Quickstart

This quickstart takes you from an empty workspace to a useful agent you can test in a chat. You can add a team, pipeline, or trigger after the basic behavior is reliable.

1. Describe the outcome

From Home, describe one real task in the builder. Be specific about the input, desired result, and what a good answer should contain.

Good starting tasks are bounded and easy to verify, such as summarizing a document, classifying incoming text, or turning structured notes into a report.

2. Create the agent

Open Agents, choose Create Agent, and provide:

• a name that describes the role;
• a short purpose that describes the outcome;
• instructions covering behavior, constraints, and output;
• a model appropriate for the task.

Save the agent before adding optional tools or advanced execution settings.

3. Add useful knowledge

Open Knowledge, stay on the My Files tab, and upload the documents the agent should use. Keep source material focused: unrelated files make retrieval noisier and make incorrect answers harder to diagnose.

A shared workspace also has a Workspace Library tab for material that members should discover and reuse. It is available even before another member joins. Personal workspaces show only My Files. See manage a workspace for invitations and workspace controls.

To give the agent a focused set of sources, choose its knowledge access from My Files, the Workspace Library, or both.

4. Test in a chat

Open Chats, start a conversation with the agent, and send a representative request. Check the answer itself and the run details behind it.

Test at least three cases:

1. a normal request;
2. an ambiguous request;
3. a request the agent should refuse or clarify.

5. Automate only after it works

When the agent behaves consistently, put it into a pipeline, a team, or a scheduled or webhook-driven flow.

Automation multiplies both reliable behavior and bad assumptions. Stabilize the agent first.