ae

ae ("awscli extended") is a CLI tool aiming to simplify various AWS tasks, taking inspiration from kubectl, docker, docker-compose etc.

Features:

• Fast and consistent interface for:
  • Connecting and getting an interactive shell on instances/containers
  • Running commands on multiple instances/containers
  • Forwarding ports to instances/containers
• Viewing logs from one or more Cloudwatch log groups
• Easily using multiple AWS accounts and assumed IAM roles
• Extendible by adding custom subcommands from local files or pip packages

Installation

Requirements:

• Python and pip
• The AWS CLI
• AWS CLI Session manager plugin

Install or upgrade using pip:

• pip install --upgrade git+ssh://git@github.com/leigholiver/ae.git

Quick start

$ ae [-p/--profile <aws cli profile>] <command> [--help]

• Connect to an EC2 instance or ECS task container
  • ae connect <name>
• Run a command on an instance/container
  • ae run-command <name> -- command
• Forward a local port to an instance/container
  • ae port-forward <name> <local port>:<remote port>
• View CloudWatch logs from one or more log groups
  • ae logs [-f] <name> [<name>...]
• Select an AWS CLI profile to use
  • ae -p <aws cli profile> <command>

Full usage instructions can be found in the documentation

Nix

Add this repo to your flake inputs

inputs = {
  ae.url = "github:leigholiver/ae";
};

If you have nixpkgs or poetry2nix in your flake inputs, you can pin ae's dependencies to your versions:

inputs = {
  ae = {
    url = "github:leigholiver/ae";
    inputs.nixpkgs.follows = "nixpkgs";
    inputs.poetry2nix.follows = "poetry2nix";
  };
};

Install the package

# nixos
environment.systemPackages = [ inputs.ae.packages.${pkgs.stdenv.hostPlatform.system}.default ];
# home-manager
home.packages = [ inputs.ae.packages.${pkgs.stdenv.hostPlatform.system}.default ];

dev/build

The flake provides a dev shell which can be entered with nix develop. In the dev shell, running ae will use the local dev version.

To build the package, run nix build, and the output will be placed in the result/ directory.