This weekend, build the system that does it.
The Monday brief, the recaps, the roll-up, the QBR deck — built on your transcripts and inbox, still running next week.
A note-taker on every call. A workspace holding every SOP you wrote. A Claude tab open all day. You still write the recap on Sunday, from Thursday's transcript.
Every module ends with a working artifact on your data: a skill, an agent, a connector, a dashboard. By Sunday the brief runs on a schedule.
Two live days, one instructor, two floor coaches, so nobody falls behind. Module 0 is a governance clinic: strict data policies don't sit anyone out.
Your capstone is a jet-black, liquid-glass command center your principal can actually open. It needs no explaining once it's on screen.
The notifications stacked up while you were in back-to-backs.
The Monday-Morning Brain: a scheduled brief and triage agent.
The nagging feeling that something got dropped.
The Meeting-Recap skill that opens the action items itself.
The quarterly deck, hand-assembled from five sources, again.
The QBR / board deck generator.
Three directives in one afternoon, no record of what was decided.
The Operating Cadence Engine: status roll-up plus decision log.
It's all in your head, and only your head.
A connected Knowledge MCP your whole team can query.
You ask Claude a question, read the answer, close the tab.
The whole weekend, from prompting to delegating.
Nothing moves the week you're on leave.
The capstone orchestrator that runs without being asked.
One orchestrator. One command center. Built on your real data, kept forever.
Run morning brief
one button · 6 agents
Meeting-Recap
transcript → clean summary
Status-Roll-Up
team updates → one report
Deck-Builder
KPIs → near-final deck
Stakeholder-Brief
one-screen pre-meeting brief
Digest
the day, in five lines
Knowledge
answers, with citations
A CLAUDE.md standing-context file: your quarter priorities, your principal's profile, your stakeholders, your decision style. Every agent and the dashboard load it automatically.
Meeting-Recap, Status-Roll-Up, Deck-Builder and Stakeholder-Brief, each invoked on demand.
Read-scoped or synthetic connectors to Gmail, Calendar, Notion and Slack — built safely for strict orgs.
Every action writes to a governance log. Built in, not bolted on.
The digest and roll-up run on a cadence, without being asked.
KPIs, open action items, the decision log and team status, auto-refreshed from the data layer. The thing your principal opens.
One trigger, in parallel: the digest, the recap, the roll-up, a dashboard refresh. You demo it live, then leave with the repo and the runbook.
The survival load first: context-switching, dropped commitments, the notes grind, the knowledge only you hold. You leave with your operator brain, a knowledge assistant, and reusable skills running on your real meetings and inbox.
Single skills become orchestrated systems: scheduled agents, n8n automations, auto-built decks, a live dashboard, async video and voice updates. Everything composes into your AI chief of staff, behind a command center you demo live.
The same task typed into a chat box, then handed over as a job with standing context behind it. Model choice, skills versus MCPs versus subagents, and the governance clinic where you pick your data tier.
A real transcript becomes a reusable Meeting-Recap skill, live: clean summary, decisions, action items with owners and due dates. Then the n8n flow that posts it to Slack and opens the tasks — running on last week's real meeting.
MCP without the engineering detour: what a server is, connecting Notion, Drive and Slack, scoping permissions, building a citation-backed assistant. Once the team can query it, you stop being the lookup.
One in-the-head process — how you run the weekly ops review — becomes a documented SOP and an executable skill. Then the Stakeholder Intelligence agent and CRM behind a one-screen pre-meeting brief, seeded with your five real stakeholders.
A scheduled agent crawls your calendar, notes, inbox and Slack into a one-page brief. Then triage on top: replies in your principal's saved voice, meetings declined or batched, focus blocks held.
A repeatable workflow: KPIs from the canonical sources, commentary drafted, variances flagged, a near-final deck out. Plus the freeze-the-data discipline, and a diff against the deck you built by hand last quarter.
A Weekly Status Roll-Up pulls team updates, normalizes them and assembles one exec-ready report. A Decision-Log agent records what was actually decided, with owners, so decisions stop getting re-made.
The week's written update becomes a script, then a HeyGen avatar video or an ElevenLabs voice briefing, automated with n8n. One script, many formats — the same narrative for exec, board and team.
Everything composes into one orchestrator, on Anthropic's official Chief of Staff Agent architecture. The operator brain as standing memory, the weekend's skills registered, subagents via the Task tool, hooks writing the audit trail. Then the command-center dashboard and the maintenance runbook.
artifacts, yours to keep
Operator Brain — a personal CLAUDE.md every agent and dashboard loads automatically.
Meeting Recap & Follow-Up — a skill + n8n flow that posts recaps and opens tasks after every meeting.
Knowledge MCP — a connected source over your real docs that your whole team can query.
SOP-skill library — a starter set that clones your judgment, one skill per recurring process.
Stakeholder Intelligence — an agent + lightweight CRM that briefs you before every meeting.
Daily/Weekly Brief + Triage — scheduled agents with your principal's saved voice profile.
QBR / Board Deck Generator — turns canonical data into a near-final deck.
Operating Cadence Engine — Weekly Status Roll-Up + a persistent Decision-Log agent and database.
Leadership Update to Video/Voice — a workflow (Claude to HeyGen/ElevenLabs, automated via n8n).
The capstone — a composed AI Chief of Staff orchestrator behind a live command center, with a runbook so it doesn't rot.
Cheat sheets — a model-choice one-pager and a data-governance one-pager, ready to forward to whoever approves this.
Assumed background — No coding background required. You write context files, build skills and wire connectors in plain language.
Day 1 opens with a 30-minute governance clinic. Under strict data policies, you build the identical system on a scoped or synthetic copy.
A laptop (Mac or Windows) and a real, in-progress workload — not a sandbox.
One price, once, for every artifact on the list above — and you keep them when you change companies. Nothing is metered. There is no back-end program to buy.
No coding background assumed: plain language and a guided terminal, with two floor coaches in the room. By Saturday lunch, “I'm not technical” stops being true.
Generic by design — that's what makes them shippable. The one you build runs on your stack, holds your principal's context and your bar, and leaves with you. You own it; you don't renew it.
Then you have the transcript and you're still doing the synthesis. This weekend is the layer after it: recaps that open their own actions, roll-ups that assemble themselves, decks that regenerate.
Module 0 is a governance clinic. Scoped, read-only or fully synthetic — you build the identical system, with the audit trail built in.
You build on your own transcripts, KPIs and workspace, not on demo data. If it doesn't run on your reality by Sunday, that's on us.
The load is the reason to come. One weekend now, against the week you rebuild by hand every week after.
You leave with a maintenance runbook and the repo, plus the prompt-update routine for when the models move. The capstone is built to survive contact with a real quarter.
Every session live, every asset delivered, no back-end upsell. The build is the credential.
What each one does on the Monday after.
Lands before you're at your desk: what slipped last week, what's time-sensitive this week, your 1:1 talking points, what you said you'd do.
Scheduled agentPulls the KPIs from the canonical sources, drafts the commentary, flags the variances, and hands you a near-final deck instead of an empty template.
WorkflowOne orchestrator behind one command center. The recap agent, the roll-up, the deck generator and the decision log, all fired from one place.
CapstoneBring two real transcripts, one messy KPI export and last quarter's deck. Leave with the system that turns all three into next week's work. 25 seats, one weekend.
Finish Day 1, and if it isn't what we promised, we refund you.
Seats are capped at 25 because one lead and two floor coaches can't reach more screens. No countdown timers, one cohort a month.