google agents-cli for ADK - it is worth learning this

This is quite a handy tool that we can use to work with ADK. It is newer than the adk tool and its repo can be found here https://google.github.io/agents-cli/guide/quickstart-tutorial/

Let's get started. 

To create an agentic scaffold quickly (this will create basic agentic code, your private virtual environment, docker file, pyproject.toml, without CICD)

agents-cli create my-agent --prototype --yes

To create use agentic ADK, uses agentic_runtime (you can choose cloud run or GKE) and with github as the CICD

agents-cli create my_agent2 -d agent_runtime  --cicd-runner github_actions --agent adk

If you would like to get a base template, you can run the following command 

agents-cli create my_agent2 -d agent_runtime  --cicd-runner github_actions --agent adk

To add a new deployment target 🔥and other examples:-

agents-cli scaffold enhance --deployment-target cloud_run

# Add Cloud Run deployment

agents-cli scaffold enhance --deployment-target cloud_run

# Add a RAG datastore

agents-cli scaffold enhance --datastore agent_platform_search

# Preview changes without applying

agents-cli scaffold enhance --deployment-target cloud_run --dry-run

To deploy it 

agents-cli deploy

Running evaluation 

# Run a specific eval set agents-cli eval run --evalset tests/eval/evalsets/custom.evalset.json # Run all eval sets in the project agents-cli eval run --all