BeeAI Agent integration to WxO with A2A[Documentation]

[BrUn3y]

Hello everyone,

I'd like to contribute to the BeeAI Framework documentation by adding an example as an alternative for integrating BeeAI with Watsonx Orchestrate (WxO) using A2A. I know an integration already exists using the BeeAI libraries, but with this new approach.

Below the documentation

Integrating a BeeAI Agent in Watsonx Orchestrate via A2A

You can connect a BeeAI agent to IBM Watsonx Orchestrate to leverage its powerful orchestration capabilities. This guide will walk you through the process of exporting your BeeAI agent and importing it into a Watsonx Orchestrate instance.

Prerequisites

Before you begin, ensure you have the following set up and ready:

• A BeeAI Agent: Your agent should be built using the A2A (Agent-to-Agent) protocol.
• BeeAI SDK: The BeeAI SDK should be installed and configured.
• WxO Instance: An active IBM Watson Orchestrate instance.
• WxO ADK: The Watson Orchestrate ADK (Application Development Kit) installed on your machine.

---

Step 1: Create Your BeeAI Agent

First, you need a BeeAI agent that communicates using the A2A protocol. If you're starting out, these resources are excellent references:

• BeeAI Agent Starter: https://github.com/i-am-bee/beeai-platform-agent-starter

---

Step 2: Expose Your Agent via a Public URL

For WxO to communicate with your agent, it must be accessible from the internet.

• Production: Deploy your agent on a server with a public IP address or a configured domain.
• Development: For development and testing, you can use a tool like ngrok to create a secure tunnel to your local machine. This will generate a temporary public URL for your agent.

---

Step 3: Create the Agent Definition File

WxO requires a YAML file to define the external agent. Create a file (e.g., my-agent.yml) with the following structure. Replace the placeholder values with your agent's specific information.

spec_version: v1
kind: external
provider: external_chat/A2A/0.2.1
name: My BeeAI Agent
title: My BeeAI Agent
nickname: My Agent
description: >
  A brief and clear description of what your agent does.
  This helps users understand its capabilities within WxO.
api_url: <YOUR_AGENT_PUBLIC_URL> # e.g., https://your-agent-domain.com or https://xxxx.ngrok-free.app
chat_params:
  agentProtocol: A2A
  stream: true

---

Step 4: Log in to Your WxO Instance

Next, use the WxO ADK to log in to your WxO environment. This command adds your instance to the ADK's environment list and activates it.

orchestrate env add -n wxo -u <YOUR_WXO_INSTANCE_URL> --type mcsp --activate

• Replace <YOUR_WXO_INSTANCE_URL> with the API URL of your WxO instance.

---

Step 5: Import the Agent into WxO

Now, import the agent definition file you created in Step 3 into your WxO environment using the following command:

orchestrate agents import -f /path/to/your/my-agent.yml

• Make sure to replace /path/to/your/my-agent.yml with the actual path to your YAML file.

---

Step 6: Verify the Import

You can verify that your agent was successfully added to your WxO environment by listing all available agents.

orchestrate agents list

Your agent should appear in the list.

---

Step 7: Finalize in the WxO UI

The final steps are completed in the Watson Orchestrate user interface:

1. Navigate to your WxO instance in your web browser.
2. Create a new, empty agent.
3. Within the new agent, use the option to import an external agent.
4. Select your newly imported BeeAI agent from the list.
5. Save the changes and begin testing the integration!