#G.9:4.3a — Worked parity slice

Preface node heading:g-9-4-3a-worked-parity-slice:92237

#What this page is

This is generated FPF reference text from the specification preface or supporting sections. It helps interpret FPF; it is not FPF Reference product documentation.

#Methodology

Use it to understand how the specification wants to be read, then return to a route, pattern, or work packet for active work. Cite generated IDs only when the wording changes the task decision.

#Content

Loop-engineering parity compares loop or harness variants only after the parity object is declared. An evaluation program, benchmark script, or dashboard is not the characteristic being improved; it is part of the evaluation or comparison procedure. The parity plan must pin the object family, baseline set, freshness window, comparator editions, budget when relevant, and outcome shape before a result can be consumed by E.23, C.19, or G.5.

• Two agentic search setups both claim bounded specialization on the same declared task family.
• The ParityPlan pins the same freshness window, threshold target, adaptation budget, prior-exposure declaration, comparator editions, and corridor-entry baseline. One setup reaches threshold sooner but shows low retention and no transfer. The other reaches threshold later, but carries reusable transfer and lower downside field.
• A CSLC-admissible ParityReport@Context therefore states what was held constant, which signals remained telemetry, and why the outcome stays a governed selected set or partial order rather than collapsing into a scalar winner. The reader can recover the practical comparison from the parity slice itself before reading any optional wiring blocks.

---

Last Updated: 2026-07-03 — upstream FPF commit f7c7e93f (github.com/ailev/FPF)