Skip to main content
Outbound research

Meeting Prep Briefing

Walk into every meeting prepared. The agent reads your calendar, researches each attendee, and emails you a briefing before the call.

30 sec to prepfor any meeting on your calendar — no skimming, no LinkedIn rabbit hole(every day on autopilot, before your first coffee)

Built across the leading AI labs — Anthropic, Google, OpenAI — picking the right model for each step.

  • Claude Agent SDK
  • NotebookLM
  • Google Calendar API
  • AgentMail
  • FastAPI
  • SQLite
InputTomorrow 10:00am — "Initial intro — discussing AI infra partnership" with elon@x.com. 30 min.

Click Run to see the result.

Sources read12Briefing built in34sRecent signals4
everymeeting

on tomorrow's calendar — briefed automatically, not just the "important" ones

~30sec each

from calendar event to a one-page briefing in your inbox

0effort

cron-triggered overnight — the brief is waiting when you wake up

Two things change. First, you stop walking into meetings cold. Every meeting gets briefed — not just the ones you remembered to research. Second, and more importantly: the quality of the questions you ask in the meeting goes up. You reference what they shipped last week, the talk they gave, the news that broke yesterday — and the meeting moves faster because you are starting from a shared baseline. The attendee notices. This is not a tool you license. It is built around your calendar, your context, your stack. Same toolbox we use every time; different shape for every business.

How it works

  1. Reads your calendar overnight

    Every morning at 7am the agent pulls tomorrow's meetings from Google Calendar and identifies the external attendees worth researching — skipping internal teammates and recurring standups.

  2. Researches each attendee from the public web

    For each external attendee the agent runs a focused research pass: recent news, interviews, podcasts, public statements from the last few weeks. Sources are cited so you can verify any claim in one click.

  3. Emails you a one-page briefing per meeting

    Claude synthesizes the research into a per-attendee briefing — who they are now, 3-5 talking points specific to your meeting, recent signals worth knowing about — and emails the lot to you before the call. NotebookLM also generates a poster-style infographic for each attendee, so the visual is right there alongside the email.

What lands in your inbox

The real artifacts the pipeline generates for the Elon meeting in the video — the briefing email itself, and the NotebookLM-generated poster infographic that arrives alongside it. Both delivered ~30 seconds after the prompt.

Email briefing for the Elon Musk meeting
The briefing email. Summary, talking points tied to your meeting context, dated recent signals, and cited sources you can verify in one click.
NotebookLM-generated infographic for the Elon Musk meeting
The NotebookLM infographic. Auto-generated from the same research session — "Tesla's Master Plan IV" with the Optimus roadmap, $7.5T valuation target, and AI compute footprint.

Why this automation

Meeting prep is one of those tasks that is obviously worth doing and yet rarely gets done well. The "important" meetings get a five-minute LinkedIn skim. The rest get nothing. By the time you are on the call, you are reconstructing the attendee's recent context from whatever you can remember plus whatever they say in the first two minutes.

This is not a story about reading faster. It is a story about doing the prep step that was not consistently happening at all — and doing it for every meeting, not just the ones with budget attached.

See it in action

Prompt → calendar event → research → emailed briefing + NotebookLM infographic + proposal slide — all generated automatically.

The transformation

Before

You skim LinkedIn five minutes before the call, miss the news that closed yesterday, and end up asking generic questions because you did not have time to dig. The "important" meetings get prepped; the rest get winged.

After

Every meeting on the calendar gets a briefing — talking points specific to your context, recent signals on each attendee, sources cited. You walk in knowing what they shipped last week and what you should actually ask.

Where this goes from here

Swap the endpoint

  • This demo emails the briefing — yours could post to Slack, drop a card into Notion, or attach itself directly to the calendar event so the brief lives next to the meeting itself.
  • Research here uses NotebookLM for synthesis. Yours could pull from your CRM (last touchpoints, deal stage), your call recordings (what you discussed last time), or any internal source — the briefing format stays the same; the inputs follow your stack.

Expand the scope

  • Today it briefs one meeting at a time. Same pipeline can compile a weekly "who you're meeting this week" digest, flag pre-call homework you still owe, or surface follow-ups based on what was discussed in prior meetings.
  • Once you have a briefing per meeting, the same agent can draft your opening message, pre-write follow-up emails for likely outcomes, or schedule the "send a thank-you" task into your task tracker — turning prep into a closed loop with the rest of your relationship work.

Every ClearSpring automation is the entry point, not the ceiling. The demo is what we'll show; the deployment is what we'll build.

Want something like this for your business?

Every business is different. We start with a short conversation about the workflow you’d most like to stop doing — then build a custom tool around it.