Scope the comp's authority and bind builds to the recorded world.

The comp rules composition; the world rules material. Adds the authority
ranking to new-work.md, rewrite and token discipline to the craft floor,
token and motion evidence checks to the finish reviewer, derived text
inks to the questionnaire seed, and stale-seed disclosure at attended
stops. From the ekihana run's confirmed drift.

Co-authored-by: Cursor <cursoragent@cursor.com>

skill/agents/impeccable-finish-reviewer.md

@@ -27,9 +27,9 @@ Expect: the original request; the confirmed user answers; the artifact path(s);
 ## Checks, in order
 
 0. **Evidence.** Before any other check, verify the required captures exist and every capture is valid. Required: the platform's full viewport set (web: `desktop.png` and `mobile.png`; native: one capture per shipped device class), plus every capture the calling brief names as required, a reported user viewport (`user-<width>.png`) included. Valid: no black or blank regions, content matching what the filename claims (a visit capture showing the About section is invalid), the document top visible where the file claims a full page, dimensions that make sense for the named viewport. A required capture that is absent fails exactly like one that is malformed: a viewport nobody captured is a viewport nobody inspected, and it cannot ship. When any capture fails, the whole review changes shape: return `disposition: recapture` as the first line, then one section, `recapture`, listing each missing or invalid file and what a valid capture of it shows, and stop. Never build a matrix on malformed evidence; a verdict derived from a broken capture launders the breakage into an approval, and the parent owes you a full re-review on valid captures, not a scoring round.
-1. **Persistence.** PRODUCT.md exists. On a comp-led build, `.impeccable/build/state.json` exists and its `comps` (or `skipped` when a surface round locked the comp), `spec`, `plates`, and `hero` phases are `closed`; a comp-led config with no state file, or a state whose `comps` phase never closed, means the comp round was skipped and the build ran from a world description alone, a material finding that outranks craft; a phase closed with a `forced` record is disclosed as a material finding unless the user downgraded the comp in words the packet quotes; a state file whose `hero.gate.score` sits under 0.72, or a missing state file, means the reproduction ran unproven, a material finding, and `.impeccable/review/hero-repro.png` must exist either way. When DESIGN.md predates this build (an extension or redesign), it matches the built world; on a new world it is written after this review by the documenter, so its absence here is not a finding. When comp-round comps exist under `.impeccable/mocks/`, an approval record exists too: the surface brief naming the approved comp, or an `approved` flag in its sidecar. Comp-round comps with no recorded pick mean the approval point was skipped, a material finding. Files under `.impeccable/mocks/decision/` are exempt: they are the direction round's dealt hand, produced before any comp round, and imply no approval whatever the build path; a code-led build has no comp round at all.
+1. **Persistence.** PRODUCT.md exists. On a comp-led build, `.impeccable/build/state.json` exists and its `comps` (or `skipped` when a surface round locked the comp), `spec`, `plates`, and `hero` phases are `closed`; a comp-led config with no state file, or a state whose `comps` phase never closed, means the comp round was skipped and the build ran from a world description alone, a material finding that outranks craft; a phase closed with a `forced` record is disclosed as a material finding unless the user downgraded the comp in words the packet quotes; a state file whose `hero.gate.score` sits under 0.72, or a missing state file, means the reproduction ran unproven, a material finding, and `.impeccable/review/hero-repro.png` must exist either way. When a seed or prior DESIGN.md predates this build, it matches the built world, and "matches" is evidence you grep, not an impression: custom properties DESIGN.md defines that no rule consumes, literals sitting a step from a defined token, geometry a named rule bans (a 999px pill against a Slightly Soft rule). Each hit is a material finding under the craft floor's token rule, and an approved comp excuses none of them: the comp rules composition; the world rules material. On a new world with no seed, DESIGN.md is written after this review by the documenter, so its absence here is not a finding. When comp-round comps exist under `.impeccable/mocks/`, an approval record exists too: the surface brief naming the approved comp, or an `approved` flag in its sidecar. Comp-round comps with no recorded pick mean the approval point was skipped, a material finding. Files under `.impeccable/mocks/decision/` are exempt: they are the direction round's dealt hand, produced before any comp round, and imply no approval whatever the build path; a code-led build has no comp round at all.
 2. **Fidelity.** Start from the measurement, then judge what it cannot: read `.impeccable/review/diff/final/report.json` (and hero) first; every region scored `missing` or `contradicted` is a matrix row in that state unless the paired crop under `regions/` shows the score is wrong, and you say why; a region scored `match` still gets your eye for lettering character and material, which the numbers do not measure. Then, against your own element inventory of the approved comp, never against the contract's summary of it: topology, reading order, focal scale, overlaps and z-order, density, signature geometry, the primary action's treatment (a CTA the comp physically works, dissolves, or stamps is a signature element; its plain-rectangle rendition is contradicted), navigation items and icons, headline levels and scale relationships. Classify every salient element: match, acceptable adaptation, missing, contradicted, or added without approval. Three rows are mandatory in every matrix. TYPE: the display lettering's character, compression, width, weight, contrast, terminals, against the comp's; a face of a different character is contradicted however the layout matches. MATERIAL: an element rendered as flat CSS or clean vector where the comp shows painted, textured, dimensional, or photographic material is contradicted regardless of placement; medium is part of the promise. GROUND: the page field's value and temperature against the comp's, sampled from pixels on both sides when tooling allows rather than judged from memory, and read as the net on-screen result where a texture or tile paints over the base color; a ground warmer or cooler than the comp's is contradicted however faithfully the layout matches, and drift toward the rendition prior (warm cream on light grounds, blue-black slate on dark) is the direction to hunt. With no approved comp, TYPE and MATERIAL do not lapse: judge them against the contract's OWN-WORLD and the world's real materials, and treat faked physicality (CSS bevels, embossing, stamped-metal or chalk effects imitating a material the page never renders) as contradicted on its face; imitation material is the single most reliable mark of machine-made design. GROUND narrows rather than lapses: with no comp to sample, a color OWN-WORLD names is the target and the same warmer-or-cooler judgment applies; when OWN-WORLD names none, there is no GROUND authority, and the review says so in place of a verdict, because a target the reviewer invents turns the check into taste. A critique-reference comp on such a build is provocation, not spec: no element matrix, no adaptation citations, no asset obligations; its one contribution is what the image dared that the build did not, and dares worth adopting enter material_fixes as ordinary ordered fixes. An adaptation counts as intentional only when it cites the user answer, surface brief, accessibility need, or product truth that forced it; an uncited deviation is a defect. A missing signature element, a changed topology, or content added without approval fails fidelity and outranks every craft point in material_fixes. When MATERIAL is contradicted on the focal element, or contradiction is the page rather than the exception, stop ordering repairs: make the first material fix a rebuild directive naming the comp regions to re-derive and the assets to produce; a list of patches against a rejected page launders the rejection into an approval. A fix that requires producing an asset says so explicitly ("produce: <region> as a raster asset"), never phrased as a style adjustment the parent will answer with CSS. The comp is the spec for composition, topology, element inventory, density, lettering character, and material; it is not a pixel spec for semantics, accessibility, or responsive reflow, and that allowance covers translation, never replacement.
-3. **Ceiling.** Against the QUALITY BAR card: name the world's native devices the build left unused, frame, depth, lettering treatment, ornament density, motion. The card governs commitment and finish, never composition.
+3. **Ceiling.** Against the QUALITY BAR card: name the world's native devices the build left unused, frame, depth, lettering treatment, ornament density, motion. A committed motion energy shows up as eased state changes in the shipped code; a recorded energy with zero transitions is a finding. The card governs commitment and finish, never composition.
 4. **Contract, promise by promise.** First verify FORM carries the seed key the concept roll printed; a contract with no seed key, or one the parent cannot corroborate, means the roll was skipped, a material fix ahead of any craft point. Then, for each of the five blocks: does the render keep the promise? Apply the memory test to the first viewport.
 5. **Truth.** Demonstration data authored and labeled synthetic; no invented commercial claims; unanswered claims present as marked placeholders, not omissions. Every raster region of the spec shipped as its plate (the spec names the file; the page references it; the region's diff row is not `missing`), not a gradient, an inline SVG, or a many-vertex `clip-path` standing in for it, and every produced asset visibly present in the screenshots; an asset applied at near-zero opacity or buried behind a wash is a compliance token, not a shipped material, and the detector's `buried-raster` and `organic-clip-path` findings in the packet are material fixes.
 6. **Floor.** Read the craft floor's Refuse list and hold the screenshots against it: kickers and eyebrows, hard offset shadows outside a neobrutalist world, glyph icons, system display faces, gradient text, side stripes, and the rest. A banned element is a material fix even when it matches nothing in the comp: the builder loaded the same ban before writing it, and fidelity to a comp cannot authorize what the floor refuses. The parent's hook findings cover this mechanically where hooks run; this check exists because hookless harnesses reach you with none, and the last two live sessions shipped five kickers past a reviewer that never looked.

skill/reference/craft-floor.md

@@ -2,6 +2,8 @@
 
 Load this after the direction is settled, and build without announcing the checklist. A pinned brief or the committed visual world overrides anything here; your own habit does not. When the design hook is active it already enforces the mechanical checks below as you edit: act on its findings instead of re-auditing each rule. <!-- rule:skill-craft-floor -->
 
+A wholesale file rewrite, after a hook block or an error recovery, starts by re-opening DESIGN.md and the token sheet, so the new file restates the recorded system rather than your memory of it; memory is where a display token becomes a hand-tuned clamp. The ranking lives in [new-work.md](new-work.md): the comp rules composition; the world rules material, and a rewrite changes neither. <!-- rule:skill-rebuild-from-the-record -->
+
 ## Verify
 
 Each of these is a check on the built result, not an intention. Run them together in the batched inspection rounds, not as separate screenshot trips; the checks share one render.
@@ -11,6 +13,7 @@ Each of these is a check on the built result, not an intention. Run them togethe
 - **Spacing:** tight groups, generous separation, more space above a heading than below it. Read the computed values. <!-- rule:skill-layout-spacing-rhythm -->
 - **Type:** body measure 65–75ch, display max 6rem, tracking floor -0.04em, balanced headings, obvious scale and weight steps. Run the real copy at every breakpoint and fix what overflows. <!-- rule:skill-typo-floor --> <!-- rule:skill-ban-text-overflow -->
 - **Motion:** one authored moment, not scattered effects and not one identical entrance on every section. Exponential ease-out from an already-visible default. Reach past transform and opacity: blur, backdrop-filter, clip-path, mask, and shadow belong to the palette when they stay smooth. <!-- rule:skill-motion-floor --> <!-- rule:skill-motion-materials-palette --> <!-- rule:skill-motion-no-section-fade -->
+- **Tokens:** a size or color a step from a defined token takes the token: `0.82rem` beside a `0.8rem` token is the token, not a new size, and inline SVG or JS-drawn strokes take `var(--role)` or `currentColor`. Wire or remove what nothing consumes before finish; done means every defined custom property is consumed by at least one rule. <!-- rule:skill-tokens-are-consumed -->
 - **States:** hover, disabled, loading, error, empty. Plus real content, working controls, responsive composition, keyboard focus. <!-- rule:skill-floor-shipping -->
 - **Browser surfaces:** the parts you did not draw still carry the design. Text selection, the caret, custom scrollbars, focus rings, underline offset, and the numerals in tabular data all ship with browser defaults that belong to no design system. Theme them from the palette. This is the cheapest signal that a page was built rather than assembled, and the one models skip most reliably. <!-- rule:skill-craft-browser-surfaces -->
 - **Copy:** the product's own language. Controls name their action; errors name the problem and the recovery. <!-- rule:skill-copy-design-material -->
@@ -37,7 +40,7 @@ Surface habits:
 - Sparklines, progress rings, and soft-shadowed rounded rectangles standing in for content. <!-- rule:skill-reflex-decorative-chrome -->
 - Monospace as a costume for "technical" rather than for code, data, or measurement. <!-- rule:skill-reflex-mono-as-technical -->
 - A system display face (Impact, Arial Black, the platform sans) as the display voice of an own-world page. Source and self-host a face whose character matches the approved lettering; the closest installed font is a failure, not a fallback. <!-- rule:skill-ban-system-display-face -->
-- Unicode glyphs or emoji standing in for an icon system. Icons are drawn, from a real library or authored SVG, in one consistent stroke and weight. <!-- rule:skill-ban-glyph-icons -->
+- Unicode glyphs or emoji standing in for an icon system. Icons are drawn, from a real library or authored SVG, in one consistent stroke and weight. When a pack icon fails to fetch, take the nearest icon from the same pack, jsDelivr as the fallback CDN; drawing a replacement from scratch is an exception the user signs off on, so the one-pack promise survives a failed fetch. <!-- rule:skill-ban-glyph-icons -->
 - Geometric masks standing in for organic contours. A circle, polygon, or radial-gradient cutout approximating a photographic subject's edge is the cheap version of the effect and reads worse than omitting it. Derive an alpha matte from the actual image, or produce a cut-out asset. <!-- rule:skill-ban-geometric-occlusion-mask -->
 - Light or dark picked by category. Pick it from the use scene: who, where, under what ambient light. <!-- rule:skill-reflex-theme-by-habit -->
 

skill/reference/document.md

@@ -451,7 +451,7 @@ This seed writes a minimal frontmatter with `name` and `description` only; no co
 
 **Questionnaire seed** (`.impeccable/design-interview/answers.json` exists from this run). The user answered every screen by eye, so the seed carries their answers as decisions, not directions. Read the answers file plus the picked cue's palette entry in `.impeccable/visual-cues/cues.json` (`palette-source` names it), and map:
 
-- **Frontmatter**: `name` and `description`, plus real `colors` (the four `palette-*` hex values under descriptive slugs; these are picked, not sampled) and real `typography` (`font-heading` and `font-body` are exact family names; give each role its family and weight intent, leave sizes for implementation). Still no `rounded`, `spacing`, or `components`: the corner and spacing answers are qualitative, and nothing is built.
+- **Frontmatter**: `name` and `description`, plus real `colors` (the four `palette-*` hex values under descriptive slugs; these are picked, not sampled) and real `typography` (`font-heading` and `font-body` are exact family names; give each role its family and weight intent, leave sizes for implementation). Derive the two text inks and record them under `colors` too: one near-black and one near-white, the pair the picker's previews already set their text in over these exact surfaces, each holding 4.5:1 against the grounds it will carry copy on, so a builder needing body-text contrast finds ink in the system instead of inventing a fifth color. Still no `rounded`, `spacing`, or `components`: the corner and spacing answers are qualitative, and nothing is built.
 - **Overview**: Creative North Star and philosophy phrased from the questionnaire's color-strategy and motion answers plus the chat references; reference the user's anti-reference directly. Name the chosen surfaces (`surface-modes`) and what each is for. Movement stays here, after the North Star, but the questionnaire asks it of a landing page and a portfolio only, so write what the keys support:
   - `motion-energy-<mode>` keys present, all agreeing: one philosophy sentence for the product, as before.
   - Keys present and disagreeing: one sentence per surface, named (*"The landing page moves on state change only; the portfolio stages entrances and drives sequences on scroll."*). The bare `motion-energy` is the leading one of the two.

skill/reference/new-work.md

@@ -40,7 +40,7 @@ The script deals three of your structures; the dice pick which three reach the u
 
 ### Create or replace the visual world
 
-**When DESIGN.md is missing**, pause before any of the work below and {{ask_instruction}} once: recommend `{{command_prefix}}impeccable document --seed`, the guided interview plus browser questionnaire, because a world established from the user's own choices beats one assigned to them; offer the skip in the same breath. *"There's no design system on record yet. Before I invent directions for [the requested surface], I can run a short interview and browser questionnaire so the visual world is built from your choices; I recommend it. Or skip it and I'll roll a direction now."* The pause is enforced by the script, not by your discipline: with no DESIGN.md on record, the step 4 roll refuses to deal and prints this same offer until the questionnaire has been put to the user. Do not roll first and offer the questionnaire after, the assignment anchors the conversation; do not run document on the user's behalf without their yes. **Accepted:** seed mode in [document.md](document.md) owns the interview and the picker; follow it. When the seed DESIGN.md is written, resume at section 1: the seed world now reads as an established world, so inherit it; the direction roll below no longer applies, and the optional surface roll remains. **Skipped:** re-run the step 4 command with `--seed-declined="<the user's verbatim skip answer>"`; the flag carries the user's own words as evidence of the skip, and the roll deals with nothing in step 4 softened. The build request itself is never a skip answer, a bare flag refuses again, and fabricating or paraphrasing the quote is a contract violation; only words the user typed after being asked qualify. When DESIGN.md exists, there is no pause and no flag.
+**When DESIGN.md is missing**, pause before any of the work below and {{ask_instruction}} once: recommend `{{command_prefix}}impeccable document --seed`, the guided interview plus browser questionnaire, because a world established from the user's own choices beats one assigned to them; offer the skip in the same breath. *"There's no design system on record yet. Before I invent directions for [the requested surface], I can run a short interview and browser questionnaire so the visual world is built from your choices; I recommend it. Or skip it and I'll roll a direction now."* The pause is enforced by the script, not by your discipline: with no DESIGN.md on record, the step 4 roll refuses to deal and prints this same offer until the questionnaire has been put to the user. Do not roll first and offer the questionnaire after, the assignment anchors the conversation; do not run document on the user's behalf without their yes. **Accepted:** seed mode in [document.md](document.md) owns the interview and the picker; follow it. When the seed DESIGN.md is written, resume at section 1: the seed world now reads as an established world, so inherit it; the direction roll below no longer applies, and the optional surface roll remains. Through every phase that follows, comp-led included, the comp rules composition; the world rules material, and this seed is that world. **Skipped:** re-run the step 4 command with `--seed-declined="<the user's verbatim skip answer>"`; the flag carries the user's own words as evidence of the skip, and the roll deals with nothing in step 4 softened. The build request itself is never a skip answer, a bare flag refuses again, and fabricating or paraphrasing the quote is a contract violation; only words the user typed after being asked qualify. When DESIGN.md exists, there is no pause and no flag.
 
 1. Name the product's unique mechanism in one sentence, the audience's real scene, its cultural home, and what this first surface must prove. Note the page this category always ships and its predictable opposite; name both as the rut and keep them out of the seven-candidate list. A brief that paints its own picture, a product name, a titled artifact, a governing metaphor, adds its literal reading to the rut: spend at most one candidate on it and derive the rest from elsewhere in the audience's world.
 2. From that cultural world, list seven concrete visual systems, artifacts, places, or rituals the audience knows by heart, each with one line on why it resonates and can carry the mechanism, ordered by resonance. The audience's world includes its graphic and screen traditions, not only its physical objects: the notation, publications, identity programs, data graphics, and interfaces it reads daily; a nameable abstract system (a school of poster, a documentation standard) is as concrete a candidate as any artifact. What would this thing look like as a physical object; what did its world look like before the web? Near-duplicates count once. When more than three of the seven share one material family, the derivation stopped at the subject's most obvious artifact; dig until the list spans at least three families.
@@ -98,6 +98,8 @@ For `shape`, return the selected direction to [shape.md](shape.md) and stop befo
 
 Build the assigned direction, not a safer interpretation of it. The form supplies structure, reading order, component conventions, and native motion; the product supplies every fact. Commit every atom: nav, buttons, inputs, and links are rebuilt in the form's vocabulary, and a stock component inside a committed form is a lapse. Land the first build fully committed; the passes that follow exist to make the committed thing clear and effective, never to dilute it. In unattended work, the safe rendition is the known risk. <!-- rule:skill-commit-every-atom -->
 
+**The comp rules composition; the world rules material.** Composition is the comp's dominion: topology, element inventory, density, region assignment. The recorded world's system stays binding through every phase: palette roles at their recorded scale of use, type tiers, corner language, motion register. Product truth in **PRODUCT.md** outranks both, so a comp never deletes a core product answer. A genuine conflict between the comp and a DESIGN.md named rule is resolved consciously and rides to the documenter's re-record at finish: adopt the deviation into the world's record, or conform the build. <!-- rule:skill-comp-rules-composition -->
+
 ### Comp-led: the comp is a measured contract
 
 When an approved comp exists, it is a spatial contract, not a mood board: only the user can downgrade its authority, in explicit words. Models systematically believe their HTML, CSS, and SVG recreation of an image succeeded when it did not, so the build runs as a state machine on disk whose gates measure the screen against the comp instead of asking you to remember it. Start it once, and let it tell you what is next:
@@ -142,7 +144,7 @@ After the second inspection round the build thread's polishing is over: no furth
 
 Then spawn the shipped finish reviewer, `impeccable-finish-reviewer` (`impeccable_finish_reviewer` in codex; `/impeccable-finish-reviewer` in Cursor; on GitHub Copilot say "Use the impeccable-finish-reviewer agent"), with the original request, confirmed answers, the artifact path, the screenshot paths, the direction contract, existing hook findings, the QUALITY BAR card and approved comp paths (a code-led build has no approved comp; the chosen decision comp rides in that slot as the critique reference, named as such), on a comp-led build the build state (`.impeccable/build/state.json`), the spec, and the diff directories (`.impeccable/review/diff/hero/` and `.impeccable/review/diff/final/`, whose side-by-side, heatmap, region pairs, and `report.json` are the fidelity evidence), the craft-floor reference path, and on a native platform the platform reference path(s), [ios.md](ios.md) / [android.md](android.md), both on adaptive, plus one line saying no detector ran, so the reviewer judges in the platform's conventions rather than the web's. The reviewer has no browser; screenshots you fail to pass are checks it cannot run. Never read the shipped agents' definition files before spawning; the harness loads them at spawn, and you owe only the input packet. Wait on any agent with one long timeout rather than a loop of short polls, and spend the wait on the next independent step. Verify the return carries the five contract sections (a recapture return carries one, its recapture list); on an empty or thrashed return, respawn once with the same inputs. This review never runs inside the build thread and never inherits it: spawn the reviewer fresh, with no forked conversation history (`fork_turns: 0` in codex); a reviewer that inherits your transcript inherits your framing, your optimism, and your abstractions, and everything it needs travels in the inputs above. Only a harness with no subagent capability at all substitutes a fresh in-thread pass after stepping fully out of the build context, run from [degraded/finish-reviewer.md](degraded/finish-reviewer.md), and a substituted or failed-and-replaced review is disclosed in one line at finish, never silently. <!-- rule:skill-finish-separate-reviewer -->
 
-Act on the disposition word; there are exactly four. **recapture**: the evidence failed, not the build. Recapture what the return names under the capture-validity rules, then run a full review over the new evidence. A review conducted on invalid evidence binds nothing, and a verdict pass may never follow it. **rebuild**: fidelity failed wholesale, not in patches. Skip the fix batch and execute the rebuild immediately: re-derive the named regions, produce the named assets, and send the result back for a fresh full review, never a verdict pass; a rebuild replaces regions wholesale, so the whole matrix runs again over the recaptures. Tell the user what is happening rather than asking permission to fix a failure. Consult the user only on a second rebuild directive, both verdicts on the table, or when rebuilding would discard content the user approved. **ship**: nothing is owed; report the verdict at its scope and continue to the documenter. **fix**: apply the material fixes in one batch, rebuild once, and recapture the same viewports over the same files. A recapture measures positions, loading, and overflow; it cannot measure whether a fix reached the quality the finding named, so send the recaptured screenshots back to the same reviewer for a verdict scoring every material fix resolved, partial, or unresolved (through the harness's agent continuation; without one, run the scoring fresh from [degraded/finish-reviewer.md](degraded/finish-reviewer.md)'s Verdict Pass). Fixes scored partial or unresolved get another batch, recapture, and verdict. Two rounds is the budget an unattended run ends at; an attended session's ceiling belongs to the user, so when the second verdict still lists open items, put the table in front of them and let them choose between shipping as it stands and funding another round. Whoever decides, stop the moment a round resolves nothing, and the reviewer's findings are the only list you work from, never your own re-opened hunt. Do not run a second detector. <!-- rule:skill-verdict-bounds-the-finish -->
+Act on the disposition word; there are exactly four. **recapture**: the evidence failed, not the build. Recapture what the return names under the capture-validity rules, then run a full review over the new evidence. A review conducted on invalid evidence binds nothing, and a verdict pass may never follow it. **rebuild**: fidelity failed wholesale, not in patches. Skip the fix batch and execute the rebuild immediately: re-derive the named regions, produce the named assets, and send the result back for a fresh full review, never a verdict pass; a rebuild replaces regions wholesale, so the whole matrix runs again over the recaptures. Tell the user what is happening rather than asking permission to fix a failure. Consult the user only on a second rebuild directive, both verdicts on the table, or when rebuilding would discard content the user approved. **ship**: nothing is owed; report the verdict at its scope and continue to the documenter. **fix**: apply the material fixes in one batch, rebuild once, and recapture the same viewports over the same files. A recapture measures positions, loading, and overflow; it cannot measure whether a fix reached the quality the finding named, so send the recaptured screenshots back to the same reviewer for a verdict scoring every material fix resolved, partial, or unresolved (through the harness's agent continuation; without one, run the scoring fresh from [degraded/finish-reviewer.md](degraded/finish-reviewer.md)'s Verdict Pass). Fixes scored partial or unresolved get another batch, recapture, and verdict. Two rounds is the budget an unattended run ends at; an attended session's ceiling belongs to the user, so when the second verdict still lists open items, put the table in front of them and let them choose between shipping as it stands and funding another round; the table states in one line that the documenter has not run and, when DESIGN.md is still the pre-build seed, names that too, so the stop reads as what it is, a paused run with an unrecorded world. Whoever decides, stop the moment a round resolves nothing, and the reviewer's findings are the only list you work from, never your own re-opened hunt. Do not run a second detector. <!-- rule:skill-verdict-bounds-the-finish -->
 
 A rebuild and a fix round share one asset rule: a raster either round creates or replaces is still asset work under [visualize.md](visualize.md)'s Produce section and keeps its **provenance** like every build raster, and a raster the round abandons is deleted in the same batch. Before either round's result goes back for review or verdict, run `node {{scripts_path}}/embed-prompt.mjs --scan <asset-dir...>` over the directories the artifact's rasters ship from and clear every file it reports by embedding what it is missing: the exact generation prompt for a produced raster, the origin for a sourced, stock, or pre-existing one. The scan only reads; deletion is reserved for rasters the round abandoned, never for a file the scan flagged. <!-- rule:skill-late-raster-provenance -->