assistant-ui + A2A Kitchen Sink

A comprehensive example showing assistant-ui connected to an A2A (Agent-to-Agent) server, demonstrating all major protocol features.

Frontend (Next.js)  -->  Bridge Backend (FastAPI)  -->  A2A Server
     :3000                    :8000                       :9999

Features

Skill	Command	A2A Features
Chat	(any message)	Streaming text via TaskStatusUpdateEvent
Artifacts	/artifacts <prompt>	Text, data, and file artifacts
Multi-step	/multistep [topic]	input-required task state
Failure	/fail	failed task state, error handling
Slow task	/slow	Long-running task, cancellation

The frontend shows a live tool UI with:

• Agent Card display (name, skills, capabilities, version)
• Task state badges with live transitions
• Streaming text updates
• Artifact rendering (code, structured data, files)
• Error states

Prerequisites

• Node.js >= 18
• Python >= 3.10
• An OpenAI API key

Setup

1. A2A Server (port 9999)

cd a2a-server
cp .env.example .env
# Edit .env and add your OPENAI_API_KEY

python -m venv .venv
source .venv/bin/activate
pip install -e .

python main.py

2. Bridge Backend (port 8000)

cd a2a-backend
cp .env.example .env

python -m venv .venv
source .venv/bin/activate
pip install -e .

python main.py

3. Frontend (port 3000)

cd frontend
npm install
npm run dev

Open http://localhost:3000. You can change the A2A server URL in the input field at the top. Use the suggestion buttons to try each skill.