fix: clarify interpretation boundary across viewing, provenance, and prompts

FPF-Spec.md

@@ -9165,6 +9165,15 @@ In both profiles:
 * `CorrespondenceModel` remains an **episteme‑level artefact**, not a new kernel‑type hidden inside A.6.3.
 * `U.EpistemicViewing` stays **view‑like**: it reveals what is already there under the correspondence; it does not perform Γ‑style constructions of new Intensions.
 
+#### A.6.3:4.4a - Conservative re-textualization boundary
+
+Translation, summary, filtering, and report belong to `U.EpistemicViewing` only when they remain **describedEntity-preserving** and **claim-conservative** in the sense of EV-3.
+
+Such outputs may later be rendered on publication surfaces, but their semantic status remains that of a conservative viewing step. If a publication adds conjectural explanatory commitments, introduces a new why-claim not already entailed by the source episteme, or turns the output into an open-question-bearing interpretive move, it is **not** an A.6.3 viewing. In that case:
+
+* faithful explanation/rendering belongs to publication/provenance discipline (`A.10` + `E.17`),
+* hypothesis-bearing interpretation belongs to `B.5.2.0` / `B.5.2`, not to `U.EpistemicViewing`.
+
 ### A.6.3:5 - Archetypal grounding (Tell–Show–Show)
 
 #### A.6.3:5.1 - Engineering system description → safety officer view (DirectEpistemicViewing)
@@ -16106,6 +16115,15 @@ A **typed, acyclic** graph disjoint from mereology. Node types: **SymbolCarrier*
 * `validatedBy` — links a claim to **empirical** evidence (tests, measurements, trials, observations).
   Both anchors terminate in the EPV‑DAG, not in the mereology graph.
 
+#### A.10:4.2a - Faithful rendering vs post-hoc narrative
+
+When a readable explanation or justification surface is generated from an existing trace, proof, model, or evidence path, the publisher **SHALL** declare whether the output is:
+
+* a **faithful rendering** of already-anchored content, or
+* a **post-hoc interpretive narrative**.
+
+Only the first may inherit anchors directly from the underlying EPV-DAG / SCR / RSCR chain. The second is a new episteme-level publication with its own evidence status. If the generated text introduces an open explanatory question or conjectural answer, it routes to `B.5.2.0` / `B.5.2` rather than masquerading as provenance itself.
+
 #### A.10:4.3 SCR / RSCR (Symbol Carrier Registers).
 Every `Γ_epist` aggregation **SHALL** emit an **SCR**: an exhaustive register of **symbol carriers** materially used in the aggregate, with id, type, version/date, checksum, source/conditions and optional `PortionOf` (A.14) for sub‑carriers.
 Every `Γ_epist^compile` **SHALL** emit an **RSCR**: SCR specialised to a **bounded context** (vocabularies, units) with publication‑grade identifiers and hashes.
@@ -27872,6 +27890,8 @@ A prompt is not yet a hypothesis. Prompt legality usually presupposes articulati
 #### B.5.2.0:4.3 - Boundary rule
 `U.AbductivePrompt` is an entry form, not an excuse to let arbitrary prose count as abductive input. Only declared prompt species may enter `B.5.2` through this form.
 
+It is also **not** a conservative view, summary, or faithful rendering of an existing claim set. If a publication merely re-textualizes or renders already-anchored commitments without introducing a genuine open question, keep it under `A.6.3` plus `A.10`/`E.17` discipline. Entry into `U.AbductivePrompt` begins only once the publication carries a real question-bearing interpretive move rather than a conservative rendering.
+
 ### B.5.2.0:5 - Archetypal Grounding
 **Tell.** An anomaly is one prompt species, not the only one.