agent-engine-cli 0.1.3

Command-line interface for managing Google Cloud Vertex AI Agent Engine deployments

Agent Engine CLI

A command-line interface to manage Agent Engine.

Installation with uv

This project is set up to work seamlessly with uv.

1. Install uv (if you haven't already):

   curl -LsSf https://astral.sh/uv/install.sh | sh
   

2. Create a virtual environment and install dependencies:

   uv venv
   source .venv/bin/activate
   uv pip install -e .
   

3. Run the CLI:

   ae --help
   

Running Without Installation

You can run the CLI directly without installing it:

# Using uv (recommended)
uv run python -m agent_engine_cli.main --help

# Or with plain Python (after installing dependencies)
python -m agent_engine_cli.main --help

Example commands:

uv run python -m agent_engine_cli.main list -p PROJECT_ID -l us-central1
uv run python -m agent_engine_cli.main get AGENT_ID -p PROJECT_ID -l us-central1

Chat with an Agent

Start an interactive chat session with a deployed agent:

# Basic usage
ae chat AGENT_ID -p PROJECT_ID -l us-central1

# With custom user ID
ae chat AGENT_ID -p PROJECT_ID -l us-central1 --user my-user-id

# With debug logging enabled
ae chat AGENT_ID -p PROJECT_ID -l us-central1 --debug

Development

To run tests:

uv pip install -e ".[dev]" # or just use the dev-dependencies if using a uv-managed lockfile workflow
# If using pure uv sync:
uv sync
uv run pytest