feat: add AG2 multi-agent framework template

agent_starter_pack/agents/ag2/.template/templateconfig.yaml

@@ -0,0 +1,24 @@
+# Copyright 2026 Google LLC
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+description: "Multi-agent system powered by AG2 framework"
+settings:
+  requires_data_ingestion: false
+  deployment_targets: ["cloud_run", "none"]
+  extra_dependencies:
+    - "ag2[openai,gemini]>=0.11.4,<1.0"
+    - "python-dotenv>=1.0.0,<2.0.0"
+  tags: ["ag2"]
+  frontend_type: "None"
+example_question: "What's the weather in San Francisco?"

agent_starter_pack/agents/ag2/README.md

@@ -0,0 +1,50 @@
+# AG2 Multi-Agent Template
+
+A multi-agent system template powered by [AG2](https://ag2.ai) (formerly AutoGen), an open-source framework with 500K+ monthly PyPI downloads.
+
+## Overview
+
+This template demonstrates a ReAct-style agent built with AG2, featuring:
+- **Tool Use**: Decorator-based function registration for LLM tool calling
+- **Multi-Agent Ready**: Easy to extend with GroupChat for multi-agent orchestration
+- **Dual LLM Support**: Vertex AI (Gemini) by default, with Google API key fallback
+
+## Architecture
+
+```
+User Message
+    |
+    v
++--------------+     +---------------+
+|  UserProxy   |---->|  Assistant    |
+|  (executor)  |<----|  (LLM agent)  |
++--------------+     +---------------+
+    |                       |
+    v                       v
++----------+         +----------+
+|  Tools   |         |  Gemini  |
+|  (Python |         |  / GPT   |
+|  funcs)  |         |          |
++----------+         +----------+
+```
+
+## Quick Start
+
+```bash
+agent-starter-pack create my-agent --agent ag2
+cd my-agent
+make install
+make playground
+```
+
+## Customization
+
+- Add tools: Define functions with `@user_proxy.register_for_execution()` + `@assistant.register_for_llm()`
+- Change model: Update `llm_config` in `app/agent.py`
+- Multi-agent: Add GroupChat (see `notebooks/getting_started.ipynb`)
+
+## Resources
+
+- [AG2 Documentation](https://docs.ag2.ai)
+- [AG2 GitHub](https://github.com/ag2ai/ag2)
+- [Agent Starter Pack Docs](https://googlecloudplatform.github.io/agent-starter-pack/)

agent_starter_pack/agents/ag2/app/__init__.py

@@ -0,0 +1,15 @@
+# Copyright 2026 Google LLC
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+"""AG2 multi-agent template for Agent Starter Pack."""

agent_starter_pack/agents/ag2/app/agent.py

@@ -0,0 +1,120 @@
+# ruff: noqa
+# Copyright 2026 Google LLC
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+"""AG2 multi-agent implementation with tool use."""
+
+import os
+from typing import Annotated
+
+from dotenv import load_dotenv
+
+from autogen import AssistantAgent, UserProxyAgent, LLMConfig
+
+load_dotenv()
+{%- if not cookiecutter.use_google_api_key %}
+
+import google.auth
+
+_, project_id = google.auth.default()
+os.environ.setdefault("GOOGLE_CLOUD_PROJECT", project_id)
+os.environ.setdefault("GOOGLE_CLOUD_LOCATION", "us-central1")
+
+llm_config = LLMConfig(
+    {
+        "model": "gemini-2.5-flash",
+        "api_type": "google",
+        "project_id": os.environ["GOOGLE_CLOUD_PROJECT"],
+        "location": os.environ["GOOGLE_CLOUD_LOCATION"],
+    }
+)
+{%- else %}
+
+llm_config = LLMConfig(
+    {
+        "model": "gemini-2.5-flash",
+        "api_key": os.environ.get("GOOGLE_API_KEY", ""),
+        "api_type": "google",
+    }
+)
+{%- endif %}
+
+
+# --- Agent Setup ---
+
+assistant = AssistantAgent(
+    name="Assistant",
+    system_message=(
+        "You are a helpful AI assistant. Use the available tools to answer "
+        "questions accurately. Reply TERMINATE when the task is complete."
+    ),
+    llm_config=llm_config,
+)
+
+user_proxy = UserProxyAgent(
+    name="User",
+    human_input_mode="NEVER",
+    max_consecutive_auto_reply=5,
+    is_termination_msg=lambda x: x.get("content", "").rstrip().endswith("TERMINATE"),
+    code_execution_config=False,
+)
+
+
+# --- Tool Definitions ---
+
+
+@user_proxy.register_for_execution()
+@assistant.register_for_llm(description="Get the current weather for a given location")
+def get_weather(
+    location: Annotated[str, "The city name to get weather for"],
+) -> str:
+    """Simulates a web search. Use it get information on weather."""
+    if "sf" in location.lower() or "san francisco" in location.lower():
+        return "It's 60 degrees and foggy."
+    return "It's 90 degrees and sunny."
+
+
+# --- Entry Point ---
+
+
+def run_agent(message: str) -> str:
+    """Run the AG2 agent with the given user message.
+
+    Args:
+        message: The user's input message.
+
+    Returns:
+        The assistant's final response text.
+    """
+    assistant.reset()
+    user_proxy.reset()
+
+    response = user_proxy.run(assistant, message=message)
+    response.process()
+
+    # Extract the summary or last assistant message
+    if response.summary:
+        return response.summary.replace("TERMINATE", "").strip()
+
+    for msg in reversed(response.messages):
+        if msg.get("role") == "assistant":
+            content = msg.get("content", "")
+            return content.replace("TERMINATE", "").strip()
+
+    return "No response generated."
+
+
+if __name__ == "__main__":
+    response = run_agent("What's the weather in San Francisco?")
+    print(response)

agent_starter_pack/agents/ag2/notebooks/getting_started.ipynb

@@ -0,0 +1,179 @@
+{
+ "cells": [
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "# Getting Started with the AG2 Agent Template\n",
+    "\n",
+    "This notebook demonstrates the AG2 (formerly AutoGen) agent template from Agent Starter Pack.\n",
+    "\n",
+    "[AG2](https://ag2.ai) is an open-source multi-agent framework with 500K+ monthly PyPI downloads and 4,300+ GitHub stars. It enables building multi-agent systems where agents collaborate, use tools, and solve complex tasks.\n",
+    "\n",
+    "## What you'll learn\n",
+    "- How the AG2 template works\n",
+    "- How to customize tools and system prompts\n",
+    "- How to add multiple agents with GroupChat"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "%pip install \"ag2[openai]>=0.11.4,<1.0\" python-dotenv -q"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Basic Usage\n",
+    "\n",
+    "The template's `run_agent()` function handles agent setup and execution:"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import os\n",
+    "os.environ[\"OPENAI_API_KEY\"] = \"your-key-here\"  # Or use Vertex AI\n",
+    "\n",
+    "from app.agent import run_agent\n",
+    "\n",
+    "response = run_agent(\"What's the weather in San Francisco?\")\n",
+    "print(response)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Customizing: Adding Your Own Tools\n",
+    "\n",
+    "You can extend the agent by registering additional tools. AG2 uses a decorator pattern for tool registration:"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "from typing import Annotated\n",
+    "from autogen import AssistantAgent, UserProxyAgent, LLMConfig\n",
+    "\n",
+    "llm_config = LLMConfig(config_list=[{\n",
+    "    \"model\": \"gpt-4o-mini\",\n",
+    "    \"api_key\": os.environ.get(\"OPENAI_API_KEY\"),\n",
+    "    \"api_type\": \"openai\",\n",
+    "}])\n",
+    "\n",
+    "assistant = AssistantAgent(\n",
+    "    name=\"Assistant\",\n",
+    "    system_message=\"You are a helpful assistant. Use tools to answer questions. Reply TERMINATE when done.\",\n",
+    "    llm_config=llm_config,\n",
+    ")\n",
+    "\n",
+    "user_proxy = UserProxyAgent(\n",
+    "    name=\"User\",\n",
+    "    human_input_mode=\"NEVER\",\n",
+    "    max_consecutive_auto_reply=5,\n",
+    "    is_termination_msg=lambda x: x.get(\"content\", \"\").rstrip().endswith(\"TERMINATE\"),\n",
+    "    code_execution_config=False,\n",
+    ")\n",
+    "\n",
+    "# Register a custom tool\n",
+    "@user_proxy.register_for_execution()\n",
+    "@assistant.register_for_llm(description=\"Calculate the square of a number\")\n",
+    "def square(n: Annotated[int, \"The number to square\"]) -> int:\n",
+    "    return n * n\n",
+    "\n",
+    "result = user_proxy.run(assistant, message=\"What is 42 squared?\").process()"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Advanced: Multi-Agent GroupChat\n",
+    "\n",
+    "AG2's key differentiator is native multi-agent orchestration. Here's how to set up a GroupChat with specialized agents:"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "from autogen import GroupChat, GroupChatManager\n",
+    "\n",
+    "researcher = AssistantAgent(\n",
+    "    name=\"Researcher\",\n",
+    "    system_message=\"You research topics thoroughly and present findings.\",\n",
+    "    llm_config=llm_config,\n",
+    ")\n",
+    "\n",
+    "writer = AssistantAgent(\n",
+    "    name=\"Writer\",\n",
+    "    system_message=\"You write clear, concise content based on research. Reply TERMINATE when done.\",\n",
+    "    llm_config=llm_config,\n",
+    ")\n",
+    "\n",
+    "coordinator = UserProxyAgent(\n",
+    "    name=\"Coordinator\",\n",
+    "    human_input_mode=\"NEVER\",\n",
+    "    max_consecutive_auto_reply=0,\n",
+    "    code_execution_config=False,\n",
+    ")\n",
+    "\n",
+    "group_chat = GroupChat(\n",
+    "    agents=[coordinator, researcher, writer],\n",
+    "    messages=[],\n",
+    "    max_round=6,\n",
+    "    speaker_selection_method=\"auto\",\n",
+    ")\n",
+    "\n",
+    "manager = GroupChatManager(\n",
+    "    groupchat=group_chat,\n",
+    "    llm_config=llm_config,\n",
+    ")\n",
+    "\n",
+    "result = coordinator.run(\n",
+    "    manager,\n",
+    "    message=\"Write a brief overview of how AI is used in weather forecasting.\",\n",
+    ").process()"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Next Steps\n",
+    "\n",
+    "- Add more tools in `app/agent.py` for your specific use case\n",
+    "- Deploy to Cloud Run using `make deploy`\n",
+    "- Explore [AG2 Documentation](https://docs.ag2.ai) for advanced patterns\n",
+    "- See [AG2 GroupChat Tutorial](https://docs.ag2.ai/docs/tutorial/conversation-patterns) for complex workflows"
+   ]
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "Python 3",
+   "language": "python",
+   "name": "python3"
+  },
+  "language_info": {
+   "name": "python",
+   "version": "3.11.0"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 4
+}