Versions Compared

Key

  • This line was added.
  • This line was removed.
  • Formatting was changed.
Note

Work in progress

The Agentic platform The AI Agent Platform allows you to create your own agents, customized to the needs of your teams, with each agent having specific instructions and tools to use.

...

  • A name, used to call the agent in the chat with “@”;

  • An avatar (or agent image) that will be displayed next to the agent’s name in the chat;

  • A brief description of what the agent is used for. This description will be displayed on the agent’s presentation card and may help users choose an agent;

  • A category that designates the context in which the agent is mostly used.

  • A series of instructions defining therole of the agent, its tasks, what data it has access to, etc. The purpose of the instructions is to guide the agent in the way they should answer;

Info

Learn how to write instructions for your agents here.

  • A model (also known as an LLM or Large Language Model), which is used to retrieve relevant information and generate the answer to the user’s query.

  • One or more data sources, including documents and external integrations (such as websites or knowledge bases), that the agent should utilize to deliver accurate and relevant information.

...

Info

Learn more about tools here.

  • One or several toolsets, which are groups of tools and/or data sources that determine what an agent can do and what data it has access to;

Info

Learn more about toolsets /wiki/spaces/AGENTIC/pages/4126670850.

  • Guardrails, which are predefined guidelines or rules ensuring that an agent operates safely and effectively.

...

Tutorial

Iframe
scrollingauto
allowfullscreentrue
srchttps://app.supademo.com/demo/cm4sljbml07nqql5j951vo2dy
width100%
framebordershow
idagents_create_agent_from_scratch
alignmiddle
langen
height500

Build your first agent

Determine your agent’s purpose

...

  • What is the primary purpose of my agent?

  • What persona should it adopt?

  • Who will be using the agent?

  • What tasks will it need to perform?

  • What data should it access?

Create a new agent

To create an agent starting from the home page, navigate to the Build > Agents menu item, then click the Create an agent button. You can also click the Create an agent button on the top right of the page.

...

You can either create a new agent from scratch or by using a template. In this example, we will create an agent

...

This page will focus on the agent’s creation from scratch by clicking the New agent/Start from scratch buttonthe button in the From scratch section.

Info

To learn how to create an agent using a template, visit this page: Create an agent using a template.

General information

Begin by providing general information about your agent, such as:

  • its name, such as Customer Support for example. The agent’s name must be short and descriptive at the same time.

  • a dedicated agent image.To edit the image,click the default logo to open the a right panel on the left where you can change the choose a background color and choose an icon among the ones provided.

  • a category, which will make it easier for users to find agents relevant to their use case. For instance, an agent with the HR type means that it is used for HR tasks.

Info

You can find the list of available categories on this page: Pre-built agents.

  • a short description of its purpose.

  • the instructions your agents should follow such as role, its tasks, what data it has access to, etc. You can write the instructions yourself or with the help of AI by clicking the Generate with AI button above the text area.

Info

Visit this page to learn how to write constructive instructions, as well as the best practices : TODOfor writing instructions: How to write instructions for your agents.

  • the AI model to use for your agent. Choose one amongst among the ones proposed in the list.

Info

Visit this page to see the available models, their characteristics as well as recommendations on choosing the right model for your tasks: How to choose an LLM.

...

Data sources

Next, you should select the data source(s), such as external integrations or documents your agent should have access . Depending on the agent’s purpose, adding data to. Make sure that you only add data sources relevant to the agent’s tasks and avoid adding irrelevant data. Remember that quality is better than quantity.

Note that data sources might not be needed depending on your agent’s purpose.

...

You have the option of:

  • adding

...

  • an external integration:

    • Click the External integrations card;

    • Select the integrations you want to add;

Info

Note that you can also create a new integration by clicking the Add

...

Info

Visit this page external integration button next to the search bar via this page.

...

  • adding a document:

    • Click the Documents card;

    • Click Add Documents, select the documents you want to add, and click Add Documents once more. Note that you can also upload new documents before adding them to the agent by clicking the Upload document button.

    • Provide a description of the documents explaining why your agent should search the documents.

...

Info

Visit these pages to learn how to create data sources for your agent to use: Data sources.

Tools

You can add one or more tools to your agent by clicking the Add a tool button.

...

the Tools card and selecting the correct tools for your use case.

...

Publish your agent

Finally, it’s time to decide who can access your agent. Click the Next button on the bottom right of the page to access the Publish step.You can share your agent with specific people and teams or give everyone access to use or edit

Here, you can define the access and permissions, the agent’s description, and the preset questions.

Access and permissions

You can choose who can edit your agent and who can chat with them:

  • Builders: anyone with an Agent builder within your organization can edit or chat with the agent.

  • PrivateCreator: the agent is will only be available to you.

  • Company: the agent is available to you and your colleagues company-wide.

  • Shared: the agent is available to specific users designated by you.

...

  • the creator of this agent for editing or chatting.

  • Users (only for chatting): anyone in my organization with an account can chat with the agent.

Agent description

You can add a short description to your agent to indicate to users what the agent can do. This description will be displayed in the chat interface and visible to all users.

Preset questions

You can add up to three preset questions, whose purpose is to guide the user on how to query the agent. You can put them in any order by dragging them in the position you want.

...

Finally, click Create on the bottom right to finalize the creation of your agent.

Test your agent

Once you have finished the configuration of your agent, you can test it by typing in the chat box in the right panel. accessing the Chat and invoking your agent to discuss with them, or you can click the Chat with agent icon next to your agent in the agent list.

...