Claude Cowork prepares a daily work dashboard
Pean reports a daily dashboard combining scheduled inbox, meeting and research tasks, with prepared email drafts, priorities and links to original records.
substack.com
Suggested prompt
Suggested prompt written by Agents In Use, not the source author. Replace bracketed inputs and review tool permissions first.
Prepare a daily dashboard from my authorized inbox, calendar and task systems. Link every item to its source, distinguish urgent actions from background reading, and leave outgoing communication as drafts for my review.
Reach people exploring AI agents
See placements →Evidence at a glance
Sourced account- Source author
- Christian Pean MD, MS
- Original account
- Read the source ↗
- Published
- 25 Mar 2026
- Source link checked
- 15 Oct 2026
- Reported result
- Pean reports a daily dashboard combining scheduled inbox, meeting and research tasks, with prepared email drafts, priorities and links to original records.
- Human involvement
- Pean acts from the dashboard and Cowork; outgoing messages are described as drafts. The source does not establish independent medical interpretation or detailed approval rules for every task.
A linked account supports attribution. It does not mean we independently reproduced its result.
Report an issue with this case
Open correction emailOpens your email app with the case link included. Nothing is sent automatically.
How the agent did it
- Connect authorized email, calendar, notes and business tools through MCP.
- Run scheduled tasks for triage, meeting preparation and relevant updates.
- Aggregate drafts, prioritized actions and source links into the daily dashboard.
Christian Pean describes using scheduled Claude Cowork tasks as a personal chief of staff. His morning run brings together inbox triage, meeting preparation, stale deal alerts, policy scanning and literature updates. The result is an interactive dashboard with prioritized actions and links back to source threads.
MCP connectors provide access to tools such as Gmail, Calendar, Slack, Granola and HubSpot. The account says email replies and team messages are prepared as drafts, while Pean acts from the dashboard and Cowork. He runs about twenty tasks across clinical, business, policy, academic and personal work, using different daily, weekly and monthly schedules.
The source also introduces a configurator for other users, but the completed deployment described here is Pean's own connected task portfolio. Its existence does not establish that a particular medical paper was interpreted correctly or that the dashboard replaced professional judgment. No independent measurement of time saved or task completeness is supplied. The practical result is a recurring coordination artifact with source links and drafts, keeping the user's action decisions visible instead of claiming that every task executes externally without review.
Prompt
Suggested prompt (written by Agents In Use, not the author's)
Source: Blog · Christian Pean MD, MS
Frequently asked questions
What does the morning dashboard contain?
Prioritized actions, prepared email drafts, meeting context and source links.
How does it reach the work systems?
The author describes MCP connectors to tools including Gmail, Calendar, Slack, Granola and HubSpot.
Does the account verify every research conclusion?
No. It describes aggregation and workflow, without an independent audit of research interpretation.
More details
About this account
Cowork executed scheduled research and briefing tasks and assembled the clinician dashboard from connected information.
Firsthand account read. Original Substack Note datePublished 2026-03-25T00:57:55+00:00. No independent completeness or medical-interpretation audit; drafts are not assumed sent. Reported outcomes were not independently reproduced. Suggested prompt written by Agents In Use.
Credit: Christian Pean MD, MS
Source published 25 Mar 2026
Source checked 15 Oct 2026
Tools and topics
Claude Cowork · Gmail · Calendar · Slack · HubSpot
Claude Cowork · Gmail · Calendar
Editorial completeness: 100/100. Measures documented fields, not agent performance or verified outcomes.