolute path starting from the root.

```markdown
@/absolute/path/to/file.md
```

--------------------------------

### Example Generated IFLOW.md Content

Source: https://github.com/iflow-ai/iflow-cli/blob/main/docs_en/configuration/iflow.md

A sample structure of an IFLOW.md file generated by the /init command.

```markdown
# Project Context

## Project Overview
This is a full-stack application based on TypeScript + React + Node.js.

## Technology Stack
- Frontend: React 18, TypeScript, Vite
- Backend: Node.js, Express, TypeScript
- Database: PostgreSQL
- Testing: Jest, React Testing Library

## Project Structure
```
src/
├── components/     # React components
├── services/      # API services
├── utils/         # Utility functions
└── types/         # TypeScript type definitions
```

## Development Standards
- Use TypeScript strict mode
- Components use functional approach
- Use ESLint + Prettier for code formatting
- Run pre-commit hooks before committing
```

--------------------------------

### Example Natural Language Prompts for iFlow CLI

Source: https://github.com/iflow-ai/iflow-cli/blob/main/README.md

Illustrative examples of natural language prompts for various tasks, including information gathering, file management, data analysis, development support, and workflow automation.

```text
> Help me find the best-rated restaurants in Los Angeles and create a 3-day food tour itinerary.
```

```text
> Search for the latest iPhone price comparisons and find the most cost-effective purchase option.
```

```text
> Organize the files on my desktop by file type into separate folders.
```

```text
> Batch download all images from this webpage and rename them by date.
```

```text
> Analyze the sales data in this Excel spreadsheet and generate a simple chart.
```

```text
> Extract customer information from these CSV files and merge them into a unified table.
```

```text
> Analyze the main architectural components and module dependencies of this system.
```

```text
> I'm getting a null pointer exception after my request, please help me find the cause of the problem.
```

```text
> Create a script to periodically backup my important files to cloud storage.
```

```text
> Write a program that downloads stock prices daily and sends me email notifications.
```

--------------------------------

### Install Commands

Source: https://github.com/iflow-ai/iflow-cli/blob/main/docs_en/examples/subcommand.md

Add commands to the CLI using the add command, with options to specify project or global scope.

```bash
# Basic syntax
iflow commands add <name or id> [--scope project|global]

# Install to project (default)
iflow commands add 123
iflow commands add 123 --scope project

# Install to global
iflow commands add commit --scope global

# Actual examples
iflow commands add 456 --scope project   # Code review tool
iflow commands add docs --scope global    # General documentation generation tool

# View more details
iflow commands -h
```

--------------------------------

### Install Agents via CLI

Source: https://context7.com/iflow-ai/iflow-cli/llms.txt

Install agents using the `iflow agent add` command, specifying the agent name and scope (project or global). List and remove agents with corresponding commands.

```bash
iflow agent add python-expert --scope project
```

```bash
iflow agent add code-reviewer --scope global
```

```bash
iflow agent list
```

```bash
iflow agent remove <name>
```

--------------------------------

### Streaming Response Example with IFlowQuery in Kotlin

Source: https://github.com/iflow-ai/iflow-cli/blob/main/docs_cn/sdk/sdk-android.md

This example demonstrates how to handle streaming responses from iFlow using the queryStream method. It prints chunks of text as they are received and indicates when the stream is complete.

```kotlin
import com.iflow.sdk.query.IFlowQuery

fun streamExample() {
    // 流式响应
    IFlowQuery.queryStream(
        text = "解释一下什么是升华现象",
        onChunk = { chunk -> print(chunk) },
        onComplete = { reason -> println("\n完成!") }
    )
}
```

--------------------------------

### DevOps Engineer Quick Start

Source: https://github.com/iflow-ai/iflow-cli/blob/main/docs_en/scenarios.md

Commands for optimizing deployment scripts and configuring infrastructure monitoring.

```bash
# 1. Deployment script optimization
iflow
> Analyze existing Docker configuration and provide optimization suggestions
> Create CI/CD pipeline configuration
> Design monitoring and alerting strategies
```

--------------------------------

### Agent Collaboration Example

Source: https://github.com/iflow-ai/iflow-cli/blob/main/docs_en/examples/subagent.md

Demonstrates how multiple agents can be chained together to perform sequential tasks, such as code review, performance optimization, and documentation generation.

```bash
# First perform code review
$code-reviewer Check current code for security issues

# Then optimize performance
$performance-expert Optimize code performance based on review results

# Finally generate documentation
$doc-writer Generate complete documentation for the optimized code
```

--------------------------------

### Manage Agent Installations

Source: https://github.com/iflow-ai/iflow-cli/blob/main/docs_en/examples/subagent.md

Commands for adding, listing, removing, and inspecting agents at project or global scopes.

```bash
# Add project-level Agent
iflow agent add <agent-name-or-id> --scope project

# Add user-level Agent (global scope)
iflow agent add <agent-name-or-id> --scope global

# Actual examples
iflow agent add python-expert --scope project
iflow agent add code-reviewer --scope global

# Other management commands
iflow agent list              # List all configured Agents
iflow agent remove <name>     # Remove specified Agent
iflow agent get <name>        # View Agent details
iflow agent online            # Browse online Agent marketplace
```

--------------------------------

### Frontend Development Use Case

Source: https://github.com/iflow-ai/iflow-cli/blob/main/docs_en/examples/subagent.md

Example of using the '$frontend-developer' agent to create a responsive user login component with form validation.

```bash
$frontend-developer Create a responsive user login component with form validation
```

--------------------------------

### Configuration Format

Source: https://github.com/iflow-ai/iflow-cli/blob/main/docs_en/examples/hooks.md

Details how to add the `hooks` configuration item in the `settings.json` file and provides an example of the overall structure.

```APIDOC
## Configuration Format

### Description
Add the `hooks` configuration item in the `settings.json` file.

### Example Configuration
```json
{
  "hooks": {
    "PreToolUse": [
      {
        "matcher": "tool_pattern",
        "hooks": [
          {
            "type": "command",
            "command": "your_command",
            "timeout": 30
          }
        ]
      }
    ],
    "PostToolUse": [
      {
        "matcher": "another_pattern",
        "hooks": [
          {
            "type": "command",
            "command": "cleanup_command"
          }
        ]
      }
    ],
    "SetUpEnvironment": [
      {
        "hooks": [
          {
            "type": "command",
            "command": "python ~/.iflow/hooks/env_enhancer.py",
            "timeout": 30
          }
        ]
      }
    ],
    "Stop": [
      {
        "hooks": [
          {
            "type": "command",
            "command": "echo 'Session ended'"
          }
        ]
      }
    ],
    "SubagentStop": [
      {
        "hooks": [
          {
            "type": "command",
            "command": "cleanup_subagent.sh"
          }
        ]
      }
    ],
    "SessionStart": [
      {
        "matcher": "startup",
        "hooks": [
          {
            "type": "command",
            "command": "echo 'Session initialized'"
          }
        ]
      }
    ],
    "SessionEnd": [
      {
        "hooks": [
          {
            "type": "command",
            "command": "python ~/.iflow/hooks/session_summary.py"
          }
        ]
      }
    ],
    "UserPromptSubmit": [
      {
        "matcher": ".*sensitive.*",
        "hooks": [
          {
            "type": "command",
            "command": "python ~/.iflow/hooks/content_filter.py"
          }
        ]
      }
    ],
    "Notification": [
      {
        "matcher": ".*permission.*",
        "hooks": [
          {
            "type": "command",
            "command": "logger 'iFlow permission request'"
          }
        ]
      }
    ]
  }
}
```
```

--------------------------------

### Data Analyst Quick Start

Source: https://github.com/iflow-ai/iflow-cli/blob/main/docs_en/scenarios.md

Commands for analyzing data files and generating reports or visualizations.

```bash
# 1. Data file analysis
iflow
> @data/sales_2024.csv Analyze this sales data and identify main trends
> Generate Python script for monthly sales reports
> Create interactive data visualization charts
```

--------------------------------

### Global Command Directory Structure

Source: https://github.com/iflow-ai/iflow-cli/blob/main/docs_en/examples/subcommand.md

Example of the directory structure for global iFlow CLI commands, showing TOML configuration files.

```bash
# Global command directory
~/.iflow/commands/
├── my-command.toml
├── code-reviewer.toml
└── project-analyzer.toml
```

--------------------------------

### Initialize Project Documentation

Source: https://github.com/iflow-ai/iflow-cli/blob/main/docs_en/examples/basic-usage.md

Use the `/init` command to automatically analyze your codebase and generate project documentation. It identifies project type, tech stack, dependencies, and configuration files.

```bash
/init
```

--------------------------------

### Combine Initialization and Build Commands

Source: https://github.com/iflow-ai/iflow-cli/blob/main/docs_en/examples/basic-usage.md

Illustrates combining the `/init` command for project setup with `!npm run build` for building the project, followed by a natural language request for AI analysis if issues arise.

```bash
# First initialize the project
/init

# Then execute related operations
!npm run build

# If encountering issues, let AI help analyze
Please help me analyze the cause of the build failure
```

--------------------------------

### Documentation Writing Scenario

Source: https://github.com/iflow-ai/iflow-cli/blob/main/docs_cn/examples/subagent.md

Generate detailed technical documentation and usage examples for an API endpoint using the '$doc-writer' agent.

```bash
$doc-writer Write detailed technical documentation and usage examples for this API endpoint
```

--------------------------------

### Project Initialization Workflow

Source: https://github.com/iflow-ai/iflow-cli/blob/main/docs_en/examples/best-practices.md

Use this sequence to provide the AI with project context and establish a development plan.

```bash
# Standard project start process
/init                    # Analyze project structure
!git status             # Check current status
Please help me create a development plan for this project
```

--------------------------------

### Initialize Project Context

Source: https://github.com/iflow-ai/iflow-cli/blob/main/docs_en/configuration/iflow.md

Run the /init command to automatically analyze the project and generate an IFLOW.md file.

```bash
# Execute in project root directory
$ iflow
> /init

# Example output
Empty IFLOW.md created. Now analyzing the project to populate it.
Analyzing project structure...
Generating customized context...
IFLOW.md has been successfully populated with project-specific information.
```

--------------------------------

### Install iFlow CLI using npm

Source: https://github.com/iflow-ai/iflow-cli/blob/main/README.md

Install the iFlow CLI globally using npm. This command also installs all necessary dependencies for your terminal.

```shell
npm i -g @iflow-ai/iflow-cli
```

--------------------------------

### Install iFlow CLI via npm

Source: https://github.com/iflow-ai/iflow-cli/blob/main/docs_en/quickstart.md

Install the latest version of iFlow CLI globally using npm, assuming Node.js 22+ is already installed.

```shell
npm i -g @iflow-ai/iflow-cli@latest
```

--------------------------------

### Browse Interactive Marketplace

Source: https://github.com/iflow-ai/iflow-cli/blob/main/docs_en/examples/subcommand.md

Enter the interactive marketplace mode to browse, search, and install commands using keyboard shortcuts.

```bash
# Enter interactive command marketplace
/commands online

# Operation examples during browsing
# 1. Use j/k or arrow keys to browse command list
# 2. Press Enter to view details of interesting commands
# 3. Press i key to directly install commands
# 4. Press / key to search for commands with specific functionality
```

--------------------------------

### List Available Agents

Source: https://github.com/iflow-ai/iflow-cli/blob/main/docs_cn/examples/subagent.md

Use the '/agents list' command to view all installed Agents and their configurations. This is useful for verifying Agent availability and troubleshooting.

```bash
/agents list
```

--------------------------------

### Install iFlow CLI using Homebrew

Source: https://github.com/iflow-ai/iflow-cli/blob/main/README.md

Install the iFlow CLI on macOS using the Homebrew package manager. This involves tapping the repository and then installing the package.

```shell
brew tap iflow-ai/iflow-cli
brew install iflow-cli
```

--------------------------------

### List and Manage Installed Commands

Source: https://context7.com/iflow-ai/iflow-cli/llms.txt

Manage your installed slash commands using `iflow commands` commands. List all commands, remove project-specific commands, or remove globally installed commands.

```bash
/commands list
```

```bash
/commands remove code-reviewer
```

```bash
/commands remove code-reviewer --scope global
```

--------------------------------

### Initialize Existing Project with iFlow CLI

Source: https://github.com/iflow-ai/iflow-cli/blob/main/README.md

Use the `/init` command in an existing project to help iFlow understand the codebase, generate documentation, and implement solutions based on requirements.

```shell
cd project1/
iflow
> /init
> Analyze the requirements according to the PRD document in requirement.md file, and output a technical document, then implement the solution.
```

--------------------------------

### Start and Use iFlow CLI

Source: https://context7.com/iflow-ai/iflow-cli/llms.txt

Initiate the iFlow CLI in your project directory. Interact using natural language, shell commands prefixed with '!', or slash commands prefixed with '/'. File references use the '@' prefix.

```bash
cd your-project/
iflow
```

```bash
> /init                                    # Analyze and document project structure
> Help me optimize this React component    # Natural language request
> !npm test                                # Execute shell command and analyze output
> @src/App.tsx Fix the bug in this file   # Reference files with @ prefix
> /clear                                   # Clear conversation history
> /quit                                    # Exit CLI
```

```bash
iflow --prompt "Generate a Python script that calculates Fibonacci"
```

```bash
iflow -i "explain this code"
```

```bash
iflow --yolo
```

--------------------------------

### Verify iFlow CLI Installation

Source: https://github.com/iflow-ai/iflow-cli/blob/main/docs_en/quickstart.md

Run this command to confirm that the iFlow CLI has been successfully installed.

```shell
iflow --version
```

--------------------------------

### Manage MCP configurations

Source: https://github.com/iflow-ai/iflow-cli/blob/main/docs_en/examples/mcp.md

View help documentation or add MCP servers using JSON configuration files.

```bash
iflow mcp --help
```

```bash
# Basic syntax
iflow mcp add-json <name> '<json-config>'

# Example: Weather API server
iflow mcp add-json weather-api '{
  "type": "stdio",
  "command": "/path/to/weather-cli",
  "args": ["--api-key", "abc123"],
  "env": {"CACHE_DIR": "/tmp"}
}'

# Verify installation
iflow mcp get weather-api
```

--------------------------------

### List Installed Commands

Source: https://github.com/iflow-ai/iflow-cli/blob/main/docs_en/examples/subcommand.md

Display all currently installed commands using the list command or its aliases.

```bash
# List all installed commands
/commands list
/commands show      # Alias
/commands local     # Alias

# Example output:
# Installed commands:
# 
# 🌍 Global Commands (2):
```

--------------------------------

### Initialize Project Memory

Source: https://github.com/iflow-ai/iflow-cli/blob/main/docs_cn/configuration/iflow.md

Use the /init command to automatically analyze the project and generate a tailored IFLOW.md file.

```bash
# 在项目根目录执行
$ iflow
> /init

# 输出示例
Empty IFLOW.md created. Now analyzing the project to populate it.
Analyzing project structure...
Generating customized context...
IFLOW.md has been successfully populated with project-specific information.
```

--------------------------------

### Dynamic Configuration Loading

Source: https://github.com/iflow-ai/iflow-cli/blob/main/docs_en/configuration/iflow.md

Demonstrates loading different configuration files based on the environment using relative paths.

```markdown
# Environment-Related Configuration

@./config/development.md
@./config/production.md
```

--------------------------------

### Example Markdown Command with Context

Source: https://github.com/iflow-ai/iflow-cli/blob/main/docs_en/examples/subcommand.md

An example of a Markdown file for a git commit command, including context derived from git commands.

```markdown
---
description: Create a git commit
---

## Context

- Current git status: !`git status`
- Current git diff (staged and unstaged changes): !`git diff HEAD`
- Current branch: !`git branch --show-current`
- Recent commits: !`git log --oneline -10`

## Your task

Based on the above changes, create a single git commit.

```

--------------------------------

### Frontend Development Scenario

Source: https://github.com/iflow-ai/iflow-cli/blob/main/docs_cn/examples/subagent.md

Create a responsive user login component with form validation using the '$frontend-developer' agent.

```bash
$frontend-developer Create a responsive user login component, supporting form validation
```

--------------------------------

### Quick Call Examples

Source: https://github.com/iflow-ai/iflow-cli/blob/main/docs_en/examples/subagent.md

Illustrates how to use the '$' symbol to quickly call different types of agents for specific tasks. Ensure the agent type is valid and the task description is clear.

```bash
$code-reviewer Perform code review on the current project
```

```bash
$frontend-developer Create a responsive navigation component
```

```bash
$python-expert Optimize this algorithm's performance
```

```bash
$data-scientist Analyze trends in this dataset
```

--------------------------------

### Install Commands via CLI

Source: https://context7.com/iflow-ai/iflow-cli/llms.txt

Install commands directly using the `iflow commands add` command, specifying the command name or ID and the scope (project or global).

```bash
iflow commands add <name-or-id> --scope project
```

```bash
iflow commands add <name-or-id> --scope global
```

--------------------------------

### Run Project Analysis

Source: https://github.com/iflow-ai/iflow-cli/blob/main/docs_en/quickstart.md

Initializes project analysis within a directory.

```shell
# In any code project directory
cd your-project/
iflow
> /init
> Analyze the structure and main features of this project
```

--------------------------------

### Import Modular Configuration Files

Source: https://github.com/iflow-ai/iflow-cli/blob/main/docs_en/configuration/iflow.md

Use the @ syntax to include external markdown files for modular configuration management.

```markdown
# Main IFLOW.md file

@./shared/coding-standards.md

@./project-specific/architecture.md

@../global/personal-preferences.md
```

--------------------------------

### Agent Management Commands

Source: https://context7.com/iflow-ai/iflow-cli/llms.txt

Manage installed agents using slash commands within the CLI. Commands include listing agents, viewing descriptions, browsing the marketplace, and initiating installation.

```bash
/agents list
```

```bash
/agents list desc
```

```bash
/agents online
```

```bash
/agents install
```

```bash
/agents refresh
```

--------------------------------

### Create .iflowignore File

Source: https://github.com/iflow-ai/iflow-cli/blob/main/docs_en/configuration/iflowignore.md

Initialize a new .iflowignore file in the project root directory.

```bash
touch .iflowignore
```

--------------------------------

### Initialize Command Directory

Source: https://github.com/iflow-ai/iflow-cli/blob/main/docs_en/examples/subcommand.md

Create the necessary directory structure to store custom command configuration files.

```bash
# Create command directory
mkdir -p ~/.iflow/commands
```

--------------------------------

### Initialize iFlow CLI

Source: https://github.com/iflow-ai/iflow-cli/blob/main/IFLOW.md

Run this command within a project directory to scan the codebase and generate the IFLOW.md documentation file.

```text
iflow
> /init
```

--------------------------------

### AgentInfo Parsing Example

Source: https://github.com/iflow-ai/iflow-cli/blob/main/docs_en/sdk/sdk-python.md

Parse detailed information from an iFlow agent ID using the AgentInfo class. This example shows how to extract agent ID, task ID, agent index, and timestamp.

```python
from iflow_sdk import AgentInfo

# Parse information from agent ID
agent_id = "subagent-task-abc123-2-1735123456789"
agent_info = AgentInfo.from_agent_id_only(agent_id)

if agent_info:
    print(f"Agent ID: {agent_info.agent_id}")
    print(f"Task ID: {agent_info.task_id}")
    print(f"Agent Index: {agent_info.agent_index}")
    print(f"Timestamp: {agent_info.timestamp}")
```

--------------------------------

### Team Collaboration Standards Example

Source: https://github.com/iflow-ai/iflow-cli/blob/main/docs_en/configuration/iflow.md

An example IFLOW.md file outlining team development standards, including code commit conventions, branching strategy, code review points, and communication guidelines.

```markdown
# Team Development Standards

## Code Commit Standards
- Use Conventional Commits format
- Must pass lint and test before committing
- Each PR must have code review
- Important changes require technical proposal discussion

## Branching Strategy
- main: Production environment code
- develop: Development environment code  
- feature/*: Feature branches
- hotfix/*: Emergency fix branches

## Code Review Points
1. Code logic correctness
2. Performance considerations
3. Security checks
4. Test coverage
5. Documentation completeness

## Communication and Collaboration
- Discuss major architectural changes before implementation
- Communicate issues promptly in the team chat
- Conduct regular technical sharing sessions
- Keep code and documentation synchronized
```

--------------------------------

### Configure IFlowClient

Source: https://github.com/iflow-ai/iflow-cli/blob/main/docs_en/sdk/sdk-python.md

Demonstrates using default settings versus custom IFlowOptions for connection management.

```python
from iflow_sdk import IFlowClient, IFlowOptions

# Use default configuration (auto-manage process)
async with IFlowClient() as client:
    await client.send_message("Your question")
    async for message in client.receive_messages():
        # Handle messages
        pass

# Use custom configuration
options = IFlowOptions(
    url="ws://localhost:8090/acp",  # WebSocket URL
    auto_start_process=True,         # Auto-start iFlow
    timeout=30.0                     # Timeout (seconds)
)
async with IFlowClient(options) as client:
    await client.send_message("Your question")
    async for message in client.receive_messages():
        # Handle messages
        pass
```

--------------------------------

### Verify Installation Requirements

Source: https://github.com/iflow-ai/iflow-cli/blob/main/docs_en/quickstart.md

Checks Node.js version and network connectivity.

```shell
# Check Node.js version
node --version  # Requires 22+

# Check network connection
curl -I https://apis.iflow.cn/v1
```

--------------------------------

### Python Project Ignore Rules

Source: https://github.com/iflow-ai/iflow-cli/blob/main/docs_en/configuration/iflowignore.md

Example configuration for a Python project.

```bash
# .iflowignore for Python project

# Python cache
__pycache__/
*.pyc
*.pyo
*.pyd

# Virtual environments
venv/
env/
.venv/

# Build files
build/
dist/
*.egg-info/

# Jupyter Notebook checkpoints
.ipynb_checkpoints/

# Testing and coverage
.pytest_cache/
.coverage
htmlcov/
```

--------------------------------

### Troubleshooting - Connection Issues

Source: https://github.com/iflow-ai/iflow-cli/blob/main/docs_en/sdk/sdk-python.md

Guide to resolving connection issues with iFlow.

```APIDOC
## Troubleshooting - Connection Issues

### Description
Steps to diagnose and resolve common connection problems when using iFlow.

### Checks

1.  **Is iFlow installed**
    ```bash
    iflow --version
    ```

2.  **Is the port occupied**
    ```bash
    # Linux/Mac
    lsof -i :8090

    # Windows
    netstat -an | findstr :8090
    ```

3.  **Manual connection testing**
    ```bash
    iflow --experimental-acp --port 8090
    ```
```

--------------------------------

### Troubleshooting Guide

Source: https://github.com/iflow-ai/iflow-cli/blob/main/docs_en/features/slash-commands.md

Common issues, solutions, and diagnostic steps for the CLI.

```APIDOC
## Troubleshooting

### Common Issues and Solutions
| Issue | Possible Cause | Solution |
|---|---|---|
| Command not recognized | Spelling error or command doesn't exist | Check spelling, use `/help` to view available commands |
| Invalid parameters | Incorrect parameter format | View command help information, use Tab completion |
| Network connection failed | Network issues or server unavailable | Check network connection, retry later |
| Insufficient permissions | Missing necessary system permissions | Check file permissions or system settings |
| Configuration file error | Incorrect configuration file format | Reconfigure or restore default settings |

### Diagnostic Steps
1. **Basic Check**: Confirm command spelling, use `/help`, check parameter format.
2. **Network Connection**: Verify network, check firewall, confirm server accessibility.
3. **Configuration Verification**: Use `/about`, check config files, re-run configuration commands.
4. **Error Troubleshooting**: View detailed error info, use `/log`, use `/bug` to report issues.

### Platform Compatibility
| Platform | Support Level | Special Considerations |
|---|---|---|
| Windows | Full support | Paths use backslashes |
| macOS | Full support | May need system permission authorization |
| Linux | Full support | Depends on terminal environment configuration |
```

--------------------------------

### Directory Reference Examples

Source: https://github.com/iflow-ai/iflow-cli/blob/main/docs_en/features/interactive.md

Usage of the @directorypath syntax to reference entire directories.

```bash
> Analyze the utility functions in the @src/utils directory
> Refactor the component structure in the @src/components directory
```

--------------------------------

### Quick iFlow CLI Experience

Source: https://github.com/iflow-ai/iflow-cli/blob/main/docs_en/scenarios.md

Run this command for a quick experience of the iFlow CLI and to analyze the current directory.

```bash
iflow` then enter `> Hello, help me analyze the current directory`
```

--------------------------------

### Execute Task in iFlow

Source: https://github.com/iflow-ai/iflow-cli/blob/main/docs_en/features/suspending-resuming.md

Example of providing an instruction to the iFlow interactive session.

```text
>>> "Write a python hello world"
```

--------------------------------

### Effective Questioning Methods

Source: https://github.com/iflow-ai/iflow-cli/blob/main/docs_en/examples/best-practices.md

Examples of how to structure queries for better AI responses.

```text
I encountered an issue while implementing user authentication. Here's my code:
[paste relevant code]

When running tests, I get the following error:
[paste complete error message]

I'm using Node.js + Express + JWT, please help me analyze the problem and provide solutions.
```

```text
Code doesn't work, please fix it
```

--------------------------------

### Create and Configure a New Command

Source: https://github.com/iflow-ai/iflow-cli/blob/main/docs_en/examples/subcommand.md

Steps to create a new custom command by making a TOML file in the `.iflow/commands` directory and defining its description and prompt.

```bash
mkdir -p ./.iflow/commands
touch ./.iflow/commands/my-command.toml
```

```toml
description = "Test command"
prompt = "This is a test command"
```

--------------------------------

### SessionStart Hook Configuration

Source: https://github.com/iflow-ai/iflow-cli/blob/main/docs_en/examples/hooks.md

Configure a SessionStart hook to run a command when a session begins, matching specific start sources like 'startup'. This can initialize the environment or set up logging.

```json
{
  "hooks": {
    "SessionStart": [
      {
        "matcher": "startup",
        "hooks": [
          {
            "type": "command",
            "command": "echo 'New session started'"
          }
        ]
      }
    ]
  }
}
```

--------------------------------

### Generate Docker Configuration Files

Source: https://github.com/iflow-ai/iflow-cli/blob/main/docs_en/scenarios.md

Configure deployment environments by generating Docker configuration files, creating Kubernetes deployment scripts, and setting up monitoring rules.

```bash
# Deployment configuration
iflow
> Generate Docker configuration files
> Create Kubernetes deployment scripts
> Configure monitoring and alerting rules
```

--------------------------------

### Flowchart Drawing Workflow

Source: https://github.com/iflow-ai/iflow-cli/blob/main/docs_en/examples/workflow.md

Commands to install and execute the Excalidraw-based flowchart generation workflow.

```shell
iflow workflow add "excalidraw-OzctqA"
```

```shell
iflow
```

```shell
/excalidraw your drawing topic
```

--------------------------------

### AI PPT Generation Workflow

Source: https://github.com/iflow-ai/iflow-cli/blob/main/docs_en/examples/workflow.md

Commands to install and execute the PPT generator workflow.

```shell
iflow workflow add "ppt-generator-v3-OzctqA"
```

```shell
iflow
```

```shell
/ppt-generator
```