Command Center Example
← Instructions
Build your Command Center

Copy a ready request into your AI with your brain folder attached. Download example ZIP

GitHub ↗
Interactive example · fictional data · 17 Nov 2026

This is a static snapshot of John’s workday. Explore the views and controls; no real accounts, messages or AI are connected.

This is a teaching example with invented data. John Doe, Doe Studio and every client, message and number below are made up. The layout follows the Command Center Egils Boitmanis runs his own business from; none of his data is on this page.

Everything here is read from one example folder, the Doe Studio brain, built with the same starter as the masterclass. Build the dashboard last: it only shows what the folder already holds.

01 / Today

Tuesday, 17 November

Start with the north star, see the road to it, then the three things that matter today.

1 task overdue 2 reply drafts waiting for approval Calendar access expires in 3 days Morning run done at 08:05

North star · 05-focus/north-star.md

Software that people trust on the first try, because it was designed from what they actually said.

Twelve-month goal: four steady clients and two teaching days a month by June 2027.

Steady clients2 / 4
Now 2Feb 3Jun 4
Teaching days a month1 / 2
Now 1Feb 1–2Jun 2

Road to June 2027

Three outcomes, the projects that carry them, and what comes later

  • Active
  • Planned
  • Later
  • Milestone
  • Done
  • Next 3 months

Scroll sideways to see the whole road to June.

Road to June 2027 Timeline from October 2026 to June 2027 with three outcome groups: clients, teaching and visibility. A table version follows the chart. Next 3 months OctNovDecJan 2027FebMarAprMayJun Clients Harbor Health discovery Brightwater retainer Third client Fourth client Teaching Northline guild session Course pilot, 3 parts Two teaching days a month Visibility Two case studies Newsletter, every 2 weeks GEO7 check, monthly Harbor Health discovery, 12 Oct to 20 Nov, active Readout, 20 Nov Readout 20 Nov Brightwater proposal, 19 Nov Brightwater retainer, 1 Dec to 28 Feb, planned Third client, 1 Dec to 15 Feb, planned Third client signed by 15 Feb Signed by 15 Feb Fourth client, March to June, later Northline guild session, 19 Nov Session 19 Nov Course pilot, 11 Jan to 5 Feb, planned Two teaching days a month, from March, later Two case studies, 1 Dec to 15 Jan, planned Newsletter every two weeks GEO7 check monthly Today Goal · Jun 2027
01
Clients

Brightwater retainer signed and running; one new client from the guild session signed by 15 Feb.

Proposal due Thursday
02
Teaching

A three-part course pilot, 11 Jan – 5 Feb, built on Thursday's session.

Session on 19 Nov
03
Visibility

Two case studies live by 15 Jan; recommended in 6 of 25 AI answers in February.

Now 4 of 24
Show the roadmap as a table
OutcomeItemWhenState
ClientsHarbor Health discovery12 Oct – 20 NovActive
ClientsBrightwater retainer (proposal 19 Nov)1 Dec – 28 FebPlanned
ClientsThird client, from the guild session1 Dec – 15 FebPlanned
ClientsFourth clientMar – JunLater
TeachingNorthline guild session19 NovMilestone
TeachingThree-part course pilot11 Jan – 5 FebPlanned
TeachingTwo teaching days a monthMar – JunLater
VisibilityTwo case studies on the website1 Dec – 15 JanPlanned
VisibilityCalm Product newsletterevery 2 weeksActive
VisibilityGEO7 check1st of each monthActive

This week · set on Monday

  1. 01Deliver the Harbor Health discovery reportFriday · client
  2. 02Run "AI for design teams" for the Northline guildThursday · teaching
  3. 03Send the Brightwater retainer proposalThursday · client

Today's three

  1. 01Write the findings section of the Harbor Health reportJohn · HH-11Doing
  2. 02Approve two reply drafts in MessagesJohn · 10 minutesWaiting on John
  3. 03Send the invoice #014 reminderMaker drafted it · HH-13 · due yesterdayOverdue

Morning run · Maker agent

08:05 read: focus, 5 project files, Gmail, WhatsApp, Facebook page, guild chat, calendar
08:05 today's three proposed (John can swap any of them)
08:06 2 replies drafted, not sent: Priya (Harbor Health), Lena (Northline)
08:06 4 watched posts: 1 question to answer on LinkedIn, 1 on the Facebook page
08:06 Checker review of synthesis v1: FIX, 2 items · Maker fixes them today (HH-10)
08:06 Folder check: marketing/ CONTEXT misses analytics.md; 06-archive/ has no CONTEXT file

Where this comes from. The north star, the three outcomes and the roadmap are in 05-focus/north-star.md, which John reviews once a month. The week comes from 05-focus/_focus-CONTEXT.md, today's three from the task tables in 05-focus/projects/, and the brief from the Maker's morning run. Nothing here is typed into the dashboard by hand.

Continue withfocus7context7

02 / Work · FOCUS7

GitHub · gtd7 ↗

One list for everything that needs doing

Every task has an ID, an owner, a project and a status. John and his two agents share the same tasks, so "what's next?" has one answer.

Inbox 3

Idea: turn the guild session into a three-part course
voice noteJohn
File Brightwater's December release plan
email attachmentMaker
Summarise an article on research ops
saved linkMaker

Open 4

Send the Brightwater retainer proposal
BW-05ThuJohn
Guild session slides
NG-03WedJohnMaker outline
Send the invoice #014 reminder
HH-13due yesterdayJohn
Outline newsletter issue 9
CP-0923 NovMaker

Doing 2

Harbor Health findings section
HH-11FriJohn
Fix the two Checker items, synthesis v2
HH-10todayMaker

Waiting 3

Analytics export from the client
HH-12asked 12 Nov5 days
Venue confirmation for Thursday
NG-04asked Mon1 day
Draft the Brightwater case study
DS-02project paused until 1 DecMaker

Done this week 3

Invoice #013 sent
BW-04MonJohn
Synthesis of P05–P06, draft v1
HH-08MonMaker
Review of synthesis v1: FIX, 2 items
HH-09MonChecker

Where this comes from. Each task lives in exactly one project file in 05-focus/projects/, with a status of OPEN, DOING, WAITING or DONE and a link to its result. The FOCUS7 skill runs them. There is no separate task app: this board is drawn from those files, and the inbox from 05-focus/_focus-CONTEXT.md.

Continue withfocus7exit7

03 / Messages · COMMS7

GitHub · comms7 ↗

Every conversation, and the comments worth an answer

COMMS7 does two jobs. It sorts every conversation by who has to act next, and it watches a short list of posts for comments that deserve a reply. The AI reads and drafts; John decides and sends.

GmailRead + drafts · 14 days
On
WhatsApp BusinessRead · 14 days
On
Facebook pageMessages + comments
On
LinkedInBrowser session, read only
Manual
Guild chatRead · 7 days
On
3Reply needed
2Waiting on them
4For information
5Closed today

The five newest open conversations

ChannelFromNext stepState
Gmail Priya ShahHarbor HealthCan we move Friday's readout to 11:00? Confirm 11:00 and move the calendar slot.Draft ready · not sent Reply needed
WhatsApp Lena OrtizNorthline guildHow many chairs and screens do you need? Answer: 40 chairs, one screen, HDMI.Draft ready · not sent Reply needed
inLinkedIn Tomás RuizindependentLiked your post on onboarding interviews. Open to a chat? Offer two times next week.Suggested · John writes it Reply needed
Gmail Ada KimBrightwaterQuestions on the retainer scope, sent by John on 12 Nov. Ada answers after their planning on 18 Nov. Nothing to do yet. Waiting on them
#Guild chat #eventsNorthline guildThursday's session was announced to the guild. No action. Kept for context. For information

Watch list rule: John's own posts from the last 14 days, plus posts by the six people in People when they are about research, onboarding or design teams. A suggested comment is a draft. John posts it himself, or skips it.

Three questions I ask in every onboarding interviewJohn's post · 11 Nov · watched until 25 Nov
6 new comments1 question to answer

Do you send the questions to people in advance?

Draft reply: "No, only the topic. The questions work best when people answer from memory, not from notes."

Thursday: AI for design teamsDoe Studio page · 14 Nov · watched until 21 Nov
2 new comments1 question to answer

Will the session be recorded?

Draft reply: "Not this time: some participants work on unreleased products. Everyone who joins gets the slides."

Harbor Health is hiring a design leadPriya Shah · client · 15 Nov · watched until 24 Nov
14 commentsOptional

Suggested comment: a short congratulation and one line on what the team does well. No client details.

Research ops is a team sportTomás Ruiz · new lead · 16 Nov · watched until 24 Nov
9 commentsOptional

Suggested comment: one practical point from John's own interview practice, or skip.

Skipped 16 Nov a general post about AI replacing designers: not a fit with the north star.

The one rule: the AI reads, sorts and drafts. Nothing is sent, posted, accepted or deleted without John's yes.

Where this comes from. 02-memory/comms/_comms-CONTEXT.md holds the sources, how far back each is read, the rhythm and the watch list. The connections and their limits are in 04-tools/_tools-CONTEXT.md. When John closes or parks a conversation, the decision goes into the project file it belongs to, so the next review does not raise it again.

Continue withmeeting7exit7

04 / Projects · FOCUS7

GitHub · gtd7 ↗

Where the focus is, and what each project needs next

FOCUS7 works the GTD way: every project serves an outcome, has one next action and keeps its tasks with a status. On Monday John picks the week's focus from these files. Everything else stays visible but quiet.

In focus this week3 of 5projects, one per outcome or close to it
Tasks in project files134 done · 2 doing · 4 open · 3 waiting
Overdue1HH-13, invoice reminder drafted, not sent
Waiting for others2Priya and Lena; one more is John's own pause

Focus map · set Monday 16 Nov

  • In this week's focus
  • Active
  • Paused, or not a project yet
01ClientsFour steady clients by June 2027 · now 2
HHHarbor Health discoveryfocus · readout Fri 20 Nov BWBrightwater retainerfocus · proposal Thu 19 Nov Third client, from the guild sessionsomeday · by 15 Feb Fourth clientsomeday · Mar–Jun
02TeachingTwo teaching days a month · now 1
NGNorthline guild sessionfocus · Thu 19 Nov Three-part course pilotsomeday · 11 Jan – 5 Feb
03VisibilityTwo case studies by 15 Jan · AI answers 6 of 25
CPCalm Product newsletteractive · issue 9 outline DSDoe Studio websitepaused until 1 Dec
Project · fileServesNext actionTasksState
HHHarbor Health discovery05-focus/projects/harbor-health-discovery.md ServesClients Next actionWrite the findings sectionJohn · Fri 20 Nov Tasks7 tasks · 1 overdue On track
BWBrightwater retainer05-focus/projects/brightwater-retainer.md ServesClients Next actionSend the retainer proposalJohn · Thu 19 Nov Tasks2 tasks On track
NGNorthline guild session05-focus/projects/northline-guild-session.md ServesTeaching Next actionFinish the session slidesJohn, the Maker drafts the outline · Wed 18 Nov Tasks2 tasks Venue not confirmed
CPCalm Product newsletter05-focus/projects/calm-product-newsletter.md ServesVisibility Next actionOutline issue 9Maker · 23 Nov Tasks1 task Issue 8 was four days late
DSDoe Studio website05-focus/projects/studio-website.md ServesVisibility Next actionDraft the Brightwater case studyMaker · from 1 Dec, needs Ada's OK Tasks1 task Paused by John until 1 Dec
  • Done
  • Doing
  • Open
  • Waiting

Waiting for

  • HH-12Booking funnel export from PriyaAsked 12 Nov · the findings can go ahead without it
  • NG-04Venue confirmation from LenaAsked 16 Nov · her question about chairs and screens has a reply drafted
  • DS-02The website restart on 1 DecJohn's own pause, so the Harbor Health report comes first

Someday · not a project yet

  • 01Third client, from the guild sessionStarts as a project when a conversation is booked
  • 02Three-part course pilotPlanned for 11 Jan – 5 Feb, built on Thursday's session
  • 01Fourth clientMarch to June 2027
  • —PROPOSAL skillAfter the third proposal written by hand

Where this comes from. One file per project in 05-focus/projects/: the goal, what finished looks like, the status, the next step and a task table. The focus map reads the three outcomes and the items that are not projects yet from 05-focus/north-star.md, and this week's focus from 05-focus/_focus-CONTEXT.md. Client facts live in 02-memory/clients/ and are linked, not copied.

Continue withfocus7context7

05 / People

GitHub · crm7 ↗

One private workspace per relationship

Pick a person to see what John knows about working with them, the projects and tasks they are part of, and every meeting with its outcome.

PS
John's private workspace

John & Priya Shah

Product lead, Harbor Health · last contact 13 Nov · next: readout on Friday

Only John. Priya cannot open this workspace. Notes and analysis stay in John's folder.

How we work together

A 30-minute check-in every Friday. She likes a short written summary before each call, and decisions confirmed by email.

What she cares about

Fewer abandoned bookings, and evidence her board will accept. If you can show me where they give up and why, that's the board slide.

Worth continuing

A redesign sprint after the readout. Not proposed yet; wait for her reaction on 20 Nov.

How conversations turn into work · last 4 calls

  • Purpose stated4/4
  • Outcome written down4/4
  • Owned next step3/4
  • Follow-up date set2/4

Counted from the MEETING7 notes, not from impressions.

HHHarbor Health discoveryOn track
  • HH-11Findings section of the reportJohnDoing · Fri
  • HH-12Analytics export of the booking funnelPriyaWaiting since 12 Nov
  • HH-13Invoice #014 reminderJohnOverdue
  • HH-10Synthesis v2 of P05–P06MakerDoing
  • 13 Nov
    Weekly check-inReadout time to be confirmed by email; analytics export promised by 18 Nov
    MEETING7 note
  • 6 Nov
    Weekly check-inInterviews P05 and P06 booked
  • 30 Oct
    Weekly check-inFirst four interviews reviewed
  • 12 Oct
    Kick-offScope agreed: 12 interviews, report, prototype
AK
John's private workspace

John & Ada Kim

Head of product, Brightwater · last contact 12 Nov · next: retainer proposal by Thursday

Only John. Ada cannot open this workspace.

How we work together

Calls only when there is a decision to make; everything else in writing.

What she cares about

The December onboarding release landing well. We need someone who keeps talking to users while we ship.

Worth continuing

Usability testing after the December release, as part of the retainer.

How conversations turn into work · last 3 calls

  • Purpose stated3/3
  • Outcome written down3/3
  • Owned next step3/3
  • Follow-up date set3/3
BWBrightwater retainerOn track
  • BW-05Send the retainer proposalJohnOpen · Thu
  • BW-04Invoice #013 for the onboarding redesignJohnDone 16 Nov
  • 12 Nov
    Retainer scope callTwo days a week for three months; testing in or out decided after 18 Nov
    MEETING7 note
  • 30 Oct
    Onboarding handoverRedesign accepted, ships in December
LO
John's private workspace

John & Lena Ortiz

Organiser, Northline design guild · last contact 16 Nov · next: venue for Thursday

Only John. Lena cannot open this workspace.

How we work together

WhatsApp for logistics, one planning call per event.

Worth continuing

A three-part course for the guild in January, if Thursday goes well.

One call so far: not enough to count conversation habits yet.

NGNorthline guild sessionVenue not confirmed
  • NG-03Session slidesJohnOpen · Wed
  • NG-04Venue confirmationLenaWaiting since 16 Nov
  • 16 Nov
    Session planning call40 seats, hands-on, no recording
    MEETING7 note
TR
John's private workspace

John & Tomás Ruiz

Independent product coach · first contact today · next: reply with two times

Only John. Tomás cannot open this workspace.

How we met

He commented on John's LinkedIn post on 11 Nov and then wrote to him. Works with early-stage health startups.

Worth continuing

Possible referrals in both directions.

No project yet. A project starts only when John approves its card.

No meetings yet.

JB
John's private workspace

John & Jonas Berg

Founder, small fintech · former client · last contact 6 Oct

Only John. Jonas cannot open this workspace.

History

John ran a discovery sprint for him in spring 2026.

Worth continuing

He mentioned a new product line in October. Check-in overdue

No open project. The spring project is archived.

  • 6 Oct
    Coffee callNew product line planned for 2027; "talk again in November"
MC
John's private workspace

John & Mei Chen

Host, Design Notes podcast · last contact 27 Oct

Only John. Mei cannot open this workspace.

Worth continuing

John is a guest in the January episode, about interviews with front-desk staff.

No project. Sending notes for the episode becomes a task when it is due.

  • 27 Oct
    Pre-callEpisode recorded on 14 Jan

Where this comes from. One file per person in 02-memory/people/, with three parts: overview, projects and tasks, meetings. Tasks are linked from the project files, never copied. Meetings link to the dated notes in 02-memory/meetings/. Phone numbers, emails and contracts stay in the email and phone apps.

Continue withmeeting7comms7

06 / Marketing · GEO7 · Analytics

GitHub · geo7 ↗

Is Doe Studio being found, and by whom?

Two views of one marketing folder. GEO7 checks whether AI assistants recommend the studio. Analytics brings the site, search and ad numbers into one weekly view.

Once a month, the same five buyer questions go to the same five AI assistants. Three answers arrive automatically through APIs; two are pasted by John. The result shows where Doe Studio is recommended, and what to fix first.

November · 1 Nov17%recommended in 4 of 24 answers
October8%2 of the same 24 pairs
Change+8 ptssame questions, same assistants
Next check1 DecFebruary target: 6 of 25

Five questions × five assistants · November

Whether each AI assistant recommended Doe Studio for each question in November
Buyer question ChatGPTmanual GeminiAPI ClaudeAPI PerplexityAPI DeepSeekmanual
1Who can run user research for a small health tech startup in Europe?———#4—
2How do I find out why users drop out of our booking flow?—named—#2—
3Freelance product designer for fintech onboarding?—————
4UX research consultant for a six-week discovery sprint?#3————
5Workshop on using AI for design teams?————error
  • #2Recommended, with its place in a numbered list
  • namedRecommended, no numbered list
  • —Not recommended
  • errorNo answer; not counted, never a zero

What to fix first · order of work

  1. P1Services page: answer question 2 in plain words, with a short FAQWebsite project, restarts 1 Dec
  2. P1Second case study: Brightwater onboardingNeeds Ada's written OK
  3. P2A recap page for the Northline sessionAfter 19 Nov
  4. P3A guest article for the guild newsletterNot started

What the assistants cite

  • 3×The Calm Product newsletter archive
  • 2×The Brightwater case study
  • 1×The Northline guild talk page

Other channels

  • 1,100newsletter subscribers51% average opens · issue 8 was four days late
  • 2,400LinkedIn followers2 posts in November

This shows the idea. Running GEO7 for real needs its own setup: the five questions from real buyers, API access within an agreed budget, and a fixed monthly routine. A place is counted only from a numbered list, and months are compared only on the same pairs.

Where this comes from. 02-memory/marketing/geo7-baseline.md holds the questions, the collection rules, the monthly results and the order of improvements. The channel numbers are in 02-memory/marketing/_marketing-CONTEXT.md. The improvements become tasks in the website project when it restarts.

One weekly view of the site, search and ads. Every Monday at 07:30 the morning run collects read-only exports: Google Analytics 4, Search Console, Google Ads and Meta Ads arrive through connectors that cannot change anything. ChatGPT Ads has no connector, so John pastes its export. Below: the last four full weeks, 19 Oct – 15 Nov, compared with the four weeks before.

Collected Monday 16 Nov, 07:30 4 sources automatic, read-only 1 source pasted by John: ChatGPT Ads
Enquiries159 contact forms + 6 booked calls +36% vs 11 in the four weeks before
Site visits1,252Google Analytics 4, all channels +11% vs 1,125
Search clicks478Search Console, Google only +16% vs 412
Ad budget used71%of one month's cap, over the four weeks −13 pts vs 84%

Enquiries per week, by channel

  • Search
  • Ads
  • Newsletter
  • Direct & referral

Weeks start on Monday. Hover a block for its number.

Table view · enquiries
Week ofSearchAdsNews­letterDirect & ref.Total
19 Oct11013
26 Oct11103
2 Nov21014
9 Nov11125
Four weeks542415
Four weeks before442111

Ads are the three platforms together: Google Ads 3, ChatGPT Ads 1, Meta Ads 0. Each enquiry is a contact form sent or a call booked, counted once, in the channel of the visit where it happened.

What this tells John

  1. 1AI assistants have started sending people. Links in AI answers brought 47 visits, up from 12, mostly to the Brightwater case study that GEO7 found cited. Two of the four direct and referral enquiries named it.A second case study is the next step · website project restarts 1 Dec
  2. 2Search is the steadiest channel: 478 clicks, up 16%, and 5 of the 15 enquiries. The plain-words services page that GEO7 asks for should help search as well.Same website restart · 1 Dec
  3. 3The newsletter only looks lower. Issue 8 went out four days late, on 14 Nov, so most of its visits fall into next week. Issue 7 and issue 8 each brought one enquiry.Issue 9 on 24 Nov · outline due 23 Nov

Every source · 19 Oct – 15 Nov

For each source: what it measures, its own numbers for the four weeks, the site visits and enquiries it brought, and its state
SourceWhat it measuresIts numbers, four weeksSite visitsEnquiriesState
Google Analytics 4automatic · read-only The whole site: visits by channel, and two enquiry events, form sent and call booked 9 contact forms · 6 booked calls 1,252+11%, was 1,125 15was 11 Collected Monday
Search Consoleautomatic · read-only Organic Google search: how often the site shows, and is clicked 21,400 impressions · 478 clicks, +16% · average position 18.4, was 19.6 520all search engines 5was 4 Rising
Google Adsautomatic · read-only Search ads: impressions, clicks, share of the monthly cap 6,850 impressions · 182 clicks, −7% · 41% of the cap · 61 clicks per enquiry 164was 170 3was 3 Running, within the cap
ChatGPT Adsmanual · pasted by John Ads in ChatGPT answers: impressions, clicks, share of the cap Since 26 Oct · 3,100 impressions · 51 clicks · 17% of the cap 46new 1new First month, too early to judge
Meta Adsautomatic · read-only Facebook and Instagram ads: impressions, clicks, share of the cap 19 Oct – 8 Nov · 9,400 impressions · 104 clicks, −36% · 13% of the cap 88was 141 0was 1 Paused 9 Nov: no enquiries in three weeks
Newsletterfrom GA4 link tags Visits from links in Calm Product issues Issue 7, 27 Oct: 71 visits in its first week · issue 8, 14 Nov: 42 in two days 136−9%, was 150 2was 2 Issue 8 four days late
Direct & referralfrom GA4 Typed addresses and links from other sites, AI answers included 47 visits from links in AI answers, was 12: Perplexity 21, ChatGPT 17, Gemini 9 298+52%, was 196 4was 1 AI links rising

The channel rows add up to the Google Analytics row: 520 + 164 + 46 + 88 + 136 + 298 = 1,252 visits and 5 + 3 + 1 + 0 + 2 + 4 = 15 enquiries. Search Console counts Google only, so its 478 clicks are fewer than the 520 visits from all search engines. Ad budget is shown only as a share of the monthly cap: 41% + 17% + 13% = 71%.

The dashboard shows the numbers; decisions go into tasks. Nothing here can change a campaign. When a number calls for action, John decides, and the decision becomes a task with an owner and a date in the project it belongs to. That is how Meta Ads were paused on 9 Nov.

Where this comes from. 02-memory/marketing/analytics.md holds the sources and how each is collected, the weekly table, the channel table and these findings. The morning run updates it every Monday from the read-only exports and John's pasted ChatGPT Ads file. Money stays in the ad accounts and the books: spend appears only as a share of the monthly cap, and results as clicks per enquiry.

Continue withads7focus7

07 / Finance · example

GitHub · cc7 ↗

What came in, what went out, what needs a decision

A small financial view for Doe Studio. Cash flow covers the last full month; unpaid invoices have their own as-of date. Every amount is invented.

October 2026 · full monthExample review · 16 Nov 20261 overdue invoice needs John

Cash collected

€12,400

October · full month

Cash paid

€7,300

Same period · all categories

Net cash flow

+€5,100

Collected minus paid · not profit

Invoices unpaid

€3,800

3 invoices · as of 17 Nov

Cash movement · three full months

Collected and paid, kept separate

Bar scale: €14,000. The table contains every value.

Show the cash-flow table
Invented cash flow · EUR
MonthCollectedPaidNet
Aug€8,600€6,100+€2,500
Sep€9,400€6,800+€2,600
Oct€12,400€7,300+€5,100

Invoice ageing · as of 17 Nov 2026

An invoice is not money received

  • Within agreed terms2 invoices€3,200
  • 1–30 days overdue1 invoice€600
  • Over 30 days overdue0 invoices€0

Total unpaid€3,800

Ageing uses agreed due dates, not the date the invoice was created.

The next useful decision

Review the €600 reminder

John checks the invoice and the drafted reminder. Sending it is still his decision; the task stays HH-13 in the existing project.

Open the task view →

Source and boundary

An approved summary, no bank connection

In this example, John approves a monthly accounting summary. Raw bank records stay in accounting. A real setup follows its owner’s privacy rules and may show only a link instead.

Design reference. The ERPNext Payments dashboard separates payments and invoice ageing. This view adapts that idea to the fictional studio; no accounting service is installed. ERPNext · GitHub ↗

Continue withcontext7exit7

08 / RADAR · example

Opportunities worth a look this week

Once a week John’s AI looks through the sources he chose for four kinds of opportunity: paid work, partnerships, speaking slots and communities to join. Each find gets a fit score, a reason and one thing to check. John decides; nothing is applied for, sent or joined automatically.

Week 47 · reviewed Mon 16 Nov31 found · 7 shown2 wait for John’s decision

Work & projects

2

1 strong fit

Partnerships

1

1 strong fit

Speaking

2

1 strong fit

Communities & events

2

0 strong fits

  1. 88%fit

    SpeakingConference call-for-papers page · deadline 5 Dec · submitted 16 Nov

    Call for speakers: regional UX conference · 25-minute practitioner talks

    Why it fits: His post “Three questions I ask in every onboarding interview” is already a talk outline.

    Check first: Travel is not covered; decide whether the trip is worth it.

    Applied, waiting
  2. 86%fit

    Work & projectsJob alert e-mail · posted 13 Nov · remote, EU hours

    Fractional research lead for a health-tech scale-up · two days a week

    Why it fits: Onboarding interviews and health clients are exactly his recent work.

    Check first: The day rate is not listed. Ask before applying.

    Worth a look
  3. 78%fit

    PartnershipsNorthline guild chat · 12 Nov · introduced by Lena Ortiz

    A two-person web studio wants a research partner for client discovery

    Why it fits: They build, John researches; referrals can go both ways.

    Check first: Who owns the client relationship, and how fees are split.

    Worth a look
  4. 74%fit

    Communities & eventsMeetup page · next session 3 Dec

    Monthly product research meetup · online · open call for 10-minute case talks

    Why it fits: A steady peer group for one evening a month.

    Check first: Whether sessions are recorded; client examples must stay anonymous.

    New
  5. 72%fit

    Work & projectsSaved search on a freelance marketplace · posted 15 Nov · fixed price

    Freelance usability study for a checkout flow · three weeks

    Why it fits: Clear scope; it fits in December, after the Harbor Health readout.

    Check first: Who recruits the test participants, and the client's past reviews.

    New
  6. 69%fit

    Communities & eventsIncubator newsletter · event on 21 Jan 2027

    Health-tech demo day · volunteer mentors wanted

    Why it fits: He meets early-stage health founders, the field Tomás Ruiz works in.

    Check first: A half-day commitment; sales pitches are not allowed.

    New
  7. 61%fit

    SpeakingHost's LinkedIn post · 10 Nov

    Podcast guest: research operations in small product teams

    Why it fits: A good audience for a small time cost.

    Check first: Overlap with the January Design Notes episode with Mei Chen.

    New

Below 50% stays hidden unless John asks. 3 finds he marked “Not for me” are hidden too, and they sharpen next week’s criteria.

John’s criteria

He sets the rules; the AI applies them

  • Work: user research, onboarding and service design; up to two days a week; from €650 a day; remote or within two hours’ travel.
  • Partnerships: complementary skills, a named client case and a clear split.
  • Speaking: practitioner talks, not paid sponsor slots; at most one trip a month.
  • Communities: active groups with a real date in the next eight weeks.

Sources he chose: two job alert e-mails, one saved freelance search, conference call-for-papers lists, the Northline guild chat and three newsletters.

Last four weeks

Which stream actually works

Invented weekly review totals
StreamFoundWorth a lookAppliedResult
Work & projects58621 call booked
Partnerships921Waiting
Speaking16531 talk accepted
Communities & events1242Joined 1 group

So far speaking brings the most results and work alerts the most noise. Statuses are John’s marks: New · Worth a look · Applied, waiting · Declined · Won · Not for me.

Boundary

The AI finds and explains. John acts.

It reads only the listed sources, when John asks or in his weekly review, and keeps each find with its link and date. It does not apply, send messages, register, join groups, create alerts or change profiles. A new source needs his OK.

After a yes

A good find moves into the existing system

An accepted opportunity becomes a project card or a person in People, with the next action in Work. The RADAR list itself stays a weekly shortlist, not a second task list.

Open People →

Where this comes from. Criteria and sources are in 02-memory/growth/radar/_radar-CONTEXT.md; each weekly list, with links, dates and John’s marks, is in 02-memory/growth/radar/2026-w47.md. Every item here is invented.

Continue withwarm7focus7

09 / Skills

GitHub · skills7 ↗

Methods the AI follows, written once

A skill is a Markdown file: when to use it, the steps, and what "done" means. Any AI that can read the folder can follow it, and the same work runs the same way the second time. The library skills live on GitHub, so anyone can install the same one.

Installed with the starter

MEETING7Friday

Meetings into decisions

One dated note per call: decisions, their words, open questions, commitments to approve.

After each call
EXIT7yesterday

Close the session

Anything that lives only in the chat goes into a file before the day ends.

End of each working day
EVAL7last week

Score one finished thing

Against its own goal, with evidence and a ranked list of what to fix first.

On request

Added from the GitHub library, after real work showed the need

Written by John

INTERVIEWyesterday

Interview synthesis

Themes with exact quotes and counts. Seven checks run before a client sees it.

On request · Maker writes, Checker reviews
PROPOSALnot written yet

Next candidate

Done by hand twice so far. It becomes a skill after the third time.

From the "Missing" lists
NEWSLETTERnot written yet

After that

An issue outline from the notes saved since the last issue.

From the "Missing" lists

Where this comes from. 03-skills/_skills-CONTEXT.md lists every trigger and where each file came from. The five starter methods are unchanged copies from the public library on GitHub (arete-aios/aios). COMMS7 and GEO7 were added from the same library when the work needed them; each has its own screen in the menu. INTERVIEW was written after three interview rounds.

Continue withcontext7eval7

09 / Skills · COMMS7

GitHub · comms7 ↗
← All skills

Who needs a reply, and what you are waiting for

COMMS7 is a written method, not an app. Every morning it reads John's chosen channels, sorts each conversation by who has to act next, drafts the replies he asks for and checks a short watch list of posts for comments worth answering. The AI reads and drafts; John decides and sends.

Open COMMS7 on GitHub

When it runs

Every morning, with a full review on Mondays

Inside the Maker's 08:05 morning run, or whenever John types COMMS7.

Who acts

You decide · AI reads and drafts · code collects

In this example, “you” is John.

What it never does

Send, post or delete

A draft stays a draft until John sends it. Nothing is accepted or promised in his name.

Goal

Know who needs a reply, what you need to do and what you are waiting for, with one clear next step for each item and no open promise lost.

How one review runs

Seven steps from the request to the saved decisions. Two of them are John's, one is plain code, and the AI does the reading, sorting and drafting in between.

  • You: round, heavy outline
  • AI: tinted box
  • Code: dashed box
  • File in John's folder
  • Dashed arrow: repeats
  • Dotted line: reads or writes

Scroll sideways to see the whole flow. The same steps follow as text.

GOAL AND SETTINGS SOURCES · CODE · AI DECISIONS AND WHAT IS SAVED GOAL Know who needs a reply, what to do and what you are waiting for. Each item gets one clear next step. TWO JOBS IN ONE RUN Conversations · who has to act next: John, them, or nobody? Watched posts · is there a question worth an answer? RHYTHM AND LIMITS Every morning in the 08:05 run; the full review on Mondays. Never sends, posts or deletes. Next morning: saved decisions + new messages 1 YOU Ask for COMMS7 chat · 08:05 run 2 AI Set the scope reads settings 3 CODE Collect the messages five channels 4 AI Read and sort whole threads 5 AI Show the few that matter drafts, not sent 6 YOU Decide and send item by item 7 AI Save and close project files Repeat per item reads reads reads writes FILE _comms-CONTEXT.md sources, days back, rhythm, watch list FILE _tools-CONTEXT.md five connections, read and draft only FILE north-star.md and this week's focus, to rank FILE 05-focus/projects/ decisions, and a date to check again

The same seven steps in words

  1. 01Ask for COMMS7You · chat or the 08:05 runType COMMS7 and say what to look at: everything, only replies, or one conversation. On weekdays the Maker's morning run starts it for him; on Mondays it is the full review.
  2. 02Set the scopeAI · reads the settingsThe AI takes the sources, how far back to read and the watch list from _comms-CONTEXT.md, plus the open items from the last review. It asks only what would change today's review.
  3. 03Collect the messagesCode · five channelsThe connections fetch the last 14 days (the guild chat 7) and new comments on watched posts. A small script removes duplicates and marks what is new since the last review; it does not judge what a message means.
  4. 04Read and sortAI · whole threadsThe AI reads each whole thread, including what John already sent, and gives it one state: reply needed, waiting on them, for information, or closed. Watched posts are checked for questions.
  5. 05Show the few that matterAI · drafts, not sentUp to five items, real deadlines first, then the north star and this week's projects. Each has a reason, one next step and its source; the replies John asked for are drafted and left unsent.
  6. 06Decide and sendYou · item by itemFor each item John replies, does it, waits, moves it to later or skips it. He sends and posts himself; the AI updates that item and shows the next one.
  7. 07Save and closeAI · project filesEach decision goes into the project file the conversation belongs to, and anything waiting gets a date to check. The AI reads the files back and closes with EXIT7, so tomorrow starts from here.

Principles

What the method holds to, every run.

Who acts next
Every conversation gets exactly one state: reply needed, waiting on them, for information, or closed.
Deadlines first
Real deadlines and consequences come first, then the north star and this week's projects; unread counts and loud messages do not decide, and no due date is invented.
Open promises stay open
“I'll send the slides tomorrow” is still open after the thank-you goes out; silence or archiving does not close it.
A draft is not a reply
Nothing counts as answered until John sends it, and a suggested comment is only a draft until he posts it or skips it.
Say what was not read
If a channel could not be read, the review says so instead of guessing what was in it.

In John's folder

The method stays the same for everyone. John's choices live in two short files, and today's run left a trail that matches the Messages screen.

Settings

02-memory/comms/_comms-CONTEXT.md

Sources
Gmail, WhatsApp Business, the Facebook page and the guild chat; LinkedIn in a session John opens
How far back
14 days, the guild chat 7; open promises however old
Rhythm
Every morning at 08:05; the full review on Mondays
Watch list
John's own posts from the last 14 days, plus posts by the six people in People about research, onboarding or design teams

Connections

04-tools/_tools-CONTEXT.md

  • Gmail · read, create drafts. Cannot send.
  • WhatsApp Business · read. Cannot send.
  • Facebook page · read messages and comments. Cannot post.
  • #Guild chat · read #events and direct messages.
  • inLinkedIn · read in a browser session John opens. Nothing is posted.

To rank, it also reads 05-focus/north-star.md and this week's focus, and a person's file in 02-memory/people/ only when a conversation needs it.

Today's run · Tuesday 17 November

08:05 read: Gmail, WhatsApp, Facebook page, guild chat; focus and project files
08:06 sorted: 3 reply needed · 2 waiting on them · 4 for information
08:06 2 replies drafted, not sent: Priya (Harbor Health), Lena (Northline)
08:06 1 reply suggested, John writes it: Tomás (LinkedIn, from Monday's review)
08:06 4 watched posts: 1 question to answer on LinkedIn, 1 on the Facebook page

Ada Kim's answer on the Brightwater scope stays “waiting on them” until their planning on 18 Nov. One general post was skipped on 16 Nov: not a fit with the north star.

What John approves

2 reply drafts waiting for approval

  • Priya Shah · confirm 11:00 for Friday's readout, then move the calendar slot himself.
  • Lena Ortiz · 40 chairs, one screen, HDMI.
  • Tomás Ruiz · John writes this one himself and offers two times next week.
  • Two draft answers on his own posts, LinkedIn and Facebook: post as they are, edit, or skip.
  • Two optional comments on Priya's and Tomás's posts: write one, or skip.

What is written back

  • Drafts · Gmail drafts for email; for the other channels on the Messages screen, ready to copy. The AI never sends them.
  • Decisions · into the project file each conversation belongs to: the new readout time goes into 05-focus/projects/harbor-health-discovery.md.
  • Waiting items · a date to check again, so Ada's answer is looked for after 18 Nov and not chased earlier.
  • Message text · stays in the channel. The folder keeps only the decision and where to find the thread.

Where this comes from. The method is the public COMMS7 skill from the library; 03-skills/_skills-CONTEXT.md records its source. John's own choices are in 02-memory/comms/_comms-CONTEXT.md and the connections in 04-tools/_tools-CONTEXT.md, so changing a source, the days read back or the watch list means editing one of those files, never the skill.

Continue withskills7meeting7

09 / Skills · GEO7

GitHub · geo7 ↗

← All skills

When buyers ask an AI, is Doe Studio named?

GEO7 is a public method for being found by the right buyers, in search and in AI answers. John uses it for one narrow question: do AI assistants recommend Doe Studio when a buyer asks? Once a month the same answers are collected and scored the same way, and the gaps become a short, ordered list of fixes for his website.

Open GEO7 on GitHub · the public method and its helper scripts

When it runs

1st of each month

Last run 1 Nov · next 1 Dec. Three answers arrive by API, two are pasted by John.

Who acts

  • Youchoose the questions and approve fixes
  • AIasks the assistants and scores the answers
  • coderepeats the same checks, the same way, every month

What it never does

  • Publish a change to the website
  • Buy anything: no new tool, no ads, no spend beyond the agreed API budget

Goal

A buyer who asks an AI assistant one of the five questions hears Doe Studio named, and that answer leads to a call with John.

How one monthly round runs

Six stages from the public method, applied to John's five questions. Code repeats the checks, AI selects and explains, John decides. The saved files are what each round leaves behind.

Scroll sideways to see all six stages.

GOAL A buyer who asks an AI assistant one of the five questions hears Doe Studio named, and that answer leads to a call with John. EVIDENCE AND INTENT 1–2 DIFFERENCE AND FIX 3–4 IMPLEMENTATION AND MEASUREMENT 5–6 audit.py reads public pages ai_visibility.py repeats the count 1 YOU + AI Check the site audit 2 YOU + AI Pick questions keywords 3 AI Who is named competitors 4 AI Brief the fix content 5 YOU Approve, build implementation 6 AI + CODE Ask and score measurement Site notes dated, with scope Page briefs P1–P3 + a check geo7-baseline.md one row per month Next month: same pairs, and the result sets the next priority
  • You decide · rounded, thick border
  • AI works · square corners, thin border
  • Code repeats a check · dashed box
  • Saved file · folded corner
  • Dashed arrow · next month's round
  1. Check the siteYou + AI

    Stage: audit

    John names the site, the market, the folder for results and the monthly budget. AI reads his public pages and notes, with the date, what each page actually answers.

  2. Pick questionsYou + AI

    Stage: keywords

    AI drafts candidate questions from real buyer emails and call notes; John keeps five, each tied to one page and one next step. The same five stay from month to month.

  3. Who is namedAI

    Stage: competitors

    From the saved answers, AI lists whom the assistants recommend instead and which pages they cite. A name with no saved answer behind it stays a guess.

  4. Brief the fixAI

    Stage: content

    Each gap becomes a short page brief: one plain answer, a real example, a small FAQ and a check that shows it is done. P1–P3 sets the order; the text stays a draft until John checks the facts.

  5. Approve and buildYou

    Stage: implementation

    John approves each change, and the website project builds it when it restarts on 1 Dec. GEO7 itself publishes nothing.

  6. Ask and scoreAI + code

    Stage: measurement

    On the 1st, the same five questions go to the same five assistants: three by API, two pasted by John. AI marks each answer placed, named or not; ai_visibility.py repeats the count and leaves errors out.

Principles

Five rules that keep one month comparable with the next.

Same pairs, month to month
Months are compared only on the same question and assistant pairs. A new question or assistant starts its own series.
A place needs a numbered list
A place such as #2 is recorded only when the answer is a numbered list. Otherwise a recommendation counts as named, without a place.
An error is never a zero
A failed or empty answer is left out and noted. That is why November reads 4 of 24, not 4 of 25.
Blockers before volume
If a page cannot be read, or answers a different question, fix that before writing anything new.
Buyer intent over broad topics
One narrow question a real buyer asks is worth more than a broad topic that many people search for.

P1–P3 is an order of work. It is judged from the buyer's path, the strength of the evidence and whether the fix can be checked. It is not a measured effect on AI answers; when the evidence is thin, the first task is a check.

In John's folder

One file holds the settings and the history. The latest run is shown in Marketing.

The file

02-memory/marketing/geo7-baseline.md

Questions
The five buyer questions, in the words buyers used, each tied to one page and one next step. A change starts a new series.
Collection rules
The 1st of each month. Gemini, Claude and Perplexity by API within the agreed budget; ChatGPT and DeepSeek pasted by John. Every answer saved with its date.
Monthly results
One row per month on the same pairs: October 2 of 24, November 4 of 24. Errors are listed, not counted.
Order of improvements
P1–P3, each with the page, who does it and the check that shows it is done.

What the 1 Nov run produced

17%recommended in 4 of 24 answers · October 8%, 2 of the same 24 pairs

  • Perplexity · #4 for question 1, #2 for question 2
  • Gemini · named for question 2, without a numbered list
  • ChatGPT · #3 for question 4
  • Question 3, fintech onboarding · no assistant named him
  • DeepSeek, question 5 · error, left out of the count

Cited most: the Calm Product newsletter archive (3×), the Brightwater case study (2×), the Northline guild talk page (1×).

Into the website project · 1 Dec

  1. P1Services page: answer question 2 in plain words, with a short FAQ
  2. P1Second case study: Brightwater onboardingNeeds Ada's written OK
  3. P2A recap page for the Northline sessionAfter 19 Nov

Each task carries the check from its brief. Outside the site: P3, a guest article for the guild newsletter, not started. The 1 Dec check still measures today's site; the February target is 6 of 25.

What John approves

  • The five questions, and any change to them
  • The API budget for the three automatic answers
  • Every page change, before it goes live
  • A client's name on a public page: the Brightwater study waits for Ada's written OK

On the 1st he also pastes the two manual answers, from ChatGPT and DeepSeek.

Where this comes from. The six stages, the rules and the helper scripts come from the public GEO7 skill, narrowed here to one question a freelancer can ask every month. John's questions, rules and results live in 02-memory/marketing/geo7-baseline.md; 03-skills/_skills-CONTEXT.md lists the skill and where it came from. The studio, the people and every number are invented.

Continue withskills7focus7

10 / Tools

GitHub · tools7 ↗

What the AI can reach, and how far

Connectors and plugins give the AI eyes on the work. Each gets the narrowest access that does the job. Sending, posting, paying and deleting always wait for a yes.

Connectors and plugins

ToolForAccessUsed byState
Gmailconnector ForClient and partner email AccessRead, create drafts. Cannot send. Used byCOMMS7, MEETING7 ConnectedChecked Mon 16 Nov
Google Calendarconnector ForThe week, meeting prep AccessRead only. Used byFOCUS7, MEETING7 Reconnect by 20 NovAccess expires in 3 days
Google Driveconnector ForClient documents and interview notes AccessRead the Clients folder only. Used byINTERVIEW, MEETING7 ConnectedChecked Mon 16 Nov
WhatsApp Businessplugin ForQuick client questions AccessRead messages. Cannot send. Used byCOMMS7 ConnectedChecked Mon 16 Nov
Facebook pageconnector ForPage messages and comments on chosen posts AccessRead messages and comments. Cannot post. Used byCOMMS7 watch list ConnectedChecked Mon 16 Nov
Guild chatconnector ForNorthline events and direct messages AccessRead #events and direct messages. Used byCOMMS7 ConnectedChecked Mon 16 Nov
Figmaconnector ForDesign files for case studies AccessRead files John shares. Used byMaker ConnectedChecked Thu 12 Nov
Zoomconnector ForCall recordings AccessRecordings John selects, nothing else. Used byMEETING7 ConnectedChecked Fri 13 Nov
LinkedInbrowser ForComments on chosen posts AccessRead in a session John opens. Nothing is posted. Used byCOMMS7 watch list Manual by choiceNo connector on purpose
Google Analytics 4connector ForSite visits by channel, enquiry events AccessRead reports only. Used byAnalytics view ConnectedLast pull Mon 16 Nov, 07:30
Search Consoleconnector ForGoogle search: impressions, clicks, position AccessRead reports only. Used byAnalytics view, GEO7 ConnectedLast pull Mon 16 Nov, 07:30
Google Adsconnector ForSearch ads results and share of the cap AccessRead reports only. Cannot change a campaign. Used byAnalytics view ConnectedLast pull Mon 16 Nov, 07:30
Meta Adsconnector ForFacebook and Instagram ads results AccessRead reports only. Cannot change a campaign. Used byAnalytics view ConnectedAds paused by John on 9 Nov
ChatGPT Adsmanual export ForAds in ChatGPT answers AccessNo connector. John pastes the weekly export. Used byAnalytics view ManualSince 26 Oct
Banknone ForBank records stay in accounting AccessNever connected. Used by— Not connectedA statement once a month, by hand

AI assistants

AssistantForAccessUsed byState
Claudemain assistant ForMaker and Checker, with the folder attached AccessReads the folder; writes drafts into it. Used byAll skills; GEO7 by API MainDesktop app with the folder; API for GEO7
ChatGPTapp ForSecond opinion AccessApp; GEO7 answers are pasted into the folder. Used byGEO7 answers pasted by John In use
GeminiAPI ForGEO7 answers AccessAPI key with a monthly cap. Used byGEO7, 1st of each month Connected
PerplexityAPI ForGEO7 answers AccessAPI key with a monthly cap. Used byGEO7, 1st of each month Connected
DeepSeekapp ForGEO7 answers AccessApp; answers pasted into the folder. Used byGEO7, pasted by John In use

Where this comes from. 04-tools/_tools-CONTEXT.md: what each tool is for, which access was granted, and when it was last checked. A tool that is not written down there is not used. Files carry over between AI apps; connections and permissions have to be granted again in each one.

Logos from Simple Icons (CC0); the trademarks belong to their owners. Services without a logo there are shown with letters.

Continue withaudit7context7

11 / Folder map

The whole brain, in one folder

This is the complete example folder, with the same layout the masterclass starter creates. Each folder has a dot for its CONTEXT7 state, and the folders that are not there yet are shown too. The folder is the brain; this page only shows it.

  • CONTEXT file lists everything, links work
  • CONTEXT file misses something
  • No CONTEXT file
  • Not there yet, or never
  • doe-studio-brain/47 filesToday's check: 11 green, 1 yellow, 1 red
    • constitution7The root files: who John is, the rules, the two agents

      CONSTITUTION.md · AGENTS.md · ONBOARDING.md · AIOS-INSTRUCTIONS.md · Agent-MAKER.md · Agent-CHECKER.md · README.md

    • 02-memory/24What the business knows
      • businesses/doe-studio/1The business profile, seven blocks
      • clients/2Harbor Health, Brightwater
      • people/7Six people: overview, projects, meetings
      • meetings/4MEETING7 notes
      • comms/1COMMS7 sources and the watch list
      • marketing/3Channels, the GEO7 baseline, analyticsanalytics.md not listed yet
      • interviews/harbor-health/5Notes P05 and P06, Maker draft, Checker review
      • proposals/Not yet: comes with the PROPOSAL skill
      • finance/Not used: records stay in accounting
    • 03-skills/7Five starter methods and INTERVIEW; COMMS7 and GEO7 by source
    • 04-tools/1Connections and their limits
    • 05-focus/7North star, this week, five project files
    • 06-archive/0Finished material, empty so farno CONTEXT file
    • 00-system/aios/1Links to the online library · starter file, not checked

Today's folder check · CONTEXT7

11 green · 1 yellow · 1 red

Morning run 08:06 · 47 files · every link resolves. The full CONTEXT7 review runs on Mondays.

Fixes, as tasks

  • 1List analytics.md in 02-memory/marketing/_marketing-CONTEXT.mdMaker, today · a one-line fix
  • 2Add 06-archive/_archive-CONTEXT.mdBefore the first project is archived

One home per thing

Every task lives in one project file

People and client files link to tasks, never copy them.

Private by default

Participants are IDs only

Recordings and contracts stay in the client's own drive.

Where this comes from. The counts and the dots are the example folder's real state. CONTEXT7 checks that each folder's CONTEXT file lists its files and that the links point at files that exist; small collections such as clients/ are listed by their parent. In the starter the constitution files sit at the root of the folder; here they are shown together as one group. A folder that is not there is shown on purpose, so the gaps are visible too.

Continue withaudit7exit7

12 / AI team

GitHub · eval7 ↗

Two agents, one brain

The Maker does the work, the Checker reviews it against the same files. Both follow the same constitution. John decides what leaves the folder.

Maker

Agent-MAKER.md · on request, and the morning run
  • Does: one task at a time, by following the skill and the constitution
  • Reads: the constitution, the skill, the client and project files
  • Never: sends anything, or approves its own work

Checker

Agent-CHECKER.md · after every draft
  • Does: PASS, or a numbered FIX list where each fix names the file and rule
  • Reads: the draft and every source file it cites
  • Never: rewrites the work

Last loop · Harbor Health interviews P05 and P06

Mon 10:40 Maker   synthesis draft v1 saved, 3 themes
Mon 11:15 Checker FIX, 2 items: one quote not word for word; one theme has only 1 participant
Tue 08:06 Maker   fixing both items for v2 (HH-10)
next     Checker reviews v2 · John decides what goes into Friday's readout

Where this comes from. Two short role cards, Agent-MAKER.md and Agent-CHECKER.md, on top of the shared AGENTS.md. Open each in its own chat with the folder attached. The loop above is saved in 02-memory/interviews/harbor-health/.

Continue withskills7exit7

13 / Health

GitHub · cc7 ↗

Is the system itself in good shape?

Green means John's own definition of healthy is met. Nothing more.

CheckResultState
Constitution within budget606 of 1,000 wordsOK
Focus updated this weekMondayOK
Weekly review closedSundayOK
Every active project has a next step5 of 5OK
CONTEXT files match the folders8 of 9 · marketing/ misses analytics.mdFix today
Every layer folder has a CONTEXT file4 of 5 · 06-archive/ has noneMissing
Calendar connectionExpires in 3 daysCheck

Privacy line, written in the constitution: Raw financial records, contracts, contact details and participant health details stay in their private systems. An approved private dashboard may show financial aggregates.

Where this comes from. The first five checks are the rules under "What healthy means" in CONSTITUTION.md; the next one is its rule to keep a CONTEXT map in each folder. Both folder checks are the dots on the Folder map. The last one comes from 04-tools/. The dashboard only reports whether each rule is met.

Continue withaudit7exit7

Your request for AI

Clipboard access is unavailable. Select the text and copy it, then paste it into your AI.