Introduction
The Scriptures here on TDS, tested the foundations of Agentic AI. I have been involved with you some concepts to help you wander in the weather for the content we see out there.
In the last two articles, we examined things like:
- How to Create Your First Agent
- What are tools and how to use your agent
- Memory and Consultation
- Watch
- Agent testing and caution
It's good! If you want to learn about it, I suggest you look at articles [1] including [2] from input.
Agentic AI is one of the most hot topics yet, and there are several possible situations. Fortunately, one thing I've seen in my experience I read about agents that these basic ideas can be driven from one to another.
For example, the class Agent from one frame is chat In some cases, or even if other, but usually with the same arguments and the same purpose of connecting in a large language model (llm).
So let's take another step on our study journey.
In this post, we will learn how to create multiple job groups, to create opportunities for AI to perform our complex tasks.
Due to consistency, I will continue to use Gnobe as our frame.
Let's do.
Many agent groups
Multi-Agent Agent Group is more than a word meaning: A group with more than one agent.
But why do we need that?
Well, I have created this simple law for me to me, if agent needs to use more than 2 or 3 tools, the time to make a team. The reason for this is what two Experts working together will do much better than sincere.
When trying to create a “Swiss-Knife Agent”, the chances of seeing backwards are high. The agent will be serious about different orders and the number of tools to deal with, so it ends up the error or retrieve a low result.
On the other hand, when making agents purposes one purpose, they will need a single tool to solve the problem, so increased working and improving the result.
Coordinating this professional team, we will use the class Team From Agno, capable of assigning appropriate functions.
Let's move on and understand what we will build next.
Design
Our project will focus on the content industry for social media. We will build a team of agents that produce Instagram posts and raise the image according to the topic provided by the user.
- The user sends the Prompt for post.
- The Coordiner sends work to Writer
- To the Internet and searched at that topic.
- This page Writer Returns the Social Media Post text.
- When a Coordinator has the first effect, guides the text in Announcement The agent, so it can quickly cause the photo of the post.
Note how we separate works well, so each agent can focus on their work. The coordinator will ensure that each agent performs its work, and will work with the last good effect.
To make our team more effective, I will limit the postage title to be done about Wine and careful food. In this way, we lose the number of information needed for our agent, and we can make it very clear and focused.
Let's stop that now.
Code
First, submit the required libraries.
pip install agno duckduckgo-search google-genaiCreate a natural variable file .env And add required API API keys and any search method you use, if required. DUCKDUCKGO does not need.
GEMINI_API_KEY="your api key"
SEARCH_TOOL_API_KEY="api key"Import information libraries.
# Imports
import os
from textwrap import dedent
from agno.agent import Agent
from agno.models.google import Gemini
from agno.team import Team
from agno.tools.duckduckgo import DuckDuckGoTools
from agno.tools.file import FileTools
from pathlib import PathCreating agents
Next, we will build the first agent. It is socially and experts in gourmet food.
- Requires a
nameSimple identification is a group. - This page
roleto tell about what is exclusive. - A
descriptionto tell an agent how they can behave. - This page
toolsthat they can use work. add_name_to_instructionsSending and answer name of the agent you worked in that work.- Explains
expected_output. - This page
modelthe agent's brain. exponential_backoffincludingdelay_between_retriesProtecting too many applications in llms (Error 429).
# Create individual specialized agents
writer = Agent(
name="Writer",
role=dedent("""
You are an experienced digital marketer who specializes in Instagram posts.
You know how to write an engaging, SEO-friendly post.
You know all about wine, cheese, and gourmet foods found in grocery stores.
You are also a wine sommelier who knows how to make recommendations.
"""),
description=dedent("""
Write clear, engaging content using a neutral to fun and conversational tone.
Write an Instagram caption about the requested {topic}.
Write a short call to action at the end of the message.
Add 5 hashtags to the caption.
If you encounter a character encoding error, remove the character before sending your response to the Coordinator.
"""),
tools=[DuckDuckGoTools()],
add_name_to_instructions=True,
expected_output=dedent("Caption for Instagram about the {topic}."),
model=Gemini(id="gemini-2.0-flash-lite", api_key=os.environ.get("GEMINI_API_KEY")),
exponential_backoff=True,
delay_between_retries=2
)Now, let's create the Illultator agent. The issues are the same.
# Illustrator Agent
illustrator = Agent(
name="Illustrator",
role="You are an illustrator who specializes in pictures of wines, cheeses, and fine foods found in grocery stores.",
description=dedent("""
Based on the caption created by Marketer, create a prompt to generate an engaging photo about the requested {topic}.
If you encounter a character encoding error, remove the character before sending your response to the Coordinator.
"""),
expected_output= "Prompt to generate a picture.",
add_name_to_instructions=True,
model=Gemini(id="gemini-2.0-flash", api_key=os.environ.get("GEMINI_API_KEY")),
exponential_backoff=True,
delay_between_retries=2
)Creating a group
Making these special specialists work together, we need to use the class Agent. We give it to the name and use argument To find the form of communication that the team will have. Agno Makes Finding Ways coordinate, route either collaborate.
Also, don't forget to use share_member_interactions=True Ensure that responses will flow well between agents. You can also use enable_agentic_contextThat makes the context of the team shared with group members.
Quarrel monitoring Beautiful if you want to use Agno's built-in Qashboard, available
# Create a team with these agents
writing_team = Team(
name="Instagram Team",
mode="coordinate",
members=[writer, illustrator],
instructions=dedent("""
You are a team of content writers working together to create engaging Instagram posts.
First, you ask the 'Writer' to create a caption for the requested {topic}.
Next, you ask the 'Illustrator' to create a prompt to generate an engaging illustration for the requested {topic}.
Do not use emojis in the caption.
If you encounter a character encoding error, remove the character before saving the file.
Use the following template to generate the output:
- Post
- Prompt to generate an illustration
"""),
model=Gemini(id="gemini-2.0-flash", api_key=os.environ.get("GEMINI_API_KEY")),
tools=[FileTools(base_dir=Path("./output"))],
expected_output="A text named 'post.txt' with the content of the Instagram post and the prompt to generate an picture.",
share_member_interactions=True,
markdown=True,
monitoring=True
)Let's race.
# Prompt
prompt = "Write a post about: Sparkling Water and sugestion of food to accompany."
# Run the team with a task
writing_team.print_response(prompt)This answer.
This is how the text file looks like.
- Post
Elevate your refreshment game with the effervescence of sparkling water!
Forget the sugary sodas, and embrace the crisp, clean taste of bubbles.
Sparkling water is the ultimate palate cleanser and a versatile companion for
your culinary adventures.
Pair your favorite sparkling water with gourmet delights from your local
grocery store.
Try these delightful duos:
* **For the Classic:** Sparkling water with a squeeze of lime, served with
creamy brie and crusty bread.
* **For the Adventurous:** Sparkling water with a splash of cranberry,
alongside a sharp cheddar and artisan crackers.
* **For the Wine Lover:** Sparkling water with a hint of elderflower,
paired with prosciutto and melon.
Sparkling water isn't just a drink; it's an experience.
It's the perfect way to enjoy those special moments.
What are your favorite sparkling water pairings?
#SparklingWater #FoodPairing #GourmetGrocery #CheeseAndWine #HealthyDrinks
- Prompt to generate an image
A vibrant, eye-level shot inside a gourmet grocery store, showcasing a selection
of sparkling water bottles with various flavors. Arrange pairings around
the bottles, including a wedge of creamy brie with crusty bread, sharp cheddar
with artisan crackers, and prosciutto with melon. The lighting should be bright
and inviting, highlighting the textures and colors of the food and beverages.After we have a text file, we can go to any LLM that we like to create pictures, and copy and paste Prompt to generate an image.
And here is the MOCKUP for the post office.
Good, I was going to say. What do you think?
Before you go
In this post, we took another step to study with Agentic Ai. This article is hot, and there are many situations found in the market. I just stopped trying to read them all and I chose one start actually building something.
Here, we were able to conven the construction of social media spaces. Now, all we have to do is select the theme, reopen, and then hold the group. After that, everything is still going to social media and creates posts.
Of course, there is an automated can be done in this flow, but it is outdoor here.
About construction agents, I recommend that you take prepared structures to start, and as you need to customize it, you can add to Langgraph, for example, allowing.
Communication and availability of internet
If you have loved this content, find out some of my activities and social media on my website:
GitHub Repository
Progress
[1. Agentic AI 101: Starting Your Journey Building AI Agents]
[2. Agentic AI 102: Guardrails and Agent Evaluation]
[3. Agno]
[4. Agno Team class]
