The task components allow you to give specific tasks to AI agents. Think of AI agents as additional team members for a moment. Each member has a different role, backstory, and goals. Just like a real team member, they are qualified to carry out many various tasks. Thanks to this component, you can assign specific tasks.
Do you have just one agent and one task? There’s no need to use task components. Simply use the goal field of the agent instead. Tasks are great for complex Flows consisting of several steps or giving several tasks to one agent.
What is the SelfManaged Task Component
The SelfManaged Task component allows you to define and assign tasks for agents to perform. This way, you gain more control and insight into how individual parts of a workflow are performed, which is handy for complex workflows. Breaking complex goals into subtasks also leads to more detailed and a higher-quality of output.
Tasks As A Part Of Crews
Let’s say you want to create and publish a long-form blog post. You’re likely to try to recreate a whole content team of agents. The work usually starts with an SEO specialist researching keywords and outlining the content. They’ll create an SEO brief, which will be passed on to the content writer. Once the writer is done, their colleague will proofread and edit the article to ensure quality. What about the featured images or infographics? A designer will help with that.
You already have at least three or four people working on creating the content. Each of them has a specific task to perform. Just like in a real team, you can make a team of AI agents using the SelfManaged Crew component. Within this team, each member will get a specific task assigned via the SelfManaged Task component.
Besides having their exact task in a team, one AI agent can perform several other tasks. Let’s take the content writer from the previous example. This agent’s main task is to write an article based on the SEO brief from the previous agent. In real life, a writer will have more tasks than just writing the article. For example, they may need to write a meta description and even a social media post. Putting these as subtasks of one task can lead to a more chaotic and less clear output.
Curious about the Flow we’re analyzing in this guide? It’s the Advanced Blog Generator, and you can easily find it in your Flow library.
The Difference Between Sequential And SelfManaged Tasks
You might have noticed there are two task components in your dashboard. The difference between these types is in the order of tasks and the level of control you get.
Let’s talk about Sequential Tasks first. Sequential tasks are performed one after another in a sequence (exact order) you specify. Once a task is done, it’s done, and the Flow moves on to the next agent. The sequential approach is great for linear processes that don’t require task repetition.
But this isn’t always the case in the real world. Let’s take a real-life content writer. They will first do research and move on to writing, but as the article unfolds, they may realize more research is needed. Understandably, they will go back and forth between research and writing tasks before finally moving on to the next step. But this is impossible with sequential crews and tasks. That’s where SelfManaged tasks come in.
With SelfManaged Tasks, the manager AI agent decides the order of tasks. When making decisions, the AI tries to copy traditional organizational hierarchies closely and assign tasks to the best available agent. This opens up the possibility of repeating tasks and creating multiple iterations of the final output.
Does your workflow follow a clear linear process? Try using sequential tasks instead. Learn more in this guide.
How To Use SelfManaged Tasks
SelfManaged tasks are tasks performed in a crew managed by an AI Agent. In other words, you give tasks to your AI team, and their own AI manager decides which team member will perform which task and when.
For each complex goal, there may be multiple agents and multiple tasks, all brought together by the SelfManaged Crew component. Before we create and assign tasks, we should define the agents first. That is, we should know our team before giving them tasks.
Setting up SelfManaged Tasks consists of four steps:
- Setting up individual AI agents
- Giving the agents tasks
- Setting up the manager agent
- Making the agents a crew
Setting Up Individual AI Agents
Each member of a real team has a role, goals, and a unique backstory that includes their past experiences, personality, and specific style. So does each AI agent.
For example, let’s focus on the content writer team member:
- The Role is your agent’s job title. In this example, being a content writer is the role.
- The Goal is what the agent does and what their ideal outcome is. The expected outcome for the content writer is a well-written article that adheres to the theme and SEO brief.
- The Backstory represents who the agent is. Whether you like it or not, you always bring your personality, way of thinking, vocabulary, and past experiences to anything you do. This is even more visible in creative work, such as content writing.
Go ahead and set up some agents. When creating a crew, always think of the end goal and which team members you need to achieve it. Then, create agents that represent these team members. You can create just one agent if you wish to give the agent multiple tasks.
Note: Do you have just one agent and one task? There’s no need to use task components. Simply use the goal field of the agent instead to give it the task.
In our content writing example, we’ve created the SEO researcher, copywriter, and proofreader agents. However, the agents and tasks always depend on your specific needs and processes. Feel free to break down the existing process even more or add new agents. For example, you can add a designer agent that will generate images to go along with your article.
Learn more about AI agents and how to use the AI Agent component
Giving Agents Tasks
In Crews, each Agent is assigned one or multiple tasks to perform. Like in a real team, each member can carry out various project-specific tasks. The task components allow you to specify and assign these tasks. Continuing with our blog creation example, we now know who our agent is. The next step is to let the agent know their task and introduce them to the team.
You’ll notice that, like with the Crew component, there are two possible task components— sequential and SelfManaged. Since these are two opposite approaches to managing agents, mixing them would make no sense. That’s why we’ll also use SelfManaged Tasks when using a SelfManaged Crew:
Unlike Sequential Tasks, SelfManaged Tasks are great for processes where you’re unsure about how to split a complex process into smaller subtasks. You can simply write the entire complex goal into a single task. The manager LLM is there to assign tasks and oversee the process, making sure each agent knows what to do and when. It can effortlessly split the main task and assign the parts to the correct agent.
In addition to the task, each agent in a Crew can also get appropriate tools, making their job easier and more accurate. In our example, the researcher uses the GoogleSearch and URL Retriever tools to control the research options.
How To Set Up SelfManaged Tasks
Each task must have a description and an agent responsible for the execution. Optionally, you can fill out the expected output field if you need the output in a specific structure or want to ensure something is included.
The task description for our content writer agent might go a little something like this:
“Given the SEO content brief, write a blog post in no more than 1500 words.
Never start paragraphs with vague statements such as "In the fast-changing field of...". Always go directly to the main information the paragraph should deliver. “Let’s take a closer look at this task description:
“Given the content brief” – The agent knows what to do with the previous output.
“Write a blog post of up to 1500 words” = The output we expect from the agent.
“Never start…..” = Giving additional custom instructions to tweak the output. These instructions can be any pointers on language, vocabulary, structure or anything else that will help the agent create what you need.
The expected output field is optional and works great when you need a clearly structured output or make sure something is included in the output. For example, our SEO researcher agent’s task is to create:
A brief in this form:
SEO friendly Title:
SEO friendly Meta description:
SEO friendly OutlineEnsuring it doesn’t forget to start the output with a title and meta description.
The final step is to connect all the tasks to the tasks handle of the SelfManaged Crew component. From there, the manager LLM will take over, ordering the tasks as necessary to achieve the expected output.
Remember that sequential crews work with sequential tasks only, and SelfManaged crews work with SelfManaged tasks only.
Making The Agents A Crew
Let’s go back to our Flow. It features three team member agents and a task for each, and manager LLM overseeing the entire process. The last step of creating a crew is letting the agents know they’re a team. This is where the SelfManaged Crew component comes into play.
The SelfManaged Crew Component
The SelfManaged Crew component represents a group of agents whose work is automatically managed by a manager LLM. The team manages itself, allowing for dynamic work and the ability to create several iterations. It’s essentially a way to tell agents they’re a team with a common goal.
There may be more than one independent team within your Flow, meaning more than one crew component distinguishing these teams from each other. In our example, we only use one crew, but we still need to bring the agents together in a crew.
Want to learn more about SelfManaged Crews? Refer to our guide to learn everything you need to know.
In our example, we only use one crew, but we still need to bring the agents together as a team:
Connect all the agents to the Agents handle of the SelfManaged Crew component.
That’s it. Just send it to output, and now you have a team of agents working in an exact order. Our Flow includes three agents: an SEO specialist, a content writer, and an editor.
The Flow used in this was the Advanced Blog Generator, which you can find in your Flow library.
Instead of vague output laden with telltale AI phrases, this Flow’s output will be well researched, inspired by top Google results, written according to a clear brief, and edited to avoid sounding like generic AI. Plus, using a group of agents instead of a single agent minimizes bottlenecks. It ensures that any issues can be diagnosed and resolved promptly by simply tweaking one of the agents.
