Agno AI

Agno is a lightweight library for building Multimodal Agents.

agno-workspace

Prerequisite

• uv

• docker

Install

Install Agno CLI globally.

uv tool install "agno[aws]"

Setup

Create a new workspace

[!NOTE] Agno Workspaces are standardized codebases for running Agentic Systems locally using Docker and in production on AWS. They help us manage our Agentic System as code.

ag ws create --template agent-api --name my-agents

This will create a folder named agent-app with the following structure:

agent-app                   # root directory
├── agents                  # your Agents go here
├── api                     # your Api routes go here
├── db                      # your database tables go here
├── Dockerfile               # Dockerfile for the application
├── pyproject.toml          # python project definition
├── requirements.txt        # python dependencies generated using pyproject.toml
├── scripts                 # helper scripts
├── utils                   # shared utilities
└── workspace               # Agno workspace directory
    ├── dev_resources.py    # dev resources running locally
    ├── prd_resources.py    # production resources running on AWS
    ├── secrets             # secrets
    └── settings.py         # Agno workspace settings

Setup existing workspace

If you clone the codebase directly (eg: if your coworker created it) - run ag ws setup to set it up locally

ag ws setup
# with debug logs
ag ws setup -d

Develop

Start workspace resources

Run ag ws up to start i.e. create workspace resources

ag ws up
# shorthand
ag ws up dev:docker
# full options
ag ws up --env dev --infra docker

Stop workspace resources

Run ag ws down to stop i.e. delete workspace resources

ag ws down
# shorthand
ag ws down dev:docker
# full options
ag ws down --env dev --infra docker

Other commands

# Patch workspace resources
ag ws patch
# Restart workspace
ag ws restart
# Run on AWS
ag ws up prd:aws

Examples

• Scheduling Agent with Nebius AI and Cal.com