Help Center / Understand and Improve Agents

Explanation

Meet Your Super Agent Command Center

Use your Super Agent to coordinate work, monitor progress, and launch repeatable tasks from one place.

5 min read
On This PageWhat a Super Agent Does
  1. What a Super Agent Does
  2. Start With a Clear Request
  3. Use the Command Center
  4. Build Reliable Workflows
  5. When to Improve It

What a Super Agent Does

Your Super Agent is the command center for AI work in your workspace. It gives you one place to assign tasks, review activity, connect tools, and understand what happened during a run.

Use it for work that needs context across conversations, files, integrations, or teammates. For example, a Super Agent can summarize customer issues, prepare follow-ups, check operational dashboards, or coordinate recurring reporting.

Start With a Clear Request

Open your Super Agent and describe the outcome you want, the source material it should use, and any constraints it should respect. A useful request includes the goal, the audience, the deadline, and the format you want back.

If the task touches a connected app, mention the app by name and start with a small check before asking the Super Agent to make changes or send messages.

Use the Command Center

The command center helps you move between the agent conversation, connected tools, recent activity, usage, and traces, the step-by-step run logs, without losing context.

Super Agent command center showing chat, connected tool chips, recent activity, and a latest trace panel
Use the command center to keep the chat, tool status, recent sessions, and trace inspection in one place while you tune agent work.

Check the latest sessions when you need to see what the Super Agent has been working on. Open traces, the step-by-step run logs, when you want to inspect tool calls, missing access, or a result that needs tuning.

Build Reliable Workflows

Turn repeated requests into reusable prompts or scheduled jobs after the Super Agent has completed the task successfully by hand.

Keep instructions practical and specific. Add approval rules for actions like sending external messages, changing customer records, or using sensitive data.

When to Improve It

Update the Super Agent when the same correction appears more than once, when a team process changes, or when a trace shows that the agent guessed instead of asking a question.

Small edits usually work best: add the missing source, define the preferred output format, or clarify when the Super Agent should ask for approval.

Was this helpful?

Your feedback helps us rank and improve Help Center articles.