coda-code-assistant 2025.8.22.910

A CLI-focused code assistant for AI-assisted development

Coda - AI Code Assistant

A powerful, modular AI code assistant that brings the best of AI-powered development directly to your terminal.

---

✨ What is Coda?

Coda is your AI pair programmer that lives in your terminal, built with a modular architecture that supports multiple AI providers including Oracle OCI GenAI, OpenAI, Anthropic, Google, and 100+ more via LiteLLM.

🏗️ Modular Architecture

Coda is built with a clean, three-layer architecture:

┌─────────────────────────────────────────────────┐
│                    Apps Layer                    │
│         (CLI, Web UI, Plugins, Custom)          │
├─────────────────────────────────────────────────┤
│                 Services Layer                   │
│    (Integration, Orchestration, Workflows)       │
├─────────────────────────────────────────────────┤
│                   Base Layer                     │
│ (Config, Theme, Providers, Session, Search, etc) │
└─────────────────────────────────────────────────┘

Base Modules

• 🔧 Config: Unified configuration management
• 🎨 Theme: Terminal UI and formatting
• 🤖 Providers: AI integration
• 💬 Session: Conversation persistence
• 🔍 Search: Semantic code search
• 📊 Observability: Logging and metrics

🚀 Quick Start

Try without installing

# Run directly with uvx (no installation needed!)
uvx --from coda-code-assistant coda

Install from PyPI

# Install with pip
pip install coda-code-assistant

# Or install with uv
uv pip install coda-code-assistant

# Run Coda
coda

Install from source

# Clone the repository with submodules
git clone --recursive https://github.com/djvolz/coda-code-assistant.git
cd coda-code-assistant

# Install with uv (recommended)
uv sync

# Run Coda
uv run coda

Using Individual Modules

from coda.base.config import Config
from coda.base.providers import ProviderFactory

# Initialize configuration
config = Config()

# Create a provider
factory = ProviderFactory(config.to_dict())
provider = factory.create("openai")

# Start chatting
response = provider.chat(
    messages=[{"role": "user", "content": "Hello!"}],
    model="gpt-4"
)
print(response["content"])

📖 Documentation

In-Repository Docs

• Integration Guide - How to use modules together
• Module Documentation - Individual module READMEs
• Example Applications - Working examples
• Diagram Rendering Guide - How to create diagrams

Wiki Documentation

For comprehensive documentation, visit our Wiki:

• Getting Started Guide - Installation and setup
• AI Modes - Specialized AI personalities
• Development Guide - Contributing and development workflow
• Architecture Guide - Deep dive into the modular design
• Troubleshooting - Common issues and solutions

🎯 Key Features

• 🌐 Provider Support: Works with Oracle OCI GenAI, Ollama, OpenAI, Anthropic, Google, and 100+ providers
• 🧩 Modular Design: Use only the components you need
• 💻 Terminal-First: Designed for developers who live in the command line
• 🧠 Smart AI Modes: Specialized modes for coding, debugging, explaining, and reviewing
• 💾 Session Management: Save, resume, and branch conversations
• 🎨 Beautiful Interface: Rich terminal UI with syntax highlighting
• 🔧 Tool Integration: File operations, web search, and more via MCP
• 📊 Diagram Rendering: Create flowcharts, UML diagrams, and graphs with diagram-renderer
• 🧪 Well-Tested: Comprehensive test suite ensures reliability

🛠️ Example Applications

Check out our example applications to see Coda modules in action:

• Simple Chatbot - Basic provider usage
• Session Manager - Persistent conversations
• Code Analyzer - AI-powered code analysis

🤝 Contributing

We welcome contributions! Please see our Contributing Guide for details.

Development

# Install development dependencies
uv sync --dev

# Run tests
uv run pytest

# Run type checks
uv run mypy coda

# Format code
uv run black .
uv run ruff check --fix

📄 License

This project is licensed under the MIT License - see the LICENSE file for details.

🔗 Links

• Full Documentation
• Issue Tracker
• Discussions
• Roadmap
• Architecture Documentation

---

Made with ❤️ by the Coda community