Warning: This repository is deprecated and no longer maintained. The project has moved to https://github.com/BlueCentre/code-agent.

An AI coding assistant for your terminal, powered by multiple LLM providers (currently Google Gemini and Ollama).

Table of Contents

• Features
• Installation
• Setup
• Usage
• Documentation
• Contributing
• License

• Interactive chat sessions in your terminal.
• Supports multiple model providers (Google Gemini, Ollama).
• Configurable default provider and model.
• Automatic context initialization from project files.
• Markdown rendering for chat output.
• Assistant can utilize tools for:
  • File operations (view, edit, list, grep, search)
  • Directory operations (list, tree)
  • System commands (run terminal commands)
  • Quality checks (linting, formatting)
  • Running tests (e.g., pytest)

For detailed installation instructions, please see the Installation Guide.

Recommended (PyPI):

uv pip install cli-code-agent

From Source:

git clone https://github.com/BlueCentre/cli-code.git
cd cli-code
# Create a virtual environment (optional but recommended)
# uv venv
# source .venv/bin/activate

# Install in editable mode
uv pip install -e .
# For development including test dependencies, use:
# uv pip install -e '.[dev]'

Configure API credentials for your desired LLM provider. See the Installation Guide for details on setup commands and using environment variables.

Example (Gemini):

cli-code setup --provider=gemini YOUR_GOOGLE_API_KEY

Configuration is typically stored in ~/.config/cli-code/config.yaml, but can be overridden by environment variables or a custom file path.

Start an interactive chat session:

# Use default provider/model
cli-code

# Specify provider (uses provider's default model)
cli-code --provider=ollama

# Specify provider and model
cli-code --provider=ollama --model llama3

Interactive Commands:

• /exit - Exit the chat session.
• /help - Display help information.

Detailed documentation is available in the docs/ directory:

• Installation & Setup
• Contributing Guide
• Code Coverage
• Changelog
• Architecture
• Context Guidelines
• Project Brainstorm

Contributions are welcome! Please see the Contributing Guide for details on setting up a development environment and submitting pull requests.

MIT