Skip to content

Python SDK for AI agent monitoring, LLM cost tracking, benchmarking, and more. Integrates with most LLMs and agent frameworks including CrewAI, Agno, OpenAI Agents SDK, Langchain, Autogen, AG2, and CamelAI

6.8 KB
Raw Blame
{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "8b2111ae",
   "metadata": {},
   "source": [
    "# Basic Agents and Teams with Agno\n",
    "\n",
    "This example demonstrates the fundamentals of creating AI agents and organizing them into collaborative teams using the Agno framework.\n",
    "\n",
    "## Overview\n",
    "\n",
    "In this example, you'll learn how to:\n",
    "- **Create specialized AI agents** with specific roles and expertise\n",
    "- **Organize agents into teams** for collaborative problem-solving\n",
    "- **Use coordination modes** for effective agent communication\n",
    "- **Monitor agent interactions** with AgentOps integration\n",
    "\n",
    "## Key Concepts\n",
    "\n",
    "### Agents\n",
    "Individual AI entities with specific roles and capabilities. Each agent can be assigned a particular area of expertise, making them specialists in their domain.\n",
    "\n",
    "### Teams\n",
    "Collections of agents that work together to solve complex tasks. Teams can coordinate their responses, share information, and delegate tasks based on each agent's expertise.\n",
    "\n",
    "### Coordination Modes\n",
    "Different strategies for how agents within a team interact and collaborate. The \"coordinate\" mode enables intelligent task routing and information sharing."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "d087e416",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Install the required dependencies\n",
    "%pip install agentops\n",
    "%pip install agno\n",
    "%pip install python-dotenv"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "39ad00cb",
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "from dotenv import load_dotenv\n",
    "\n",
    "import agentops\n",
    "from agno.agent import Agent\n",
    "from agno.team import Team\n",
    "from agno.models.openai import OpenAIChat"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "f733e281",
   "metadata": {},
   "outputs": [],
   "source": [
    "load_dotenv()\n",
    "os.environ[\"OPENAI_API_KEY\"] = os.getenv(\"OPENAI_API_KEY\", \"your_openai_api_key_here\")\n",
    "os.environ[\"AGENTOPS_API_KEY\"] = os.getenv(\"AGENTOPS_API_KEY\", \"your_agentops_api_key_here\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "fb37819a",
   "metadata": {},
   "outputs": [],
   "source": [
    "agentops.init(auto_start_session=False, tags=[\"agno-example\", \"basics\", \"agents-and-teams\"])"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e954b898",
   "metadata": {
    "vscode": {
     "languageId": "raw"
    }
   },
   "source": [
    "## Creating Agents and Teams\n",
    "\n",
    "Now let's create our specialized agents and organize them into a collaborative team:\n",
    "\n",
    "### Step 1: Create Individual Agents\n",
    "We'll create two agents with different specializations:\n",
    "- **News Agent**: Specializes in gathering and analyzing news\n",
    "- **Weather Agent**: Specializes in weather forecasting and analysis\n",
    "\n",
    "### Step 2: Form a Team\n",
    "We'll combine these agents into a team using the \"coordinate\" mode, which enables:\n",
    "- Intelligent task routing based on agent expertise\n",
    "- Information sharing between agents\n",
    "- Collaborative problem-solving\n",
    "\n",
    "### Step 3: Execute Tasks\n",
    "The team will automatically delegate tasks to the most appropriate agent(s) based on the query.\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c12702d0",
   "metadata": {},
   "source": [
    "Here's the code to implement this:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "f872be2e",
   "metadata": {},
   "outputs": [],
   "source": [
    "def demonstrate_basic_agents():\n",
    "    \"\"\"\n",
    "    Demonstrate basic agent creation and team coordination.\n",
    "\n",
    "    This function shows how to:\n",
    "    1. Create specialized agents with specific roles\n",
    "    2. Organize agents into a team\n",
    "    3. Use the team to solve tasks that require multiple perspectives\n",
    "    \"\"\"\n",
    "    tracer = agentops.start_trace(\n",
    "        trace_name=\"Agno Basic Agents and Teams Demonstration\",\n",
    "    )\n",
    "\n",
    "    try:\n",
    "        # Create individual agents with specific roles\n",
    "        # Each agent has a name and a role that defines its expertise\n",
    "\n",
    "        # News Agent: Specializes in gathering and analyzing news information\n",
    "        news_agent = Agent(\n",
    "            name=\"News Agent\", role=\"Get the latest news and provide news analysis\", model=OpenAIChat(id=\"gpt-4o-mini\")\n",
    "        )\n",
    "\n",
    "        # Weather Agent: Specializes in weather forecasting and analysis\n",
    "        weather_agent = Agent(\n",
    "            name=\"Weather Agent\",\n",
    "            role=\"Get weather forecasts and provide weather analysis\",\n",
    "            model=OpenAIChat(id=\"gpt-4o-mini\"),\n",
    "        )\n",
    "\n",
    "        # Create a team with coordination mode\n",
    "        # The \"coordinate\" mode allows agents to work together and share information\n",
    "        team = Team(\n",
    "            name=\"News and Weather Team\",\n",
    "            mode=\"coordinate\",  # Agents will coordinate their responses\n",
    "            members=[news_agent, weather_agent],\n",
    "        )\n",
    "\n",
    "        # Run a task that requires team coordination\n",
    "        # The team will automatically determine which agent(s) should respond\n",
    "        response = team.run(\"What is the weather in Tokyo?\")\n",
    "\n",
    "        print(\"\\nTeam Response:\")\n",
    "        print(\"-\" * 60)\n",
    "        print(f\"{response.content}\")\n",
    "        print(\"-\" * 60)\n",
    "\n",
    "        agentops.end_trace(tracer, end_state=\"Success\")\n",
    "\n",
    "    except Exception as e:\n",
    "        print(f\"An error occurred: {e}\")\n",
    "        agentops.end_trace(tracer, end_state=\"Error\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "ca13c9b0",
   "metadata": {},
   "outputs": [],
   "source": [
    "demonstrate_basic_agents()"
   ]
  }
 ],
 "metadata": {
  "jupytext": {
   "cell_metadata_filter": "-all",
   "main_language": "python",
   "notebook_metadata_filter": "-all"
  },
  "kernelspec": {
   "display_name": "agentops (3.11.11)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.11.11"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}