Make AI agent reference

this article contains reference information on make ai agent in the ai agents configuration tab and the make ai agents app make ai agent reference /#ai agents configuration , on the left sidebar , is where you can create and configure ai agents make ai agent reference /#make ai agents app , in {{scenario singular}} builder, is the app for using your ai agent it includes the run an agent module ai agents configuration this section outlines the fields in the ai agents configuration tab system prompt the system prompt is your agent's core instructions it outlines its purpose, behavior, and any guidelines or constraints it should follow the agent follows these instructions across all workflows the improve button on the lower right hand side is an optional prompt improvement feature use it to improve your existing prompt with ai and enable more consistent outcomes after you click improve , you can leave the field empty for an auto improvement or suggest specific improvements click save to apply the changes context in context , you can upload external information to improve the agent's knowledge base examples include internal knowledge bases or reference tables {{product name}} stores the files in context in a rag database, enabling long term ai memory current limitations for uploading in context txt, pdf, docx, csv, and json files only 20 mb maximum per file 50 files maximum per team 100 files maximum per organization 20 files maximum per ai agent you can also view your uploaded context files in the make ai agents app in the {{scenario singular}} builder system tools system tools are {{scenario plural lowercase}} your agent uses to make decisions and complete tasks these are tools the agent can always access across all workflows changing system tools impacts all {{scenario plural lowercase}} using an agent when an agent processes requests, it uses tool names and descriptions to choose the right tool for the task similarly, it uses scenario inputs and outputs docid 7odicud0ke9 tr4xwuiqn names and descriptions to choose the right data your agent uses tools at two levels system tools in the ai agents configuration tab for core tools, and additional tools in the make ai agents > run an agent module for scenario level tools testing & training testing & training is a chat interface that enables direct communication with your agent, eliminating the need to connect to external tools like telegram or slack use it to test, debug, and refine your agent's responses when you interact with your agent in testing & training , the agent can show you the tools it uses and their inputs and outputs you can click the new chat button to start a new chat make ai agents app this section outlines the make ai agents app modules and their settings the make ai agents > run an agent module allows you to create ai agents provide additional {{scenario plural lowercase}} to the agent send messages to the agent, either as additional instructions to modify its context or messages to give it tasks agent select the agent you want to use or click the create agent button to create a new agent if you want to create a new agent in the connection field, select your connection in the agent name field, fill in your agent's identification in the model field, select the language model and version provided by your ai service provider in the system prompt field, define the agent's purpose and constraints the system prompt defines the agent in all the {{scenario plural lowercase}} it uses in each {{scenario singular lowercase}} , you can further customize the agent's description with additional instructions context in context , you can view the context files the agent uses for reference thread id in thread id , you can specify a custom thread id to keep all of your interactions with the agent in the same communication thread if you leave the thread id field empty, a new thread id appears in the output additional system instructions in additional system instructions , you can share extra contextual information with the agent dynamically provide context by mapping data from the {{scenario singular lowercase}} to this field messages in messages , send tasks and specific requirements to the agent you can send multiple user messages at once timeout in timeout , set the maximum number of seconds the agent can execute until it is terminated if you leave the timeout field empty, the timeout is 300 seconds (5 minutes) the maximum timeout is 600 seconds (10 minutes) continue scenario run while agent is working enable this option to forward the agent's response to a specific url sometimes the agent might take a while to respond if the run an agent module doesn't get a response in 180 seconds, it would output the moduletimeouterror and stop the {{scenario singular lowercase}} if you think the agent could take a long time to respond, enable this option in the webhook url field, specify a webhook url if you don't have a webhook ready create a webhooks docid 1yhunj8jvzyxip9cf3ps1 add the webhook to a {{scenario singular lowercase}} set the {{scenario singular lowercase}} scheduling to immediately activate the {{scenario singular lowercase}} paste the webhook url in the webhook url field of the run an agent module when the run an agent module runs, it replies immediately with a notification that {{product name}} will forward the agent's reply to the specified webhook when the agent finishes its tasks, the custom webhook trigger receives its reply with continue scenario run while agent is working enabled, as the run an agent module doesn't receive data from the agent, the data is unavailable for mapping in the {{scenario singular lowercase}} the agent's reply is forwarded to the webhook if you want to use the agent's results in a {{scenario singular lowercase}} , you need to move the processing to the {{scenario singular lowercase}} with the webhook