Echo AI CLI

🔮 An intelligent AI terminal with autonomous agents, multi-provider support, and advanced output optimization

✨ What Makes Echo Special

Echo isn't just another AI CLI - it's an intelligent terminal that thinks before it acts. With autonomous agents that analyze your prompts and optimize them automatically, Echo delivers better results with less effort.

🆕 What's New in Latest Version

✨ Interactive Welcome Experience

• Guided Setup: Beautiful CLI interface that walks you through provider configuration
• Smart Detection: Automatically detects existing configurations with ✅ indicators
• Connection Testing: Real-time API key validation during setup
• Persistent Storage: Never lose your configurations - they're saved permanently

⚡ Groq Provider - Ultra-Fast Inference

• Lightning Speed: Experience blazing-fast responses with Groq's optimized infrastructure
• Full Integration: Complete Groq SDK implementation with streaming support
• Llama Models: Access to Llama 3 70B, 8B, Mixtral 8x7B, and Gemma variants
• Production Ready: Robust error handling and authentication

🔧 Configuration Persistence

• Smart Storage: Configurations persist between CLI sessions automatically
• Flexible Updates: Keep existing API keys or update them individually
• No Re-entry: Set up once, use forever - no more repeated configuration
• Multi-Provider: Seamlessly manage multiple AI providers simultaneously

🚀 Features

• 🤖 Intelligent Agents: Autonomous prompt optimization and output enhancement
• 🌍 6 AI Providers: Claude, OpenAI, Gemini, Groq, Meta AI, and OpenRouter (100+ models)
• ⚡ Smart Routing: Automatically suggest the best provider for your task
• 💬 Interactive Chat: Real-time streaming conversations with context awareness
• 📝 Code Integration: AI-assisted file editing with diff preview and backups
• 🎯 Output Optimization: Context-aware formatting and post-processing
• 🔧 Smart Configuration: Global and project-specific settings with validation
• 🚀 Ultra-Fast: Groq integration for lightning-speed responses
• 🎪 Interactive Welcome: Beautiful guided setup experience
• 💾 Smart Persistence: Configuration saved automatically, never re-enter API keys
• 🔄 Seamless Updates: Keep existing configs or update individual providers

📦 Installation

# Install globally from NPM (recommended)
npm install -g echoai

# Or clone and build (development version)
git clone https://github.com/vijeet-shah/echo-ai-cli.git
cd echo-ai-cli
npm install
npm run build

# Make executable (replace with your project path)
chmod +x dist/cli.js

🏃‍♂️ Quick Start

1. Interactive Welcome Experience

NEW! Echo now features a beautiful interactive welcome that guides you through setup:

# Launch Echo's interactive welcome
./dist/cli.js
# or 
echoai

This opens an interactive menu where you can:

• 🚀 First-time setup: Guided provider configuration with connection testing
• ✅ Existing configs: See providers marked as "Already configured"
• 🔑 API key management: Keep existing keys or update them easily
• 💬 Direct access: Jump into chat, file editing, or agent optimization

2. Quick Provider Setup

# Interactive welcome handles setup, but you can also use:
echoai config setup

# Or manually configure (now with persistent storage!)
echoai config set claude.key sk-ant-your-api-key-here
echoai config set groq.key gsk_your-groq-key-here  # ⚡ Ultra-fast inference
echoai config set meta.key your-meta-ai-key-here
echoai config set openrouter.key sk-or-your-key-here  # 🌐 100+ models via unified API

3. Experience the Power

# Direct command with automatic optimization
echoai "explain machine learning to me"

# Ultra-fast responses with Groq
echoai "quick code snippet" --provider groq

# Interactive chat with agent optimization
echoai  # → Select "💬 Start interactive chat session"

4. Let Echo's Agents Optimize Your Prompts

# Automatic optimization and execution
echoai agents run "help me write a React component"

# See how agents would optimize your prompt  
echoai agents optimize "debug this code"

# Direct execution (agents work behind the scenes)
echoai "explain how machine learning works"

# Interactive agent testing (from welcome menu)
echoai  # → Select "🤖 Use intelligent agents for optimization"

🤖 Intelligent Agents System

Echo's breakthrough feature is its autonomous agents that analyze and optimize your prompts automatically:

Available Agents

🔧 CodeOptimizer Agent

• Optimizes prompts for programming tasks
• Suggests best providers for code generation
• Adds structure and best practices guidance
• Supports all major programming languages

✨ PromptEnhancer Agent

• Enhances general queries for better responses
• Adds context and output formatting
• Incorporates expert role-playing
• Customizes based on user preferences

Agent Commands

# List all available agents
echoai agents list

# Optimize any prompt with agents
echoai agents optimize "write a Python script to analyze data"

# Auto-optimize and execute (recommended)  
echoai agents run "help me design a database schema" --level expert --examples

# View agent optimization details
echoai agents optimize "fix my code" --format detailed

# Interactive agent exploration (recommended for beginners)
echoai  # → "🤖 Use intelligent agents for optimization" → "🧪 Test prompt optimization"

📚 Usage Examples

Multi-Provider Support

# Ultra-fast responses with Groq (⚡ NEW: Fully functional!)
echoai "quick code snippet" --provider groq

# Advanced reasoning with Claude
echoai "complex analysis task" --provider claude --model claude-3-opus

# Code generation with Meta AI  
echoai "generate Python function" --provider meta --model code-llama-70b

# Creative tasks with OpenAI
echoai "write a creative story" --provider openai --model gpt-4

# 100+ models via OpenRouter (🌐 NEW: Cost-effective unified access!)
echoai "explain quantum computing" --provider openrouter --model anthropic/claude-3.5-sonnet

# Interactive provider selection (NEW!)
echoai  # → Welcome shows "✨ Ready to go! You have X providers configured"

Intelligent File Operations

# AI-assisted file editing with agent optimization
echo edit src/components/Header.tsx --task "add TypeScript interfaces"

# Agent-optimized code review
echo agents run "review this code for security issues" --file auth.js --format detailed

# Multi-file context with smart optimization  
echo "optimize this implementation" --file src/app.js --file src/utils.js --provider claude

Advanced Agent Features

# Customize agent behavior
echo agents run "explain quantum computing" \\
  --format structured \\
  --level beginner \\
  --examples \\
  --provider claude

# Let agents choose the best provider
echo agents run "write performant database queries"
# → Agents automatically suggest Meta AI with Code Llama

# Complex optimization with confidence scoring
echo agents optimize "build a machine learning pipeline" --format detailed
# → Shows confidence scores and optimization strategies

⚙️ Configuration

Provider Setup

Echo supports 6 major AI providers:

Recommended: Use Interactive Welcome

echoai  # → Guided setup with connection testing and persistent storage

Manual Configuration (Advanced)

# Claude (Anthropic) - Best for analysis and reasoning
echoai config set claude.key sk-ant-your-key

# OpenAI - Excellent for creative and complex tasks  
echoai config set openai.key sk-your-key

# Groq - Ultra-fast inference (⚡ NOW FULLY WORKING!)
echoai config set groq.key gsk_your-key

# Meta AI - Powerful Llama and Code Llama models
echoai config set meta.key your-together-ai-key

# OpenRouter - 100+ models via unified API (🌐 NEW!)
echoai config set openrouter.key sk-or-your-key

# Gemini - Google's multimodal AI (coming soon)
echoai config set gemini.key your-key

Configuration Management (NEW!)

# View all configured providers with masked API keys
echoai  # → "⚙️ Manage providers and configuration" → "👀 View current config"

# Update existing API key while keeping other settings  
echoai  # → Select configured provider → "🔑 Update API key"

# Check configuration status
echoai  # → Welcome shows: "✨ Ready to go! You have X providers configured"

Agent Preferences

# Set default output preferences for agents
echo config set defaults.outputFormat structured
echo config set defaults.explanationLevel intermediate
echo config set defaults.includeExamples true

# Configure agent behavior
echo config set agents.enableAutoOptimization true
echo config set agents.preferredOptimizer CodeOptimizer

🏗️ Architecture

┌─────────────────┐    ┌─────────────────┐    ┌─────────────────┐
│   Echo CLI      │────│  Agent System   │────│   AI Providers  │
│   (Commander)   │    │  (Optimization) │    │   (5 Providers) │
└─────────────────┘    └─────────────────┘    └─────────────────┘
         │                       │                       │
         │              ┌─────────────────┐              │
         │──────────────│  Smart Routing  │──────────────│
         │              │  (Auto-Select)  │              │
         │              └─────────────────┘              │
         │                       │                       │
┌─────────────────┐    ┌─────────────────┐    ┌─────────────────┐
│   File Manager  │────│   Config Mgmt   │────│   Agent Store   │
│   (VS Code Int) │    │   (Multi-tier)  │    │   (Extensible)  │
└─────────────────┘    └─────────────────┘    └─────────────────┘

🤖 Agents Deep Dive

How Agents Work

1. Input Analysis: Agents analyze your prompt for context, complexity, and intent
2. Optimization: Apply domain-specific enhancements and structure
3. Provider Selection: Suggest the best AI provider for your specific task
4. Execution: Run the optimized prompt with optimal settings
5. Post-Processing: Apply formatting and enhancement to the output

Agent Scoring System

echo agents optimize "write a sorting algorithm" --format detailed

Output includes:

• Confidence Score: How well the agent understands your request (0-100%)
• Optimization Applied: List of enhancements made to your prompt
• Provider Recommendation: Best AI provider for this specific task
• Estimated Tokens: Approximate cost/usage prediction

Extending with Custom Agents

Echo's agent system is designed to be extensible. Each agent implements a simple interface:

interface Agent {
  name: string;
  capabilities: AgentCapability[];
  canHandle(context: AgentContext): boolean;
  optimize(context: AgentContext): Promise<AgentResult>;
  postProcess?(result: string, context: AgentContext): Promise<string>;
}

📊 Provider Comparison

Echo's agents automatically recommend the best provider for each task

🌐 OpenRouter: The Game Changer

NEW! OpenRouter integration provides unprecedented access to AI models:

Why OpenRouter?

• 💰 Cost Efficiency: Often 50-90% cheaper than direct provider APIs
• 🎯 Model Variety: 100+ models from 20+ providers with one API key
• ⚡ Easy Comparison: Test Claude, GPT, Llama, Gemini seamlessly
• 🔄 Unified Interface: Consistent experience across all models
• 📊 Transparent Pricing: Clear per-token costs for budget control

Popular Models Available

# Premium models at lower costs
echoai "complex task" --provider openrouter --model anthropic/claude-3.5-sonnet
echoai "creative writing" --provider openrouter --model openai/gpt-4-turbo
echoai "code generation" --provider openrouter --model meta-llama/codellama-70b-instruct

# Explore cutting-edge models
echoai "analysis task" --provider openrouter --model google/gemini-pro-1.5
echoai "fast response" --provider openrouter --model mistralai/mixtral-8x7b-instruct

Get your OpenRouter API key at openrouter.ai and access the world's largest AI model marketplace!

🛠️ Development

Prerequisites

• Node.js 18+
• npm, yarn, or pnpm

Setup

git clone https://github.com/vijeet-shah/echo-ai-cli.git
cd echo-ai-cli
npm install
npm run build
npm link

Adding New Agents

1. Create agent in src/agents/specialized/
2. Implement the Agent interface
3. Register in EchoAgentManager
4. Add tests and documentation

Development Scripts

npm run dev          # Development mode
npm test             # Run tests
npm run build        # Build for production
npm run type-check   # TypeScript validation
npm run lint         # Code linting

📖 API Reference

Core Classes

• EchoAgentManager: Manages agent registration and optimization
• ConfigManager: Handles multi-tier configuration
• ProviderManager: Manages AI provider integration
• FileManager: Provides file operations with backup

Agent Types

• AgentContext: Input context for agent optimization
• AgentResult: Optimized prompt with metadata and suggestions
• AgentCapability: Defines what an agent can do

🌟 Examples & Tutorials

Code Generation

# Agent-optimized code generation
echo agents run "create a REST API with authentication" \\
  --provider meta \\
  --model code-llama-70b \\
  --level expert

# Result: Agents add structure, best practices, and examples

Data Analysis

# Smart provider selection for data tasks
echo agents run "analyze this CSV for trends" \\
  --file sales_data.csv \\
  --format detailed

# Result: Agents choose OpenAI/Claude and add analysis framework

Creative Writing

# Creative tasks with optimal settings
echo agents run "write a compelling product description" \\
  --format creative \\
  --examples

# Result: Agents enhance prompt and suggest GPT-4

🤝 Contributing

We welcome contributions! Echo's agent system is designed for community extension.

Contribution Areas

• New Agents: Create specialized agents for different domains
• Provider Integration: Add new AI providers
• Agent Capabilities: Enhance existing agent intelligence
• Performance: Optimize agent decision-making
• Documentation: Improve guides and examples

📄 License

MIT License - see LICENSE file for details.

🙏 Acknowledgments

• OpenAI for pushing the boundaries of AI capabilities
• Groq for ultra-fast inference infrastructure
• Meta for open-source Llama models
• The community for feedback and contributions

📞 Support & Community

• 🐛 Issues: GitHub Issues
• 💬 Discussions: GitHub Discussions
• 📚 Docs: Full Documentation
• 🔮 Agent Marketplace: Coming soon!