M6 · Free playbook
Your calendar should defend your priorities, not be a public buffet. M6 protects deep-work, pre-screens prospects, books the right people without back-and-forth, and runs five recurring ops automations — invoicing, payment reminders, review requests — without you thinking about them.
TL;DR
AI Calendar Ops is five things wired together: a defender (Reclaim), a booking link (Cal.com), a Claude pre-screen for the routing form, an orchestrator for five recurring ops automations, and a priority taxonomy that encodes what actually matters to you. The stack costs about $60 to $100/mo and takes roughly four hours of focused work across seven days.
The hard part is not the automation. The hard part is the priority kickoff — getting honest about what you protect and what you don't. Most owners enter intake protecting deep-work; almost no owners enter intake protecting self-care. The Day 1 push-back is the work.
What follows is the seven-day plan, the exact stack, honest tradeoffs between DIY and paid, the QA scenarios, and the safety rails — including the one sandbox-mode rule that has caused real customer-facing damage when skipped.
The 7-day plan
Each day fits in a single focused block. The Day 1 priority kickoff is the work — every other day is wiring it into Reclaim, Cal.com, and Zapier.
Day 1
~75 min
This is the only sixty-minute call in the install. Walk through the priority taxonomy — deep work, inner circle, high-value meetings, standard meetings, admin, self-care, buffer. You will under-protect self-care in the intake form; the engineer will push back gently. The taxonomy you agree on at the end of Day 1 is what Reclaim enforces forever.
Day 2
~60 min
Reclaim Lite or Business at $10–$18/mo. Cal.com Pro at $15/mo. Zapier Pro at $30/mo if not already running for another module. Anthropic API key with a usage cap. Connect Google or Outlook calendar to both Reclaim and Cal.com via OAuth. The two tools share the calendar but each owns a different slice — Reclaim defends, Cal.com books.
Day 3
~75 min
Configure working hours, decompress days if any, deep-work habits with Maximum priority, self-care habits (gym, lunch, family dinner), buffer rules between meetings, and decompress windows after marathon days. Test the priority cascade by manually booking a low-priority meeting during a deep-work block — Reclaim must refuse or reshuffle. If it doesn't, the priority isn't set to Maximum.
Day 3–4
~75 min
Create event types per meeting type — intro, active client check-in, vendor, partner, inner circle. Set buffer before and after each, cap meetings per day, set max future booking. Build the routing form: name, email, company, topic (50-char minimum), have-we-worked-together. Routing logic: prior-history yes → auto-confirm; inner-circle list → auto-confirm; everything else → Claude pre-screen.
Day 4–5
~90 min
Build the Zapier scenario that takes a routing-form submission, runs a Claude classifier (qualified prospect / sales pitch / free-advice ask / ambiguous owner-only), and either confirms the booking or sends a custom decline in your voice. Then pick the highest-leverage ops automation — usually invoice send or payment reminder — and wire it. Test end-to-end with sandbox mode before letting it touch real customers.
Day 5–6
~75 min
Wire the other ops automations you opted into: contract follow-up, review request, anniversary message. Critical step: before any test that fires a real CRM or billing event, confirm in writing that Stripe is in Test Mode, QuickBooks is in Sandbox Company, and HoneyBook is in Test Project Mode. After tests pass, flip sandboxes back to live and confirm in writing again. This is the one place M6 installs have caused real customer-facing damage.
Day 6–7
~60 min
Generate 10 sample ops messages — three payment reminders, two contract follow-ups, three review requests, two anniversary messages. Rate each 1–5 on voice fidelity. Bar: 9/10 at 4 or 5. Run all 10 base QA scenarios. Then flip Zapier ON. Send yourself a test booking; confirm the routing works. Schedule Day 7, Day 14, Day 30 check-ins.
The core product idea
M6's value isn't scheduling — it's encoding the owner's actual priorities into the calendar. The taxonomy below is the default; your intake refines it.
Deep work
Strategy, writing, design
Defended; meetings cannot displace
Inner circle
Spouse, kids, family, top 3 clients
Always-yes if asked; auto-confirmed
High-value meetings
Active prospects, partner conversations
Batched on Tuesday/Thursday
Standard meetings
Vendor calls, routine ops
Lower priority; batched on Tuesday/Thursday
Admin / errands
Email batches, expense reports
Scheduled into low-energy windows
Self-care
Exercise, lunch, family dinner
Protected as non-negotiable
Buffer
15-min gaps, 30-min lunch
Auto-inserted
The stack [Verified 2026-05-23]
This is the default stack for paid M6 installs. Reclaim and Cal.com pricing have both moved in 2026 — re-verify against vendor pages within seven days of any paid install.
Calendar defender
$10–$18/user/mo
Best-in-class calendar defense in 2026. The habits system encodes priorities better than Motion (which is more task-management-flavoured) and the smart-1:1s logic auto-reschedules low-priority meetings when high-priority work needs the slot. Default for owners who want their calendar to fight for them.
Booking link
$15/mo
Pre-screening + routing logic is stronger than Calendly Pro for the same money. The routing form supports our Claude pre-screen flow cleanly. If you already have Calendly Pro, no point switching — the install works on either; Cal.com is the better default for new installs.
Pre-screen + ops orchestrator
$30/mo (or shared with M2/M3)
The Claude pre-screen runs as a Zapier scenario between Cal.com and your calendar. The five ops automations all run as Zapier scenarios. If you already have Zapier from M2, reuse it. Make.com works too; Zapier is the simpler default for owners who'll touch the automations themselves.
Pre-screen + ops drafts
~$5–$15/mo
The pre-screen classifier needs a model with actual judgement, not pattern-matching. Sonnet 4.6 catches sales pitches dressed up as introductions; Haiku doesn't. Drafts for payment reminders and contract follow-ups also need Sonnet for voice fidelity. Total spend stays single-digit dollars/month at SMB volume.
Priority KB
Free
Three KBs by Operator: business KB (M1/M2/M4), personal voice KB (M5), priority KB (M6). The priority KB lists your priority tiers, your inner circle, your hard-no meeting blocks, your self-care non-negotiables. Keeping it separate from voice prevents the pre-screen classifier from drifting into draft-tone decisions.
Invoice + payment integration
Existing
The ops automations send invoices, payment reminders, and review requests through whatever financial tools you already use. We do not install a new financial tool. Sandbox mode is mandatory during install testing — see the safety rails section.
All-in monthly
For a small-business owner running calendar defense + a routing form + three of the five ops automations, expect roughly $60–$100 all-in. Reclaim + Cal.com is the floor; Zapier + Claude API is the variable.
DIY or paid — honestly
Most playbook PDFs end with “or just buy our thing.” This one is honest about when DIY is the better answer.
You already use Reclaim or Motion and just want to layer the pre-screen on top. You're comfortable in Zapier and have an Anthropic API key. You can spare four hours over a week. You're willing to enforce the sandbox-mode rule during install testing without a second pair of eyes — this is the one rule no DIY install can skip safely.
You want a real person walking through the priority kickoff conversation — the Day 1 taxonomy work is awkward to do alone and the engineer's push-back on under-protected self-care is half the value. You don't want strangers in your calendar but you want help getting the rules right the first time.
You'd rather pay for the install than learn Reclaim's habits system and build five Zapier scenarios from scratch. You want the ops automations launched in phases over thirty days, not all at once. Most owners booking 8+ meetings/week with the wrong people land here — the pre-screen alone is usually worth the install on Pilot pricing.
QA — the ten scenarios
Run each scenario manually. If any fails, fix the pre-screen prompt or the priority rules and re-run. Sandbox mode is mandatory for the ops automation tests.
Risks + safety rails
Sandbox mode for financial automations. Before any test that touches your CRM or billing tools, confirm in writing (Slack with timestamp) that Stripe is in Test Mode, QuickBooks is in Sandbox Company, and HoneyBook is in Test Project Mode. After tests pass, confirm in writing again that all three are flipped back to live. Sandbox-mode-stuck installs have wasted two to three client-days of “why aren't my invoices going out”.
Double-booking on calendar sync lag. Reclaim and Cal.com both write to the same calendar; a race condition can produce a double-booked slot if sync lags. Set min-notice on Cal.com to 24 hours minimum to give the sync window time to settle.
FDCPA + payment-reminder tone. If the payment reminder reads like a debt collector, you have legal exposure under the Fair Debt Collection Practices Act and state analogues. Use soft language on the first reminder. Firm up only at T+14d. Never threaten, never imply credit reporting, never call repeatedly — and review the templates with counsel if you're in a regulated industry.
Pre-screen too aggressive. Rejecting legit prospects because the classifier is overtuned is the most common Day-7 issue. Tune the classifier toward allow-by-default; the cost of a false decline is much higher than the cost of a sales pitch slipping through.
DIY the whole thing free. Or pay $500 and Iris is on a call walking through the priority kickoff. Or pay $997 and we install M6 with three ops automations live in seven days.
Free
Download the PDF
This playbook as a PDF you can print and check off.
Get the PDF →
$500
Iris-Assist
You do the clicks; Iris is on the line. One 90-min call. You keep every credential.
Book a session →
$997 + $97/mo
Iris Build Pilot
We install M6 with three ops automations live. Seven days. 30-day money-back.
See Pilot →