feat: support multiple A2AStarletteApplication on a single port

samples/python/agents/a2a_multi_agent_on_single_port/README.md

@@ -0,0 +1,85 @@
+# Example: Using a2a-python SDK Without an LLM Framework
+
+This repository demonstrates how to set up and use the [a2a-python SDK](https://github.com/google/a2a-python) to create a simple server and client, without relying on any agent framework.
+
+## Overview
+
+- **A2A (Agent-to-Agent):** A protocol and SDK for communication with AI Agents.
+- **This Example:** Shows how to support multiple A2AStarletteApplication instances or AgentExecutor implementations on a single port.
+
+## Prerequisites
+
+- Python 3.13+
+- [uv](https://github.com/astral-sh/uv) (for fast dependency management and running)
+- An API key for Gemini (set as `GEMINI_API_KEY`)
+
+## Installation
+
+1. **Clone the repository:**
+
+   ```bash
+   git clone <this-repo-url>
+   cd <repo-directory>
+   ```
+
+2. **Install dependencies:**
+
+   ```bash
+   uv pip install -e .
+   ```
+
+3. **Set environment variables:**
+
+   ```bash
+   export GEMINI_API_KEY=your-gemini-api-key
+   ```
+
+   Or create a `.env` file with:
+
+   ```sh
+   GEMINI_API_KEY=your-gemini-api-key
+   ```
+
+## Running the Example
+
+### 1. Start the Server
+
+```bash
+uv run --env-file .env python -m src.no_llm_framework.server.__main__
+```
+
+- The server will start on port `9999`.
+
+### 2. Run the Client
+
+In a new terminal:
+
+```bash
+uv run --env-file .env python -m src.no_llm_framework.client --question "What is A2A protocol?"
+```
+
+- The client will connect to the server and send a request.
+
+### 3. View the Response
+
+- The response from the client will be saved to [`response.xml`](./response.xml).
+
+## File Structure
+
+- `src/no_llm_framework/server/`: Server implementation.
+- `src/no_llm_framework/client/`: Client implementation.
+- `response.xml`: Example response from the client.
+
+## Troubleshooting
+
+- **Missing dependencies:** Make sure you have `uv` installed.
+- **API key errors:** Ensure `GEMINI_API_KEY` is set correctly.
+- **Port conflicts:** Make sure port 9999 is free.
+
+## Disclaimer
+
+Important: The sample code provided is for demonstration purposes and illustrates the mechanics of the Agent-to-Agent (A2A) protocol. When building production applications, it is critical to treat any agent operating outside of your direct control as a potentially untrusted entity.
+
+All data received from an external agent—including but not limited to its AgentCard, messages, artifacts, and task statuses—should be handled as untrusted input. For example, a malicious agent could provide an AgentCard containing crafted data in its fields (e.g., description, name, skills.description). If this data is used without sanitization to construct prompts for a Large Language Model (LLM), it could expose your application to prompt injection attacks.  Failure to properly validate and sanitize this data before use can introduce security vulnerabilities into your application.
+
+Developers are responsible for implementing appropriate security measures, such as input validation and secure handling of credentials to protect their systems and users.

samples/python/agents/a2a_multi_agent_on_single_port/pyproject.toml

@@ -0,0 +1,24 @@
+[project]
+name = "no-llm-framework"
+version = "0.1.0"
+description = "Use A2A without any agent framework"
+readme = "README.md"
+authors = [{ name = "prem", email = "[email protected]" }]
+requires-python = ">=3.13"
+dependencies = [
+    "a2a-sdk>=0.3.0",
+    "asyncclick>=8.1.8",
+    "colorama>=0.4.6",
+    "fastmcp>=2.3.4",
+    "google-genai",
+    "jinja2>=3.1.6",
+    "rich>=14.0.0",
+    "starlette>=0.47.2"
+]
+[build-system]
+requires = ["hatchling"]
+build-backend = "hatchling.build"
+
+[project.scripts]
+a2a-server = "no_llm_framework.server.__main__:main"
+a2a-client = "no_llm_framework.client.__main__:main"

samples/python/agents/a2a_multi_agent_on_single_port/src/no_llm_framework/client/__main__.py

@@ -0,0 +1,50 @@
+import asyncio
+from typing import Literal
+
+
+import asyncclick as click
+import colorama
+from no_llm_framework.client.agent import Agent
+
+
+@click.command()
+@click.option('--host', 'host', default='localhost')
+@click.option('--port', 'port', default=9999)
+@click.option('--mode', 'mode', default='streaming')
+@click.option('--question', 'question', required=True)
+async def a_main(
+    host: str,
+    port: int,
+    mode: Literal['completion', 'streaming'],
+    question: str,
+):
+    """Main function to run the A2A Repo Agent client.
+
+    Args:
+        host (str): The host address to run the server on.
+        port (int): The port number to run the server on.
+        mode (Literal['completion', 'streaming']): The mode to run the server on.
+        question (str): The question to ask the Agent.
+    """  # noqa: E501
+    agent_index = 1
+    agent = Agent(
+        mode='stream',
+        token_stream_callback=None,
+        agent_urls=[f'http://{host}:{port}/{agent_index}'],
+    )
+    async for chunk in agent.stream(question):
+        if chunk.startswith('<Agent name="'):
+            print(colorama.Fore.CYAN + chunk, end='', flush=True)
+        elif chunk.startswith('</Agent>'):
+            print(colorama.Fore.RESET + chunk, end='', flush=True)
+        else:
+            print(chunk, end='', flush=True)
+
+
+def main() -> None:
+    """Main function to run the A2A Repo Agent client."""
+    asyncio.run(a_main())
+
+
+if __name__ == '__main__':
+    main()