dherman/amazon-q-developer-cli

Amazon Q CLI brings IDE-style autocomplete and agentic capabilities to your terminal.

😍 Features

• 🔮 Auto Completion: IDE-style completions to hundreds of popular CLIs like git, npm, docker, and aws.
• 💬 Natural Language Chat: Interact with your terminal using natural language to ask questions, debug issues, or explore the codebase.
• 🧠 Contextual Awareness: Integrates context from your local development environment, so answers are tailored to your specific code and setup.
• 🤖 Agentic Execution: Let Amazon Q take action: generate code, edit files, automate Git workflows, resolve merge conflicts, and more — with your permission.

⚡️ Installation

• macOS:
  • DMG: Download now
  • Homebrew: brew install amazon-q
• Linux:
  • Ubuntu/Debian
  • AppImage
  • Alternative Linux builds
• Windows:
  • Follow the discussions for
    Windows
  • Or use it on Windows with WSL
• Remote machines
  • Autocomplete in SSH

🚀 Start Contributing

Prerequisites

• MacOS
  • Xcode 13 or later
  • Brew

1. Clone repo

git clone https://github.com/aws/amazon-q-developer-cli-autocomplete.git

2. Setup

Hassle-free setup:

npm run setup

Or if you'd like to DIY:

sudo apt update
sudo apt install build-essential pkg-config jq dpkg curl wget cmake clang libssl-dev libgtk-3-dev libayatana-appindicator3-dev librsvg2-dev libdbus-1-dev libwebkit2gtk-4.1-dev libjavascriptcoregtk-4.1-dev valac libibus-1.0-dev libglib2.0-dev sqlite3 libxdo-dev protobuf-compiler

curl --proto '=https' --tlsv1.2 -sSf https://sh.rustup.rs | sh
rustup default stable

rustup toolchain install nightly
cargo install typos-cli

rustup target add x86_64-apple-darwin
rustup target add aarch64-apple-darwin

echo 'eval "$(mise activate zsh)"' >> "${ZDOTDIR-$HOME}/.zshrc"

echo 'eval "$(mise activate bash)"' >> ~/.bashrc

echo 'mise activate fish | source' >> ~/.config/fish/config.fish

mise trust
mise install

pnpm install --ignore-scripts

3. Start Local Development

To compile and view changes made to q chat:

cargo run --bin chat_cli

If you are working on other q commands, just append -- <command name>. For example, to run q login, you can run cargo run --bin chat_cli -- login

To run tests for the Q CLI crate:

cargo test -p chat_cli

To format Rust files:

cargo +nightly fmt

To run clippy:

cargo clippy --locked --workspace --color always -- -D warnings

💡 Quick Tip for Onboarding

Use Q CLI to help you onboard Q CLI!

Start a q chat session:

q chat

Once inside q chat, you can supply project context by adding the codebase-summary.md file:

/context add codebase-summary.md

This enables Q to answer onboarding questions like:

• “What does this crate do?”

• “Where is X implemented?”

• “How do these components interact?”

Great for speeding up your ramp-up and navigating the repo more effectively.

🏗️ Project Layout

Several projects live here:

• autocomplete - The autocomplete react app
• dashboard - The dashboard react app
• figterm - figterm, our headless terminal/pseudoterminal that
  intercepts the user’s terminal edit buffer.
• q_cli - the q CLI, allows users to interface with Amazon Q Developer from
  the command line
• fig_desktop - the Rust desktop app, uses
  tao/wry
  for windowing/webviews
• fig_input_method - The input method used to get cursor
  position on macOS
• vscode - Contains the VSCode plugin needed
  for the Amazon Q Developer for command line to work in VSCode
• jetbrains - Contains the VSCode plugin
  needed for the Amazon Q Developer for command line to work in Jetbrains IDEs

Other folder to be aware of

• build-scripts/ - Contains all python scripts to build,
  sign, and test the project on macOS and Linux
• crates/ - Contains all internal rust crates
• packages/ - Contains all internal npm packages
• proto/ -
  protocol buffer message
  specification for inter-process communication
• tests/ - Contain integration tests for the projects

Below is a high level architecture of how the different components of the app and
their IPC:

🛡️ Security

See CONTRIBUTING for more information.

📜 Licensing

This repo is dual licensed under MIT and Apache 2.0 licenses.

“Amazon Web Services” and all related marks, including logos, graphic designs, and service names, are trademarks or trade dress of AWS in the U.S. and other countries. AWS’s trademarks and trade dress may not be used in connection with any product or service that is not AWS’s, in any manner that is likely to cause confusion among customers, or in any manner that disparages or discredits AWS.