Continuation criteria, read in Leg B to decide when Model B moves on, either miss one of their Leg A turn's asks or demand what that turn cannot produce, such as a later turn's file.
Do this instead
Tick every ask in each Leg A prompt against a clause, and move any clause that needs a later file or ask to the turn that brings it.
Where it happened
2 real tasks. Each shows the spot and what it should have been; click any line to open the task there.
the words at faultthe row or field at faultmissing from the taskwhat is righta task file
Example 10Small Business & Personal Docs · 4 Leg A turns
How it breaks the standardAt Leg A turn 2 the only file handed over is the calendar, and the user sets the travel at about three days a week himself. No correct response to Leg A turn 2 can meet the clause, so in Leg B it would keep you hinting Model B toward a directive not yet handed over.
Each card states whether its piece still carries a correctiona seventh clause, beside the six it has
How it breaks the standardIn Leg B you move Model B past milestone 1 once its six clauses hold, and a response that never checks corrections meets every one. Here the right answer is none of the three, so a response that skips the check, or pins another piece's Slack correction on one, moves on unchecked. This costs points.
Continuation criteria are present and checkable but under-specified — they would let the simulator advance while a requirement of that turn is still unmet.
In Leg B, the simulator is you.
Why it matters
In Leg B you are the simulator, and continuation criteria are what you advance on: once Model B's response meets them you move to the next milestone, and until it does you send another hint. A missing clause lets Leg B move on with a requirement unmet, and every later turn builds on the hole. A clause the turn cannot satisfy does the opposite: it keeps you hinting Model B toward something the user has not handed over yet.
Example 10Stopping rules that do not gate the turnCosts points
Small Business & Personal Docs4 Leg A turnstask 6aa34dd1625bb5b220c6b1cb
From the taskMilestones 8, 9
8
The user has sent the travel rules his employer issued and wants the schedule rebuilt on the window they impose instead of whatever pattern was assumed until now.
Leg A turn2modifierdeferred_assetassets delivered.jpg
the travel days in the model reflect only the document the user attached, with no residual figures from a prior assumed pattern
the week by week view moves with it
both counts move with it
9
The user has sent the travel rules his employer issued and wants the schedule rebuilt on the window they impose instead of whatever pattern was assumed until now.
Leg A turn3modifierrevisionassets delivered.jpg
the travel days in the model come from the document the user attached
the earlier assumed pattern is gone rather than offered alongside it
the week by week view moves with it
both counts move with it
From the taskLeg A prompt, turn 2, to Model A
Leg A turn 2
I dug up Jaylen's school calendar for this year and marked it up (see attached IMG_0814_calendar.jpg). My carline mornings are highlighted on it and the custody handovers with Terrance are circled, with the Poplarville details in the margin. Since the regional job runs 60% to 70% travel, about three days a week across the five sites, update dashboard.html with a schedule conflict calendar and a week-by-week matrix across the school year. Map the travel load against what is marked on that calendar, and tell me exactly how many school mornings and how many handovers end up with a travel collision.
From the taskLeg A prompt, turn 3, to Model A
Leg A turn 3
Corwin sent over the Southeast region's Q3 travel directives (see attached IMG_0819_travel_policy.jpg). Read it and apply what it actually requires, because it is stricter than what you have modelled and it changes the math. On that schedule we would be out about $200 a month for before-school care, $150 a month for the Poplarville handovers, and $180 a month on meals the company will not cover. Update dashboard.html so those costs come off the raise in the cash comparison, and show me the true adjusted monthly delta and what it does to our savings.
Task fileIMG_0819_travel_policy.jpg
IMG_0819_travel_policy.jpginput file
Click the image to see it at full size, and again to fit it back.
Task fileIMG_0814_calendar.jpg
IMG_0814_calendar.jpginput file
Click the image to see it at full size, and again to fit it back.
Example 11Stopping rules that do not gate the turnCosts points
Visual Learning4 Leg A turnstask 6aa02f48755f41e02b01483f
From the taskMilestone 1
1
`Build the years-in-review from the drop
Leg A turn1modifieropeningassets deliverednone
- Make the named year-in-review page from the month-stamped exports in this drop and the desk notes, with type/subject read off each opened graphic, read vs confirmed marked, and every count opening the files behind it. - The named HTML file exists. - Each month-stamped export in this drop has a card. - Chart type and subject on those cards come from the opened graphic. - Each headline count opens the set of files that make it up. - Each card is marked read or confirmed. **Follow-up prompt:** `(opening prompt, verbatim, do not paraphrase)
No clause checks which pieces still carry a correction.
From the taskLeg A prompt, turn 1, to Model A
Leg A turn 1
Nadine wants a year-in-review of the dailies for the desk... not a recap from my head. The exports are in this drop, named by month(2026-02...and so on); desk_review_notes.md is how I usually sort what actually ran. Build years_in_review.html from those files: what ran, what it was about, which chart types we reached for, and which pieces still carry a correction. Mark each one read or confirmed. I need the counts on the page to click through to the files behind them.