V1 in five steps, on one page

Six teams are done or closed. Six jobs are left, and every one of them is buildable now except one that waits on your five answers. Then the loop runs once for real.

22tickets open on GitHub (22 closed)
6teams still to build
5steps until V1 is proven
1thing broken: the live reader, today
Update 21-11, midday: the whole build wave is in.

Done and checked: T1 tickets, T2 live reader, T3 coach (it wrote a real brief, graded B), T5 Umgesetzt with the live check, T10 switch (built, the key is not live yet) and T11 openers (closed: 7.7% are not recognised, and that is accepted). The T6 judge design is written too.

Nothing is blocked by an agent. What is left is one five-point answer from you, then building.

One thing is broken right now: the live reader reads nothing.

One chat with 6,000 messages costs more than the $0.30 day on its own, so the reader stops there every minute and never gets to the rest. It is a cost wall, not a logic error. The fix is noted on #37.

The road to V1, in order

Five steps. Each one needs the one before it, so this is a real sequence, not a wish list.

1

You answer the 5 judge pointsHow the judge decides: #38. This is the only step that waits on a person.

you · #38
2

T6 builds the judgeThe design is done; the build starts the moment the five points are in.

agents · T6
3

You give the yes for A33.writer and A34.liveBoth cost cents each. They unlock wording that a person approves and a live check.

you · cents each
4

T8 runs the fake full cycleThe whole loop once on fake data: idea, wording, approval, Umgesetzt, verdict. #35.

agents · T8
5

You and Max run the first real cycleSame loop, real chats, real booking. When it lands, V1 is proven. #36.

you + Max · #36
Alongside, not blocking: the switch test with Max (#31) is done 3 times already. It now also checks that the AI's answer fits what was actually said while the switch was off. That runs beside steps 1 to 5 and never holds them up.

Who feeds whom, and where each one stands

Read left to right. An arrow means "this team's output is what the next one needs." Green is done, blue is building now, grey waits for its input.

Team order: T1 done, T2 T3 T5 done, T4 and T7 building now, T6 T8 T9 next, ending in the first real cycle STEP 1 · done BUILDING NOW AFTER YOUR 5 POINTS AFTER YOUR YES OR A CHECK FINISH T1 gave ticket numbers to T2 T1 gave ticket numbers to T5 T1 gave ticket numbers to T3 T5 gave the time the new wording was first seen live T6 judge is needed for the test cycle T3 gave one real coach brief, graded B Test cycle passes, then the real cycle time seen live real brief works on fake data T1 Tickets: done. GitHub matches reality, cap raised to $60 T1 Tickets done · GitHub matched cap $30 → $60 T2 Live reader built (#25): but reads nothing today, cost wall, fix on #37 T2 Live reader done · reads nothing now (#37) T5 Umgesetzt + live check: done T5 Umgesetzt button + live check · done T3 Coach: done, one real brief graded B T3 Coach real brief, grade B · done T4 Lists (#39, #40, #42): building now T4 Lists #39 #40 #42 T7 Close the built tickets (#17, #25, #26) T7 Close built checks only · #17 #25 #26 T6 Judge: design done, build after your 5 points (#38) T6 Judge design done · build on #38 T8 Fake full cycle (#35), needs your A33.writer and A34.live yes T8 Fake cycle whole loop, fake data · #35 + your yes (cents) T9 Control room (#43), needs your 7 answers T9 Control room see every agent live · #43 + your 7 answers T10 Switch: built, the key is not live yet T10 Switch built · key not live T11 Openers: closed, 7.7% not recognised and accepted T11 Openers closed · 7.7% accepted T12 Backup (#13), not urgent, needs your 2nd Supabase project T12 Backup not urgent · 2nd Supabase #36 First real cycle: V1 is proven #36 first real cycle V1 proven YOU + MAX · the only lane that can move today without an agent Dani's open yeses You 5 judge points (#38) · yes to A33.writer + A34.live #8 screen · #21 B · #24 reader · #43 7 questions · #13 Max's lane Max ✓ openers · ✓ tag (ai_off) · ✓ KI Flow fixes ✓ switch test 3× · now checks the answer fits You and Max run the real cycle you pick, Max pastes
done, checked against the code done, with a caveat agents building now waits for its input needs a person

What is left, one line each

Each team has its own files and branch, so two teams never edit the same thing. Six teams are finished and are not listed here.

T6 Judge

#38after your 5 points

The design is written: it decides on how many people reply to the changed message, with bookings shown beside it. The build is one team, and it starts the day the five points are answered.

Hands on: the verdict at the end of every cycle.

T4 Lists

#39#40#42building now

The said-yes list leaves out known clients, lists never time out, and the daily "must be zero" rules go in. Nothing here needs a person.

Hands on: lists people can trust without re-reading them.

T7 Close the built ones

#17#25#26checks only

Builds nothing. Runs the missing checks on work that is already finished and closes the tickets that pass, so the open list finally means something.

Hands on: a GitHub list that matches reality.

T8 Fake full cycle

#35after T6 + your yes

The whole loop once on fake data: idea, wording, approval, Umgesetzt, verdict. Nothing real is touched, so a mistake costs nothing.

Hands on: proof the loop works before it runs on real chats.

T9 Control room

#43after your 7 answers

Every agent and its cost live on the dashboard, with a pause switch per part. Spec is written; it waits on the seven answers.

Hands on: you can see and stop anything.

T12 Backup

#13not urgent

The nightly backup goes live as soon as there is a second Supabase project to write into. It is the only job here that can wait without hurting anything.

Hands on: a second copy of everything.

Your list and Max's list

Your list is what unlocks the agents; Max's list is essentially done.

You

  • The 5 judge points (#38)unlocks step 2, the judge build
  • Yes to A33.writer and A34.live (cents each)unlocks the fake cycle, T8
  • #8 the Umgesetzt screencloses what T5 built
  • #21 point B: may the chat promise a transformation?closes the rules file
  • #24: the reader questions under 95%closes the reader
  • #43: the 7 control-room questionsunlocks T9
  • #13: a 2nd free Supabase project (no card)unlocks the backup, T12

Max

  • Send the opener textsdone 20-11, and T11 is closed
  • Name the takeover tagdone: it is ai_off, and kiaus means friends
  • The 2 KI Flow fixesdone 20-11
  • Switch test with you, 3 passesdone 3 times · #31
  • Keep the switch test going: does the AI's answer fit what was said while it was off?alongside, not blocking
  • Run the first real cycle with youthe last step, #36

Every ticket, in one of three piles

22 open, 22 closed on GitHub. Most of the open ones are built and wait on a yes or a check, not on code.

Built: only needs checking or closing

#17Numbers screen: built, T7 checks it
#25Live reader: built, T7 checks it
#26Translation: built, T7 checks it
#8Umgesetzt screen: needs your yes
#21Rules file: needs point B
#24Reader under 95%: needs your answer

Agents build now

#39 #40 #42Lists (T4)
#38Judge build (T6), after your 5 points
#35Fake full cycle (T8)
#43Control room (T9), after 7 answers
#13Backup (T12), not urgent

Waits on you or Max

#385 judge points: you
#437 control-room questions: you
#132nd Supabase project: you
#37Reader cost fix: noted, not scheduled
#31Switch test, still running: both
#36First real cycle: both, last

Checking for logic errors without trusting the AI

Your question. Each check rests on code, the CRM or a person, not on a model's opinion.

1

Count it twiceSeparate code counts every number again. Last run: 1,068 numbers, 0 differences.

exists
2

Compare with the CRMA sample of our numbers checked against the CRM itself.

T7 finishes it
3

"Must be zero" rulese.g. nobody on the said-yes list has an appointment. If the count is ever above 0, the logic is broken.

going in with T4
4

Planted answersFake data where we know the right result. The judge ran 1,000 times on it.

exists
5

Your answer keyThe reader is scored against the 50 chats you graded by hand.

exists
6

Read 10 rows yourselfOnce a week per list. One read already showed only 1 in 6 on the said-yes list were real misses.

new · you or Max
7

Did they book?The final test: people on the lists booking, according to the CRM's own appointments.

new · you + Max
The rule from now on: every new list or number comes with one "must be zero" rule and one hand check. An AI review can be added on top, but it never replaces them.