Create your first agent

This tutorial takes you from the Agents dashboard to a tested agent. Choose a source whose contents you can confirm yourself. Then you can tell whether the agent uses grounded facts instead of confident-sounding text.

1. Create the agent

Sign in at agentstack.build and open your workspace. The dashboard lists its agents.

You need permission to create agents in the selected workspace. The full walkthrough also needs permission to add sources and edit the agent. To view the final conversation in Activity, you need Activity access. Confirm the organization in the workspace switcher before you continue. Agents and their knowledge belong to that organization.

1. Select New AI agent. In an empty workspace, you can also select Create Agent in the empty state.
2. In Create New Agent, enter a Name such as Customer Support Agent.
3. Select Create Agent.

AgentStack creates the agent and opens its Analytics page. You can return to the Agents dashboard to select the agent later. Selecting its card also opens Analytics. If the dialog says that the workspace reached its agent limit, reduce the number of active agents or upgrade the plan. Do this before you create another agent. Scheduling an unneeded active agent for deletion removes it from this limit.

2. Add one reliable source

In the agent sidebar, expand Sources and select Content. The Content page supports Website, Pages, Files, Notion, and Q&A sources.

For a simple first test:

1. Keep the Website tab selected.
2. Under Crawl Website, enter a public URL that contains a fact you know, such as a shipping or returns page.
3. Set Max Pages low enough to include only the relevant page, not an entire unrelated site.
4. Select Fetch.
5. Wait until the website status becomes Completed. If the status becomes Failed, inspect the displayed error before you test the agent.

The website table shows its status and page count. The Pages tab shows the pages that AgentStack crawled and indexed. While the status is Crawling or Processing, do not treat the crawl as ready knowledge.

If a website is not the right source, a focused Q&A pair gives one clear answer. Use Files or Notion when the source material already exists in those formats. The full Sources guide describes the available options.

3. Give the agent a narrow job

Open Playground. The left panel contains a Model selector and a System Prompt editor. Start with a short instruction that covers:

• the audience and job
• which subjects are in scope
• the desired answer style
• what to do when the sources do not support an answer
• when to ask a clarifying question or offer human help

For example:

You are the support agent for Acme Store. Answer customer questions using the
available company knowledge. Keep answers concise and do not invent policies.
If the knowledge does not contain the answer, say that you are unsure and
provide the configured support contact or explain the manual next step.

Select Apply & Test to use the edited model and prompt in the Playground chat. This applies the changes to the test session only. It does not save them as the agent's configuration. After the test behaves correctly, select Save to Agent. When there are unsaved changes, Reset restores the last saved model and prompt.

A newly created agent has no actions. It cannot transfer a conversation or perform another handoff action until you configure that capability. Until then, write the prompt to offer only a support contact or another manual route that exists.

4. Test supported and unsupported questions

Ask at least three kinds of question:

1. A direct question whose answer appears verbatim in the source.
2. A paraphrase that requires retrieving the same fact with different wording.
3. A question that the source does not answer.

Confirm the facts in each response, not only the tone. A good unsupported response follows the boundary that you wrote in the system prompt. It does not fabricate a plausible policy. If a known answer is missing, confirm three things:

• the source is Completed
• the relevant page appears under Pages
• the page contains the needed text

Use the response-quality checklist for a more thorough evaluation.

5. Prepare a first deployment

When the acceptance questions pass, open Settings → General and confirm the saved model, system instruction, and Public setting. A non-public agent does not render its embed widget on external sites.

Then open Deploy. The Chat Widget tab provides the embed script, while Domain Protection can restrict the widget to one hostname per line. Enter bare hostnames such as example.com, without https:// or a *. wildcard. When you want to allow all domains, leave the domain list empty. Follow the widget installation guide for placement and verification.

After you install the widget, open the customer site in a private browser window. Repeat one supported question and one unsupported question. When the live widget matches the Playground behavior, and the conversation appears in Activity, your first agent is ready for a limited launch.