How Do I See What Needs My Action Today With an AI Agent?

On October 7, 2026, I asked a chief of staff AI agent for a dashboard of what needs my action right away and what I've let slip through the cracks (12:27). I didn't tell it what the dashboard should look like, and the session shows what happened. The file on the session page spells out the format.

What did I ask the AI agent to show me?

I asked for a view of what needs action, and I named what slips.

From the recording, 12:17. Additionally, I want to be able to manage my calendar and not miss anything and not miss replying to critical emails at the end of day. I also want to ensure that you are, as my chief of staff agent, providing a dashboard so that I can visually see things that I need to take action on right away and things that maybe do that I've let slip through the cracks.

Two groups sit inside that request: the things that need action right away, and the things that have already slipped. The brief in the file on the session page has a part for the second group, called open loops.

What happens when the format is left open?

I asked for a dashboard and did not say what a dashboard would hold. The recording shows the result.

From the recording, 13:05. Now you'll notice I didn't tell my agent how to give me this information outside of a dashboard. So it'll ask me clarifying questions in a moment.

As shown October 7, 2026, the AI agent came back with multiple-choice questions, and I said option D, other, takes a typed answer instead (17:26 to 17:42). That is the screen where the AI agent asked its questions. The recording does not read them out.

What is in a three-part brief?

The file on the session page, chief-of-staff-agent.md, answers the question the recording left open. It says what goes in the brief, in this order. Each part is quoted below as the file words it.

From the file on the session page, part one. List what's on my calendar today, in time order. For each one, one line: what it is, who it's with, and anything I should walk in already knowing.

From the file on the session page, part two. For each email that needs a response, give me: Who it's from and what they're asking. A drafted reply, written in my voice, saved as a Gmail draft if Gmail is connected, or written out in this chat if it isn't. One line on how urgent it is and why.

From the file on the session page, part three. The things that are still open and nobody has closed: a follow-up I owe someone, a decision I haven't made, a thing I said I'd send and haven't. Name each one plainly. Don't soften it into a suggestion if it's actually overdue.

Step 6 of the session page tells you to type "Give me today's brief." and then describes what to expect: a short brief in three parts, with each draft marked as a draft. The file also says the AI agent never sends an email until the words "send it" are said in that exact turn, and How Do I Make an AI Agent Wait for My Go-Ahead Before It Sends Anything? covers that rule.

Which two routines does the build name?

Between the recording and the session page, two moments in the day come up.

• Morning. Step 7 of the session page lists "Each morning" with the same prompt, "Give me today's brief."
• End of day. In the dictation I asked it not to let me miss "replying to critical emails at the end of day" (12:23). Later in the build, I said my AI agent had created a skill called an end of day sweep (40:41).

I did not read out what is inside that skill on the recording, so I do not describe how it is built here.

Where do you stand if your inbox is not connected?

The file covers it. If Gmail and Calendar are not connected, it says to work from what is pasted, and it gives the sentence to open with. The session page also carries a made-up inbox for the same purpose. Both are linked from the Chief of Staff session page.

What goes wrong

One thing came up, and I named it. Instructions without access limit the result. I described it with a new hire.

From the recording, 6:05. However, if I'm doing an onboarding and I leave out critical information and I restrict access and I don't provide access to systems, then that new hire isn't going to achieve as much.

What the recording does not show: my own dashboard, which I chose not to open on screen because it holds information about people I work with (55:04), and the contents of the end of day sweep.

Where this comes from

This article grows from the session I taught live with Women Applying AI on October 7, 2026. The quoted passages from the recording are my own words, taken verbatim. The passages marked as from the file on the session page come from the files published with that session. The lists and steps around them were drafted by my team's AI agents from those words and my published pages, and checked against the recording before publishing.

Build Your First AI Agent in One Hour: the full session with the clips, the prompts and the transcript.

Related reading: How Do You Dictate Goals to an AI Agent?, How to Create Skills for an AI Agent, and the Chief of Staff session page.