Nebula AI CLI Assistant

A powerful, multi-provider AI assistant designed specifically for developers. Nebula supports multiple AI providers and offers advanced features like context management, interactive chat, code analysis, and more.

🚀 Features

Multi-Provider AI Support

• OpenAI (GPT-4, GPT-4o, GPT-3.5-turbo)
• Anthropic (Claude 3.5 Sonnet, Claude 3 Opus)
• Google (Gemini 1.5 Pro, Gemini 1.5 Flash)
• Groq (Llama 3.1, Mixtral - Ultra-fast inference)
• Ollama (Local models - Free, no API key needed)
• Cohere (Command R+, Command R)
• Mistral AI (Mistral Large, Medium, Small)

Advanced Context Management

• Add files and directories to conversation context
• Automatic project context detection (package.json, README, etc.)
• Git integration (branch, commits, diff analysis)
• Environment detection and tool availability

Developer-Focused Features

• Code Analysis: Explain, debug, review, and optimize code
• Interactive Chat: Persistent conversation sessions
• Quick Commands: Fast access to common tasks
• Template Generation: Boilerplate code for common patterns
• Search Integration: Built-in grep with ripgrep support
• File Editing: Safe find/replace operations
• GitHub Integration: PR creation and management

📦 Installation

Global Installation (Recommended)

npm install -g nebula

Local Development

git clone <repository>
cd nebula/cli
npm install
npm link

🔧 Setup

Quick Setup

After installation, run the interactive setup:

nebula provider setup

Manual Provider Configuration

OpenAI

export OPENAI_API_KEY="sk-..."
nebula provider setup openai

Anthropic Claude

export ANTHROPIC_API_KEY="sk-ant-..."
nebula provider setup anthropic

Google Gemini

export GOOGLE_API_KEY="..."
nebula provider setup google

Groq (Fast Inference)

export GROQ_API_KEY="gsk_..."
nebula provider setup groq

Ollama (Local, Free)

# Install Ollama first: https://ollama.ai
ollama pull llama3.2
nebula provider setup ollama

💬 Usage

Basic Chat

# Simple question
nebula chat "How do I optimize this React component?"

# With file context
nebula chat "Review this code" --file src/component.jsx

# With project context
nebula chat "Help me debug this issue" --context

# Interactive mode
nebula interactive

Quick Commands

# Explain code
nebula explain "const [state, setState] = useState([])"

# Debug issues
nebula debug "Getting 'Cannot read property' error" --file app.js

# Code review
nebula review --file src/utils.js

# Generate templates
nebula template react-component --file NewComponent.jsx

Context Management

# Add files to context
nebula context add src/app.js
nebula context add src/ --pattern "**/*.js" --exclude "node_modules/**"

# List current context
nebula context list

# Clear context
nebula context clear

Provider Management

# List all providers
nebula provider list

# Setup new provider
nebula provider setup anthropic

# Switch models
nebula model set gpt-4o --provider openai
nebula model list

Advanced Features

# Different chat modes
nebula chat "Optimize this function" --mode code-review
nebula chat "Explain async/await" --mode learning
nebula chat "Design a microservice" --mode architecture

# Save conversations
nebula chat "Help with deployment" --save
nebula history show

# Custom parameters
nebula chat "Creative story" --temperature 0.9 --max-tokens 2000

🛠️ Commands Reference

Core Commands

• nebula chat <prompt> - Chat with AI assistant
• nebula interactive - Start interactive chat session
• nebula status - Show current configuration

Provider Management

• nebula provider setup [provider] - Configure AI provider
• nebula provider list - List all available providers
• nebula model set <model> - Set default model
• nebula model list [provider] - List available models

Context Management

• nebula context add <path> - Add file/directory to context
• nebula context remove <path> - Remove from context
• nebula context list - Show current context
• nebula context clear - Clear all context

Quick Commands

• nebula explain <code> - Explain code or concept
• nebula debug <issue> - Help debug problems
• nebula review <code> - Code review and suggestions
• nebula template [type] - Generate code templates

Utilities

• nebula grep <pattern> - Search files (uses ripgrep if available)
• nebula edit replace <file> - Safe find/replace operations
• nebula pr create - Create GitHub pull requests
• nebula history show - Show conversation history

🎯 Chat Modes

Nebula supports specialized chat modes for different use cases:

• default - General purpose assistance
• code-review - Focus on code quality and best practices
• debugging - Systematic problem-solving approach
• architecture - System design and architectural guidance
• learning - Educational explanations with examples
• devops - Infrastructure and deployment guidance

nebula chat "How to scale this service?" --mode architecture
nebula chat "Explain closures" --mode learning

📝 Templates

Generate boilerplate code for common patterns:

# Available templates
nebula template

# Generate specific template
nebula template react-component
nebula template dockerfile --file Dockerfile
nebula template github-workflow --file .github/workflows/ci.yml

Available templates:

• react-component - React functional component with TypeScript
• express-route - Express.js route handler
• python-class - Python class with docstrings
• dockerfile - Multi-stage Docker configuration
• github-workflow - GitHub Actions CI/CD pipeline
• terraform-module - Terraform module structure

🔍 Context Features

File Context

# Add single file
nebula context add src/app.js --alias "main-app"

# Add directory with filters
nebula context add src/ --pattern "**/*.{js,ts}" --exclude "*.test.js"

Project Context

Automatically detects and includes:

• Package.json information
• README content
• Programming languages used
• Framework detection (React, Vue, Express, etc.)

Git Context

When in a git repository, includes:

• Current branch
• Recent commits
• File changes (staged/unstaged)
• Git status

🔧 Configuration

Configuration is stored in your OS-specific config directory:

• macOS: ~/Library/Preferences/nebula/
• Linux: ~/.config/nebula/
• Windows: %APPDATA%/nebula/

Environment Variables

# Provider API Keys
export OPENAI_API_KEY="sk-..."
export ANTHROPIC_API_KEY="sk-ant-..."
export GOOGLE_API_KEY="..."
export GROQ_API_KEY="gsk_..."
export COHERE_API_KEY="..."
export MISTRAL_API_KEY="..."

# Ollama Configuration
export OLLAMA_HOST="http://localhost:11434"

🚀 Advanced Usage

Batch Processing

# Process multiple files
for file in src/*.js; do
  nebula review --file "$file" --raw >> review-results.txt
done

Custom Workflows

# Code review workflow
nebula context add src/
nebula chat "Review this codebase for security issues" --mode code-review --save

# Debugging workflow
nebula context add logs/error.log
nebula debug "Application crashes on startup" --context

Integration with Other Tools

# With git hooks
git diff --cached | nebula chat "Review these changes"

# With CI/CD
nebula chat "Optimize this Dockerfile" --file Dockerfile --raw

🔒 Security & Privacy

• API keys are stored locally in encrypted configuration
• No conversation data is sent to Nebula servers
• All communication is direct with your chosen AI provider
• Local models (Ollama) keep everything on your machine

🐛 Troubleshooting

Common Issues

API Key Not Working

nebula status  # Check configuration
nebula provider setup <provider>  # Reconfigure

Ollama Connection Issues

# Check if Ollama is running
curl http://localhost:11434/api/tags

# Start Ollama
ollama serve

Context Too Large

nebula context clear
nebula context add specific-file.js  # Add only relevant files

Model Not Available

nebula model list  # Check available models
nebula model set <available-model>

📚 Examples

Code Review Example

nebula context add src/auth.js
nebula review "Focus on security vulnerabilities" --mode code-review

Debugging Example

nebula context add error.log
nebula debug "Server returns 500 error on POST requests" --context

Learning Example

nebula explain "async function fetchData() { const response = await fetch('/api'); return response.json(); }" --mode learning

Architecture Example

nebula context add README.md
nebula chat "Design a scalable architecture for this project" --mode architecture

🤝 Contributing

1. Fork the repository
2. Create a feature branch
3. Make your changes
4. Add tests if applicable
5. Submit a pull request

📄 License

MIT License - see LICENSE file for details

🆘 Support

• GitHub Issues: Report bugs and request features
• Documentation: Check the wiki for detailed guides
• Community: Join discussions in GitHub Discussions

Made with ❤️ for developers by developers