A status-report iteration that converged on "lead with the decision-relevant number" becomes a prompt that produces that report every week without the Monday rewrite.
Recurring deliverables are where iteration pays the highest dividend — and loses the most when it stays in the chat. A leadership report iterated to its converged form (one-number opening, three sections, one page, never "various changes") is a weekly task solved permanently — IF the convergence escapes the conversation. This setup loads that iteration: the generic-opening rejection, the one-number correction, the three-section format, the one-page constraint, and the "this is the final version" declaration, distilled into the prompt the next Monday starts from.
Converge once
The iteration happens a single time — rejections and corrections doing the design.
Distill the convention
Three sections, one page, one-number opening — conventions become prompt rules.
Run it weekly
Each Monday starts at the converged format; the iteration never repeats.
The rejected draft rides along in the AVOID block as 'Too generic — leadership stops reading at "various changes"', so the model is told explicitly not to reproduce that opening. Paired with the 'Keep the one-number opening' requirement, the prompt fences off the path you already walked back instead of leaving the model to rediscover it each Monday.
That block reads '[No format decisions detected — specify the output shape if it matters]' because the iteration never nailed down an explicit output shape, and PRESERVED REQUIREMENTS confirms 'Format decisions: 0'. If layout matters, type the shape into that line yourself before running; the one-page CONSTRAINT still holds, but section order and headings are left open until you fill it in.
It packages the converged iteration into the reusable prompt you see, ending with 'Distillation confidence: High' — it does not run the report. You paste that prompt into your own assistant each Monday with fresh metrics; the builder only bottles the requirements the iteration earned, and you still own reviewing the generated report before it ships.
What a converged conversation contains: a rejected voice, an accepted framing, a hard constraint, a format decision, and a final-version moment — each mapped to its prompt section.
The result you finally liked lives inside the chat. Distill the iterations into one prompt: corrections become requirements, the accepted version becomes the quality bar.
In feedback threads the rejections carry the knowledge: the mocking empty-state enthusiasm, the passive phrasing — converted into the avoid-rules that define the style.
End a working session like a shift change, not an abandonment: state captured, decisions logged, next step named — ready for the next session to pick up.
Turns a reusable API request prompt into a clean variable set covering endpoint, method, auth, payload, response format, and error handling.
A reusable AI agent task template with variables for objective, context, available tools, constraints, success criteria, failure handling, and output format.
Distill an iterated conversation into one reusable prompt — corrections become requirements, rejections become an avoid-list.