Commands Reference

The AIGNE CLI is your primary tool for managing the entire lifecycle of your AI Agent projects. From creation and testing to deployment and monitoring, there's a command for every step. This page provides a high-level overview of the main commands available.

For a deep dive into each command's specific options, arguments, and usage examples, follow the links in the summary table below.

Developer Workflow#

The following diagram illustrates a typical development workflow using the AIGNE CLI commands:

Command Summary#

The AIGNE CLI provides a set of commands to streamline your development process. Here is a summary of the core commands:

Command	Description	Details
create	Initializes a new AIGNE project with a standard directory structure and configuration files.	View aigne create details
run	Starts an interactive chat loop with a specified Agent, allowing you to test its behavior in real-time.	View aigne run details
test	Executes automated tests within your project to ensure your Agent functions correctly.	View aigne test details
serve-mcp	Serves your Agent as an MCP (Model Context Protocol) server, enabling integration with external applications.	View aigne serve-mcp details
observe	Launches the AIGNE Observability server to monitor and analyze your Agent's performance data.	View aigne observe details

Global Options#

Before diving into specific commands, remember these handy global flags that you can use with the base aigne command:

• aigne --help: Displays a list of all available commands and their descriptions.
• aigne --version: Shows the currently installed version of the AIGNE CLI.

You now have a map of the AIGNE CLI's capabilities. The best way to learn is by doing.

Next, let's get started by scaffolding a new application with the aigne create command.