Manage AI agents

when you create multiple ai agents, managing them through scenarios can become quite a challenge to make managing ai agents easier, we added a new tab to the {{product name}} platform navigation called ai agents the ai agents tab shows you the list of ai agents that available in your team ai agents are shared across all team members, like connections if you want to have a private agent, use a team where you are the only member in the ai agents tab, you can c reate and duplicate agents c onfigure agents d elete agents create agents to create a new ai agent, click the button create agent in the top right corner a create agent pop up appears in the connection , select your ai service provider connection or click add create a connection pop up appears in the connection type drop down, select your ai service provider fill in the rest of the form according to the documentation for creating a connection to the ai service provider check the list of ai service providers https //www make com/en/help/ai in make click save to finish creating the connection in the agent name field, specify name for the agent in the model field, select the ai service model that the agent should use for reasoning in the system prompt field, define the agent's purpose and constraints we recommend providing a short system prompt for the agent and leverage the additional system instructions to customize the agent for a specific {{scenario singular lowercase}} we recommend starting with a short system prompt and checking your resource (token) usage in your ai service provider if your agent is using a lot of tools or is processing long descriptions your ai service usage might increase click save to confirm agent settings next, to add the agent's tools and context , click configure on the right hand side of your agent in the make ai agents tab in context , click add to upload files that the agent can use context allows you to add external information that enhances the agent's knowledge base, enabling it to effectively achieve its goal examples include internal knowledge bases or reference tables the files in context are stored in a database and enable long term ai memory current limitations for uploading in context txt, pdf, docx, csv, and json files 20 mb maximum per file 50 files maximum per team 100 files maximum per organization 20 files maximum per ai agent in system tools , click add to select the {{scenario plural lowercase}} the agent can use click save you have created an ai agent you can now use it in a run an agent module, or you can continue configuring it duplicate agents if you need to create a new agent based on an existing agent, you can create a duplicate of an agent to set up the new agent faster to create a copy of an agent, click the three dots next to the configure button and select duplicate configure agents when using ai agents, you will need to adapt them to your processes or update them to improve their efficiency to configure an agent, go to the ai agents tab and click configure alternatively, you can click the configuration link in the run an agent module settings options for configuring or reconfiguring your agent u pdate agent's system prompt in system prompt add background information or data for the agent to use in context add {{scenario plural lowercase}} the agent can use as its tools in system tools the {{scenario plural lowercase}} you select in the agent's settings are always available to the agent, in addition to {{scenario plural lowercase}} you add in the run an agent module keep in mind that updating agent settings influences all run an agent modules where you use the agent if you would need to revert changes, you need to put them back manually to make further changes to the agent's model, click the agent settings button the pop up form allows you to c hange the name of the agent c hange the language model the agent is using if the ai service has different models i f the ai service supports configuring the model, you can change the model configuration the configuration options are available based on the agent's language model example configuration options maximum tokens the maximum amount of tokens that the agent can generate and send back to the run an agent module maximum steps the maximum number of iterations the agent can run before providing the final reply to your request maximum history the maximum number of messages that the agent adds to its context from the communication thread this setting influences the agent's context only if you specify the thread id in the run an agent module settings you can't change an agent's ai service provider or agent's connection if you need to use a different ai service provider or if you need to change agent's connection, you have to create a new agent instead delete agents to delete an agent, navigate to the agent settings, click the button with the three dots and select delete deleting an ai agent is irreversible if you remove an ai agent, all modules that use the agent stop working