Zive Agents are AI-powered systems you can set up to make your life easier and become more productive and efficient. You can see agents as your work companions for a all kinds of use cases, helping you automate tasks, support your decision making, analyzing data and much more.
Tip: This article is focused on the functionality of Zive Agents. If you are looking for ideas and examples of agents, take a look at our dedicated article Inspiration for Zive Agents.
How to create a Zive Agent
Agents have their own section which you find in your navigation bar. On the starting page you find an overview of the available agents with your favorites starred at the top. To create an agent you find the button "Create new" in the top right corner.
After the clicking the button you can choose in which space want to create the agent. This defines who can access your agent after you published it.
Your space (Personal):
Every user has the permission to create an agent in their personal space
Agents created in your personal space are only available to you
Company space:
Only users with the platform role "Moderator" can create agents in the company space
Agents created in the company space are available to all users
Team spaces (if existent):
Only users assigned as "Managers" of the respective team space can create agents in the space
Agents created in a team space are available to all members of the respective space
You can also navigate into a space and create an agent there. To learn more about Spaces and their configuration, read our dedicated help article.
In the slider that opens you can give the basic information about your new agent. You need to provide a name and a description and can select an icon an the color for your agent. You also have the possibility to define Managers that should be able to edit the agent. All of the information can be adjusted afterwards as well. Click on "Continue" to start defining the detailed settings for your new agent.
How to configure a Zive Agent
Tip: Use the preview on the right side to test how your adjustments to the configurations impact the behavior and results of your agent.
Behavior
The tab "Behavior" is the core of your agent, where you can define its purpose and task.
Instruction: Describe the identity of your agent and what you expect it to be doing or how it should behave.
Steps (optional): Whenever an agent should perform a workflow or follow a process in a specific order, you should use steps to define this.
Tip: If possible, breaking down complex tasks into steps makes it easier for the agent to understand and follow them.
Include knowledge: Provides you the possibility to add relevant context to the agent. You have different sources you can choose from:
Collection: Attach an entire knowledge collection to your agent. Especially if you have larger sets of documents you want to define as the scope for an agent (e.g. to research in), it makes sense to create a collection and attach it to to an agent afterwards.
Select from company knowledge: If you have a couple relevant content items, the agent should consider, and those items are already available in the platform.
Upload local files: Relevant files that are not indexed into the platform, but available on your device, you can quickly upload them.
Note: Local files uploaded into an agent become accessible for all users with access to this agent. Deleting an agent removes local files uploaded from the platform again.
The token usage always indicates how much of the context window is consumed by the knowledge you attached.
As long as you are below the limit all of the attached context is always loaded into the content.
As soon as the limit is exceeded the agent automatically gets equipped with the "search company knowledge" capability, and will restrict this search to only the content in the attached collection.
Capabilities
In the tab "Capabilities" you can define what the agent should be able to do.
Search company knowledge: This lets you define whether or not the agent should perform an internal search. For example there might be cases where only certain content or knowledge needs to be considered very strictly. Here it could make sense to attach this in the Knowledge tab and deactivate the search capability.
Search web: Allows the agent to search relevant information from the web. Should be activated if the agent needs external, up-to-date information in order to perform its task.
Data analyst: This enables the agent to execute code, e.g. in order to create graphs or analyze data. Whenever you expect your agent to perform complex tasks or solve complex problems this capability will be required to achieve ideal results.
Tip: More is not always better! There might be simple use cases where deselecting a capability can lead to better results. Also be aware that response times of your agent can be impacted by the capabilities.
Additionally you can select which model should be used by the agent.
No specific model: Gives users of the agent the possibility to select the model they would like to use individually, just like with the Zive Assistant.
Auto modus: Forces the auto modus and removes the possibility for users of the agent to switch to a specific model.
One specific model: Also here you set the model for the agent and only this model will be used. Users are not able to adjust the model when using the agent.
Tip: If the agent is instructed to a specific task or use case, usually it makes sense to test which model delivers the best results and define this for all users. This way you can ensure consistent quality answers for users of the agent.
Appearance & visibility
Here you can adjust the look and feel and the settings you While most of the settings at the tops are pretty straight forward two things to highlight:
Conversation starters: Conversation starters are extremely valuable from a user perspective, especially if you create an agent for your team or the entire company. Always keep in mind that other users usually don't have any access to the settings of the agent. Even with configured welcome message it's hard for them to understand what exactly the agent can do and how they need to phrase their prompt.
Tip: When configuring agents take the perspective of a user who sees it for the first time and is interested in trying it out. Is it clear what the agent can do and how to get started?
Access settings
Hidden: This toggle gives you the possibility to hide your agent while you are setting it up or testing it. While this is toggled on, the agent is only visible to the creator and Managers (if set).
Managers: Through adding users as managers for an agent you allow them to edit the agent you created. This is great if you want to collaborate with others on creating an agent. Deleting an agent is only possible for the creator of it.
Space: You can always adjust in which space an agent belongs. This allows you to quickly adjust the visibility and accessibility to a new user group.
How to delete an agent
Note: Agents can only be deleted by Moderators.
When entering the settings of an agent you find a button "Delete" at the bottom left. You will be asked to confirm this action as once deleted an agent can't be restored. So think twice before doing this.
How to use Zive Agents
Organize your agents
Users can find all created agents that are not hidden as soon as they select "Agents" in the navbar on the left. Users can mark agents as a favorite by clicking the star icon, which saves them in the section "Your favorite agents" at the top of the page. This allows easy access to the most relevant and most frequently needed agents.
Work with agents
You can easily access an agent by clicking on it. If you already used the agent it will bring you back to your last conversation. In the top right corner you find a button to start a new thread. You can then use one of the prompts that were predefined as conversation starters or enter your prompt at the bottom.
The agent will perform provide an answer to your request following the instructions that were defined in the settings.
Attach files to an agent conversation
There might be use cases that require you to add a relevant file to a conversation with a Zive agent. You can always do this by clicking on "Attach" below the input field. Here you can select to search for a file within Zive to add it or upload a file from your local device. It works in the same way as the Prompt Attachment feature in the Zive Assistant.
Note: Files attached to a thread are not indexed into the Zive platform and not accessible for other users.
Manage your agent threads
All your conversations with an agent are saved, to enable you to jump back into a thread whenever you need it. Your most recent threads are displayed on the left side to allow quick access. For a full list just click on "Show all" which opens a slider where you find all your conversation with this agent. Here you also have the possibility to delete threads from the past.
