An overview of agents and how to configure them to perform autonomous tasks within your workflows.
Researcher
agent will excel at gathering information, whereas a Writer
agent will be great at creating a proposal from that research report.Attribute | Description | Examples |
---|---|---|
Name | Describes the agent’s role. | "Content Creator" , "Market Research Analyst" |
Personality | Describe the agent’s personality and how it should behave. | "You are an expert researcher. You specialize in finding the most relevant information about a given topic." |
Agent Inputs | Add inputs that can be used by your agents. Each input has a name and value. Value can be passed from connected nodes. | Input Name: Company Website, Value: google.com |
Task | Provide the agent with a task and all necessary information for completing it. Optionally, you can reference Agent Inputs here. | "Analyze this company website and identify its key competitors." , "Write a blog post about the benefits of using AI." |
Expected Output | A detailed description of what the task’s output looks like. | "A list of competitor websites and a summary of their offerings." |
Tools | Enhance agent capabilities by giving them tools to complete the task. | Google Search , Get Linkedin Company Data |
AI Model (Settings) | Choose the LLM model for this agent. | "gpt-4o" , "claude-3.5-sonnet" |
Structured Output | Ensures the model will always generate responses that adhere to your supplied schema. | |
Loop Mode | Select a list from agent inputs. This agent will run once for each item in the provided list. |