Skip to content

The design language that makes your AI harness better at design.

JavaScript

Sync generated provider output

github-actions[bot]875a38935bd0
diff --git a/.agent/skills/impeccable/SKILL.md b/.agent/skills/impeccable/SKILL.md
index 34cee8b0e..59c839aab 100644
--- a/.agent/skills/impeccable/SKILL.md
+++ b/.agent/skills/impeccable/SKILL.md
@@ -13,7 +13,7 @@ This skill gives you the tools and permission to make design that earns the name
Core principles:
- Finish the job. No hedging, no shortcuts: the deliverable is complete, except assets the user must provide.
- Scale ambition to the surface's mode (see Modes). Persuade and Experience get full commitment: bold, distinct work with a point of view. On Operate and Read, ambition shows as precision and restraint, and patterns the visitor already knows earn their place.
-- Verify in bounded passes, not a loop, and the ceiling covers the whole cycle: screenshots, defect scans, micro-edits, and rebuilds alike. Build fully, inspect once with a batched round (desktop and mobile together on the web; the shipped device classes on a native platform), fix everything it shows in one batch, confirm with at most one more round, and stop polishing. Open-ended self-QA burns the user's money doing worse what the finish handoffs do better.
+- Verify by looking at the rendered result at the sizes it ships at, and batch the checks: fix what one look shows before looking again, rather than screenshotting after every tweak.
## Setup
diff --git a/.agent/skills/impeccable/reference/craft-floor.md b/.agent/skills/impeccable/reference/craft-floor.md
index 9653b9808..64a195eba 100644
--- a/.agent/skills/impeccable/reference/craft-floor.md
+++ b/.agent/skills/impeccable/reference/craft-floor.md
@@ -4,7 +4,7 @@ Load this after the direction is settled, and build without announcing the check
## 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.
+Each of these is a check on the built result, not an intention. Run them together in each batched look at the render, not as separate screenshot trips; the checks share one render.
- **Contrast:** body and placeholder text ≥4.5:1, large text ≥3:1. On colored surfaces tint secondary text from that hue or the foreground; never gray.
- **Depth:** shadows carry an offset and a soft blur. A zero-offset colored halo is decoration.
diff --git a/.agent/skills/impeccable/reference/new-work.md b/.agent/skills/impeccable/reference/new-work.md
index 2bf012c4e..c5287c65b 100644
--- a/.agent/skills/impeccable/reference/new-work.md
+++ b/.agent/skills/impeccable/reference/new-work.md
@@ -133,7 +133,7 @@ Preserve semantics, accessibility, performance, responsiveness, project conventi
## 7. Inspect and finish
-Inspect the surface's target sizes in one batched screenshot round: desktop and mobile on the web; on a native platform (`ios` / `android` / `adaptive`), the shipped device classes per OS, captured from the simulator or emulator the way the platform reference's Verifying the build section describes. When the harness reports the user's actual viewport (an in-app browser's size, a named resolution), add that width to the set: the width that breaks is the one the user sees first. Critique the render against the user's request and the direction contract, fix material gaps, and confirm with one final round; two rounds is the ceiling, and fixes batch between them rather than earning per-tweak screenshots. The ceiling bounds inspection and never closes an open build-phase gate: a gate stays open until it passes or the route it names is taken. On a comp-led build, run `.agent/skills/impeccable/scripts/impeccable comp-diff --comp <approved comp> --build .impeccable/review/desktop.png --spec .impeccable/build/spec.json --out-dir .impeccable/review/diff/final` and read its region rows and paired crops as the critique: the side-by-side is the view the build thread never has on its own, and a region it scores missing or contradicted is a fix whatever the page looks like from memory. Never judge fidelity from one full-page thumbnail; it hides exactly the failures that matter. On a Persuade surface, verify the mode did its job: a first-time visitor should know what this is, why it matters, and what to do within seconds, in the form's own vocabulary.
+Inspect the surface's target sizes in one batched screenshot round: desktop and mobile on the web; on a native platform (`ios` / `android` / `adaptive`), the shipped device classes per OS, captured from the simulator or emulator the way the platform reference's Verifying the build section describes. When the harness reports the user's actual viewport (an in-app browser's size, a named resolution), add that width to the set: the width that breaks is the one the user sees first. On anything interactive, the round drives every interaction and reachable state (open, closed, hover, focus, error, empty, overlong content) and checks each for overlap and clipping, not only the resting screen. Critique the render against the user's request and the direction contract, fix material gaps, and confirm with one final round; two rounds is the ceiling, and fixes batch between them rather than earning per-tweak screenshots. The ceiling bounds inspection and never closes an open build-phase gate: a gate stays open until it passes or the route it names is taken. On a comp-led build, run `.agent/skills/impeccable/scripts/impeccable comp-diff --comp <approved comp> --build .impeccable/review/desktop.png --spec .impeccable/build/spec.json --out-dir .impeccable/review/diff/final` and read its region rows and paired crops as the critique: the side-by-side is the view the build thread never has on its own, and a region it scores missing or contradicted is a fix whatever the page looks like from memory. Never judge fidelity from one full-page thumbnail; it hides exactly the failures that matter. On a Persuade surface, verify the mode did its job: a first-time visitor should know what this is, why it matters, and what to do within seconds, in the form's own vocabulary.
A capture is evidence only when it is valid, and you validate before you send. Settle or disable entrance motion first: an element hidden by animation timing reads as a missing element and gets fixed into a regression. Capture full-page shots from the document top. Capture the comp comparison at the comp's own pixel dimensions. Then open every file once and confirm it shows what its name claims: no black or blank regions, no wrong section behind a right filename, no half-loaded state. A malformed capture sent onward costs the whole round; the reviewer answers it with `disposition: recapture` and nothing it reviewed binds.
diff --git a/.agent/skills/impeccable/reference/overdrive.md b/.agent/skills/impeccable/reference/overdrive.md
index 89be08b5c..b3dfb708c 100644
--- a/.agent/skills/impeccable/reference/overdrive.md
+++ b/.agent/skills/impeccable/reference/overdrive.md
@@ -21,7 +21,7 @@ Skipping this step risks building something embarrassing that needs to be thrown
### Verify in the Browser
-Ambitious effects rarely work on the first try, so verify them in the browser rather than from code: an effect that technically runs can still stutter, ease wrong, or land on the wrong frame. SKILL.md's bounded passes apply: one batched inspection, one fix batch, at most one confirming round. Point them at the effect's timing, easing, and smoothness, then stop and name what still falls short instead of opening another round.
+Ambitious effects rarely work on the first try, so verify them in the browser rather than from code: an effect that technically runs can still stutter, ease wrong, or land on the wrong frame. Aim the checks at timing, easing, and smoothness, batch the fixes between looks, and keep refining until the effect is extraordinary.
---
diff --git a/.agents/skills/impeccable/SKILL.md b/.agents/skills/impeccable/SKILL.md
index 3903ff726..4a0e0239f 100644
--- a/.agents/skills/impeccable/SKILL.md
+++ b/.agents/skills/impeccable/SKILL.md
@@ -10,7 +10,7 @@ This skill gives you the tools and permission to make design that earns the name
Core principles:
- Finish the job. No hedging, no shortcuts: the deliverable is complete, except assets the user must provide.
- Scale ambition to the surface's mode (see Modes). Persuade and Experience get full commitment: bold, distinct work with a point of view. On Operate and Read, ambition shows as precision and restraint, and patterns the visitor already knows earn their place.
-- Verify in bounded passes, not a loop, and the ceiling covers the whole cycle: screenshots, defect scans, micro-edits, and rebuilds alike. Build fully, inspect once with a batched round (desktop and mobile together on the web; the shipped device classes on a native platform), fix everything it shows in one batch, confirm with at most one more round, and stop polishing. Open-ended self-QA burns the user's money doing worse what the finish handoffs do better.
+- Verify by looking at the rendered result at the sizes it ships at, and batch the checks: fix what one look shows before looking again, rather than screenshotting after every tweak.
## Setup
diff --git a/.agents/skills/impeccable/reference/craft-floor.md b/.agents/skills/impeccable/reference/craft-floor.md
index ae213ce80..43da17bca 100644
--- a/.agents/skills/impeccable/reference/craft-floor.md
+++ b/.agents/skills/impeccable/reference/craft-floor.md
@@ -4,7 +4,7 @@ Load this after the direction is settled, and build without announcing the check
## 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.
+Each of these is a check on the built result, not an intention. Run them together in each batched look at the render, not as separate screenshot trips; the checks share one render.
- **Contrast:** body and placeholder text ≥4.5:1, large text ≥3:1. On colored surfaces tint secondary text from that hue or the foreground; never gray.
- **Depth:** shadows carry an offset and a soft blur. A zero-offset colored halo is decoration.
diff --git a/.agents/skills/impeccable/reference/new-work.md b/.agents/skills/impeccable/reference/new-work.md
index f0bf69d09..6720fa3b1 100644
--- a/.agents/skills/impeccable/reference/new-work.md
+++ b/.agents/skills/impeccable/reference/new-work.md
@@ -133,7 +133,7 @@ Preserve semantics, accessibility, performance, responsiveness, project conventi
## 7. Inspect and finish
-Inspect the surface's target sizes in one batched screenshot round: desktop and mobile on the web; on a native platform (`ios` / `android` / `adaptive`), the shipped device classes per OS, captured from the simulator or emulator the way the platform reference's Verifying the build section describes. When the harness reports the user's actual viewport (an in-app browser's size, a named resolution), add that width to the set: the width that breaks is the one the user sees first. Critique the render against the user's request and the direction contract, fix material gaps, and confirm with one final round; two rounds is the ceiling, and fixes batch between them rather than earning per-tweak screenshots. The ceiling bounds inspection and never closes an open build-phase gate: a gate stays open until it passes or the route it names is taken. On a comp-led build, run `.agents/skills/impeccable/scripts/impeccable comp-diff --comp <approved comp> --build .impeccable/review/desktop.png --spec .impeccable/build/spec.json --out-dir .impeccable/review/diff/final` and read its region rows and paired crops as the critique: the side-by-side is the view the build thread never has on its own, and a region it scores missing or contradicted is a fix whatever the page looks like from memory. Never judge fidelity from one full-page thumbnail; it hides exactly the failures that matter. On a Persuade surface, verify the mode did its job: a first-time visitor should know what this is, why it matters, and what to do within seconds, in the form's own vocabulary.
+Inspect the surface's target sizes in one batched screenshot round: desktop and mobile on the web; on a native platform (`ios` / `android` / `adaptive`), the shipped device classes per OS, captured from the simulator or emulator the way the platform reference's Verifying the build section describes. When the harness reports the user's actual viewport (an in-app browser's size, a named resolution), add that width to the set: the width that breaks is the one the user sees first. On anything interactive, the round drives every interaction and reachable state (open, closed, hover, focus, error, empty, overlong content) and checks each for overlap and clipping, not only the resting screen. Critique the render against the user's request and the direction contract, fix material gaps, and confirm with one final round; two rounds is the ceiling, and fixes batch between them rather than earning per-tweak screenshots. The ceiling bounds inspection and never closes an open build-phase gate: a gate stays open until it passes or the route it names is taken. On a comp-led build, run `.agents/skills/impeccable/scripts/impeccable comp-diff --comp <approved comp> --build .impeccable/review/desktop.png --spec .impeccable/build/spec.json --out-dir .impeccable/review/diff/final` and read its region rows and paired crops as the critique: the side-by-side is the view the build thread never has on its own, and a region it scores missing or contradicted is a fix whatever the page looks like from memory. Never judge fidelity from one full-page thumbnail; it hides exactly the failures that matter. On a Persuade surface, verify the mode did its job: a first-time visitor should know what this is, why it matters, and what to do within seconds, in the form's own vocabulary.
A capture is evidence only when it is valid, and you validate before you send. Settle or disable entrance motion first: an element hidden by animation timing reads as a missing element and gets fixed into a regression. Capture full-page shots from the document top. Capture the comp comparison at the comp's own pixel dimensions. Then open every file once and confirm it shows what its name claims: no black or blank regions, no wrong section behind a right filename, no half-loaded state. A malformed capture sent onward costs the whole round; the reviewer answers it with `disposition: recapture` and nothing it reviewed binds.
diff --git a/.agents/skills/impeccable/reference/overdrive.md b/.agents/skills/impeccable/reference/overdrive.md
index b07399f3f..9f8fc8eec 100644
--- a/.agents/skills/impeccable/reference/overdrive.md
+++ b/.agents/skills/impeccable/reference/overdrive.md
@@ -21,7 +21,7 @@ Skipping this step risks building something embarrassing that needs to be thrown
### Verify in the Browser
-Ambitious effects rarely work on the first try, so verify them in the browser rather than from code: an effect that technically runs can still stutter, ease wrong, or land on the wrong frame. SKILL.md's bounded passes apply: one batched inspection, one fix batch, at most one confirming round. Point them at the effect's timing, easing, and smoothness, then stop and name what still falls short instead of opening another round.
+Ambitious effects rarely work on the first try, so verify them in the browser rather than from code: an effect that technically runs can still stutter, ease wrong, or land on the wrong frame. Aim the checks at timing, easing, and smoothness, batch the fixes between looks, and keep refining until the effect is extraordinary.
---
diff --git a/.claude/skills/impeccable/SKILL.md b/.claude/skills/impeccable/SKILL.md
index 406d4e296..7520c5827 100644
--- a/.claude/skills/impeccable/SKILL.md
+++ b/.claude/skills/impeccable/SKILL.md
@@ -12,7 +12,7 @@ This skill gives you the tools and permission to make design that earns the name
Core principles:
- Finish the job. No hedging, no shortcuts: the deliverable is complete, except assets the user must provide.
- Scale ambition to the surface's mode (see Modes). Persuade and Experience get full commitment: bold, distinct work with a point of view. On Operate and Read, ambition shows as precision and restraint, and patterns the visitor already knows earn their place.
-- Verify in bounded passes, not a loop, and the ceiling covers the whole cycle: screenshots, defect scans, micro-edits, and rebuilds alike. Build fully, inspect once with a batched round (desktop and mobile together on the web; the shipped device classes on a native platform), fix everything it shows in one batch, confirm with at most one more round, and stop polishing. Open-ended self-QA burns the user's money doing worse what the finish handoffs do better.
+- Verify by looking at the rendered result at the sizes it ships at, and batch the checks: fix what one look shows before looking again, rather than screenshotting after every tweak.
## Setup
diff --git a/.claude/skills/impeccable/reference/craft-floor.md b/.claude/skills/impeccable/reference/craft-floor.md
index 9653b9808..64a195eba 100644
--- a/.claude/skills/impeccable/reference/craft-floor.md
+++ b/.claude/skills/impeccable/reference/craft-floor.md
@@ -4,7 +4,7 @@ Load this after the direction is settled, and build without announcing the check
## 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.
+Each of these is a check on the built result, not an intention. Run them together in each batched look at the render, not as separate screenshot trips; the checks share one render.
- **Contrast:** body and placeholder text ≥4.5:1, large text ≥3:1. On colored surfaces tint secondary text from that hue or the foreground; never gray.
- **Depth:** shadows carry an offset and a soft blur. A zero-offset colored halo is decoration.
diff --git a/.claude/skills/impeccable/reference/new-work.md b/.claude/skills/impeccable/reference/new-work.md
index 24ffb896c..cd0fe5b3d 100644
--- a/.claude/skills/impeccable/reference/new-work.md
+++ b/.claude/skills/impeccable/reference/new-work.md
@@ -135,7 +135,7 @@ Preserve semantics, accessibility, performance, responsiveness, project conventi
## 7. Inspect and finish
-Inspect the surface's target sizes in one batched screenshot round: desktop and mobile on the web; on a native platform (`ios` / `android` / `adaptive`), the shipped device classes per OS, captured from the simulator or emulator the way the platform reference's Verifying the build section describes. When the harness reports the user's actual viewport (an in-app browser's size, a named resolution), add that width to the set: the width that breaks is the one the user sees first. Critique the render against the user's request and the direction contract, fix material gaps, and confirm with one final round; two rounds is the ceiling, and fixes batch between them rather than earning per-tweak screenshots. The ceiling bounds inspection and never closes an open build-phase gate: a gate stays open until it passes or the route it names is taken. On a comp-led build, run `.claude/skills/impeccable/scripts/impeccable comp-diff --comp <approved comp> --build .impeccable/review/desktop.png --spec .impeccable/build/spec.json --out-dir .impeccable/review/diff/final` and read its region rows and paired crops as the critique: the side-by-side is the view the build thread never has on its own, and a region it scores missing or contradicted is a fix whatever the page looks like from memory. Never judge fidelity from one full-page thumbnail; it hides exactly the failures that matter. On a Persuade surface, verify the mode did its job: a first-time visitor should know what this is, why it matters, and what to do within seconds, in the form's own vocabulary.
+Inspect the surface's target sizes in one batched screenshot round: desktop and mobile on the web; on a native platform (`ios` / `android` / `adaptive`), the shipped device classes per OS, captured from the simulator or emulator the way the platform reference's Verifying the build section describes. When the harness reports the user's actual viewport (an in-app browser's size, a named resolution), add that width to the set: the width that breaks is the one the user sees first. On anything interactive, the round drives every interaction and reachable state (open, closed, hover, focus, error, empty, overlong content) and checks each for overlap and clipping, not only the resting screen. Critique the render against the user's request and the direction contract, fix material gaps, and confirm with one final round; two rounds is the ceiling, and fixes batch between them rather than earning per-tweak screenshots. The ceiling bounds inspection and never closes an open build-phase gate: a gate stays open until it passes or the route it names is taken. On a comp-led build, run `.claude/skills/impeccable/scripts/impeccable comp-diff --comp <approved comp> --build .impeccable/review/desktop.png --spec .impeccable/build/spec.json --out-dir .impeccable/review/diff/final` and read its region rows and paired crops as the critique: the side-by-side is the view the build thread never has on its own, and a region it scores missing or contradicted is a fix whatever the page looks like from memory. Never judge fidelity from one full-page thumbnail; it hides exactly the failures that matter. On a Persuade surface, verify the mode did its job: a first-time visitor should know what this is, why it matters, and what to do within seconds, in the form's own vocabulary.
A capture is evidence only when it is valid, and you validate before you send. Settle or disable entrance motion first: an element hidden by animation timing reads as a missing element and gets fixed into a regression. Capture full-page shots from the document top. Capture the comp comparison at the comp's own pixel dimensions. Then open every file once and confirm it shows what its name claims: no black or blank regions, no wrong section behind a right filename, no half-loaded state. A malformed capture sent onward costs the whole round; the reviewer answers it with `disposition: recapture` and nothing it reviewed binds.
diff --git a/.claude/skills/impeccable/reference/overdrive.md b/.claude/skills/impeccable/reference/overdrive.md
index fd8f544f3..01a1cb131 100644
--- a/.claude/skills/impeccable/reference/overdrive.md
+++ b/.claude/skills/impeccable/reference/overdrive.md
@@ -21,7 +21,7 @@ Skipping this step risks building something embarrassing that needs to be thrown
### Verify in the Browser
-Ambitious effects rarely work on the first try, so verify them in the browser rather than from code: an effect that technically runs can still stutter, ease wrong, or land on the wrong frame. SKILL.md's bounded passes apply: one batched inspection, one fix batch, at most one confirming round. Point them at the effect's timing, easing, and smoothness, then stop and name what still falls short instead of opening another round.
+Ambitious effects rarely work on the first try, so verify them in the browser rather than from code: an effect that technically runs can still stutter, ease wrong, or land on the wrong frame. Aim the checks at timing, easing, and smoothness, batch the fixes between looks, and keep refining until the effect is extraordinary.
---
diff --git a/.cursor/skills/impeccable/SKILL.md b/.cursor/skills/impeccable/SKILL.md
index cdcb8696d..4e3c05377 100644
--- a/.cursor/skills/impeccable/SKILL.md
+++ b/.cursor/skills/impeccable/SKILL.md
@@ -10,7 +10,7 @@ This skill gives you the tools and permission to make design that earns the name
Core principles:
- Finish the job. No hedging, no shortcuts: the deliverable is complete, except assets the user must provide.
- Scale ambition to the surface's mode (see Modes). Persuade and Experience get full commitment: bold, distinct work with a point of view. On Operate and Read, ambition shows as precision and restraint, and patterns the visitor already knows earn their place.
-- Verify in bounded passes, not a loop, and the ceiling covers the whole cycle: screenshots, defect scans, micro-edits, and rebuilds alike. Build fully, inspect once with a batched round (desktop and mobile together on the web; the shipped device classes on a native platform), fix everything it shows in one batch, confirm with at most one more round, and stop polishing. Open-ended self-QA burns the user's money doing worse what the finish handoffs do better.
+- Verify by looking at the rendered result at the sizes it ships at, and batch the checks: fix what one look shows before looking again, rather than screenshotting after every tweak.
## Setup
diff --git a/.cursor/skills/impeccable/reference/craft-floor.md b/.cursor/skills/impeccable/reference/craft-floor.md
index 9653b9808..64a195eba 100644
--- a/.cursor/skills/impeccable/reference/craft-floor.md
+++ b/.cursor/skills/impeccable/reference/craft-floor.md
@@ -4,7 +4,7 @@ Load this after the direction is settled, and build without announcing the check
## 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.
+Each of these is a check on the built result, not an intention. Run them together in each batched look at the render, not as separate screenshot trips; the checks share one render.
- **Contrast:** body and placeholder text ≥4.5:1, large text ≥3:1. On colored surfaces tint secondary text from that hue or the foreground; never gray.
- **Depth:** shadows carry an offset and a soft blur. A zero-offset colored halo is decoration.
diff --git a/.cursor/skills/impeccable/reference/new-work.md b/.cursor/skills/impeccable/reference/new-work.md
index ff2ebb27d..85862f90a 100644
--- a/.cursor/skills/impeccable/reference/new-work.md
+++ b/.cursor/skills/impeccable/reference/new-work.md
@@ -133,7 +133,7 @@ Preserve semantics, accessibility, performance, responsiveness, project conventi
## 7. Inspect and finish
-Inspect the surface's target sizes in one batched screenshot round: desktop and mobile on the web; on a native platform (`ios` / `android` / `adaptive`), the shipped device classes per OS, captured from the simulator or emulator the way the platform reference's Verifying the build section describes. When the harness reports the user's actual viewport (an in-app browser's size, a named resolution), add that width to the set: the width that breaks is the one the user sees first. Critique the render against the user's request and the direction contract, fix material gaps, and confirm with one final round; two rounds is the ceiling, and fixes batch between them rather than earning per-tweak screenshots. The ceiling bounds inspection and never closes an open build-phase gate: a gate stays open until it passes or the route it names is taken. On a comp-led build, run `.cursor/skills/impeccable/scripts/impeccable comp-diff --comp <approved comp> --build .impeccable/review/desktop.png --spec .impeccable/build/spec.json --out-dir .impeccable/review/diff/final` and read its region rows and paired crops as the critique: the side-by-side is the view the build thread never has on its own, and a region it scores missing or contradicted is a fix whatever the page looks like from memory. Never judge fidelity from one full-page thumbnail; it hides exactly the failures that matter. On a Persuade surface, verify the mode did its job: a first-time visitor should know what this is, why it matters, and what to do within seconds, in the form's own vocabulary.
+Inspect the surface's target sizes in one batched screenshot round: desktop and mobile on the web; on a native platform (`ios` / `android` / `adaptive`), the shipped device classes per OS, captured from the simulator or emulator the way the platform reference's Verifying the build section describes. When the harness reports the user's actual viewport (an in-app browser's size, a named resolution), add that width to the set: the width that breaks is the one the user sees first. On anything interactive, the round drives every interaction and reachable state (open, closed, hover, focus, error, empty, overlong content) and checks each for overlap and clipping, not only the resting screen. Critique the render against the user's request and the direction contract, fix material gaps, and confirm with one final round; two rounds is the ceiling, and fixes batch between them rather than earning per-tweak screenshots. The ceiling bounds inspection and never closes an open build-phase gate: a gate stays open until it passes or the route it names is taken. On a comp-led build, run `.cursor/skills/impeccable/scripts/impeccable comp-diff --comp <approved comp> --build .impeccable/review/desktop.png --spec .impeccable/build/spec.json --out-dir .impeccable/review/diff/final` and read its region rows and paired crops as the critique: the side-by-side is the view the build thread never has on its own, and a region it scores missing or contradicted is a fix whatever the page looks like from memory. Never judge fidelity from one full-page thumbnail; it hides exactly the failures that matter. On a Persuade surface, verify the mode did its job: a first-time visitor should know what this is, why it matters, and what to do within seconds, in the form's own vocabulary.
A capture is evidence only when it is valid, and you validate before you send. Settle or disable entrance motion first: an element hidden by animation timing reads as a missing element and gets fixed into a regression. Capture full-page shots from the document top. Capture the comp comparison at the comp's own pixel dimensions. Then open every file once and confirm it shows what its name claims: no black or blank regions, no wrong section behind a right filename, no half-loaded state. A malformed capture sent onward costs the whole round; the reviewer answers it with `disposition: recapture` and nothing it reviewed binds.
diff --git a/.cursor/skills/impeccable/reference/overdrive.md b/.cursor/skills/impeccable/reference/overdrive.md
index 89be08b5c..b3dfb708c 100644
--- a/.cursor/skills/impeccable/reference/overdrive.md
+++ b/.cursor/skills/impeccable/reference/overdrive.md
@@ -21,7 +21,7 @@ Skipping this step risks building something embarrassing that needs to be thrown
### Verify in the Browser
-Ambitious effects rarely work on the first try, so verify them in the browser rather than from code: an effect that technically runs can still stutter, ease wrong, or land on the wrong frame. SKILL.md's bounded passes apply: one batched inspection, one fix batch, at most one confirming round. Point them at the effect's timing, easing, and smoothness, then stop and name what still falls short instead of opening another round.
+Ambitious effects rarely work on the first try, so verify them in the browser rather than from code: an effect that technically runs can still stutter, ease wrong, or land on the wrong frame. Aim the checks at timing, easing, and smoothness, batch the fixes between looks, and keep refining until the effect is extraordinary.
---
diff --git a/.dsh/skills/impeccable/SKILL.md b/.dsh/skills/impeccable/SKILL.md
index d194242b9..cff91a577 100644
--- a/.dsh/skills/impeccable/SKILL.md
+++ b/.dsh/skills/impeccable/SKILL.md
@@ -11,7 +11,7 @@ This skill gives you the tools and permission to make design that earns the name
Core principles:
- Finish the job. No hedging, no shortcuts: the deliverable is complete, except assets the user must provide.
- Scale ambition to the surface's mode (see Modes). Persuade and Experience get full commitment: bold, distinct work with a point of view. On Operate and Read, ambition shows as precision and restraint, and patterns the visitor already knows earn their place.
-- Verify in bounded passes, not a loop, and the ceiling covers the whole cycle: screenshots, defect scans, micro-edits, and rebuilds alike. Build fully, inspect once with a batched round (desktop and mobile together on the web; the shipped device classes on a native platform), fix everything it shows in one batch, confirm with at most one more round, and stop polishing. Open-ended self-QA burns the user's money doing worse what the finish handoffs do better.
+- Verify by looking at the rendered result at the sizes it ships at, and batch the checks: fix what one look shows before looking again, rather than screenshotting after every tweak.
## Setup
diff --git a/.dsh/skills/impeccable/reference/craft-floor.md b/.dsh/skills/impeccable/reference/craft-floor.md
index 9653b9808..64a195eba 100644
--- a/.dsh/skills/impeccable/reference/craft-floor.md
+++ b/.dsh/skills/impeccable/reference/craft-floor.md
@@ -4,7 +4,7 @@ Load this after the direction is settled, and build without announcing the check
## 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.
+Each of these is a check on the built result, not an intention. Run them together in each batched look at the render, not as separate screenshot trips; the checks share one render.
- **Contrast:** body and placeholder text ≥4.5:1, large text ≥3:1. On colored surfaces tint secondary text from that hue or the foreground; never gray.
- **Depth:** shadows carry an offset and a soft blur. A zero-offset colored halo is decoration.
diff --git a/.dsh/skills/impeccable/reference/new-work.md b/.dsh/skills/impeccable/reference/new-work.md
index 21d8342f8..8722bb976 100644
--- a/.dsh/skills/impeccable/reference/new-work.md
+++ b/.dsh/skills/impeccable/reference/new-work.md
@@ -133,7 +133,7 @@ Preserve semantics, accessibility, performance, responsiveness, project conventi
## 7. Inspect and finish
-Inspect the surface's target sizes in one batched screenshot round: desktop and mobile on the web; on a native platform (`ios` / `android` / `adaptive`), the shipped device classes per OS, captured from the simulator or emulator the way the platform reference's Verifying the build section describes. When the harness reports the user's actual viewport (an in-app browser's size, a named resolution), add that width to the set: the width that breaks is the one the user sees first. Critique the render against the user's request and the direction contract, fix material gaps, and confirm with one final round; two rounds is the ceiling, and fixes batch between them rather than earning per-tweak screenshots. The ceiling bounds inspection and never closes an open build-phase gate: a gate stays open until it passes or the route it names is taken. On a comp-led build, run `.dsh/skills/impeccable/scripts/impeccable comp-diff --comp <approved comp> --build .impeccable/review/desktop.png --spec .impeccable/build/spec.json --out-dir .impeccable/review/diff/final` and read its region rows and paired crops as the critique: the side-by-side is the view the build thread never has on its own, and a region it scores missing or contradicted is a fix whatever the page looks like from memory. Never judge fidelity from one full-page thumbnail; it hides exactly the failures that matter. On a Persuade surface, verify the mode did its job: a first-time visitor should know what this is, why it matters, and what to do within seconds, in the form's own vocabulary.
+Inspect the surface's target sizes in one batched screenshot round: desktop and mobile on the web; on a native platform (`ios` / `android` / `adaptive`), the shipped device classes per OS, captured from the simulator or emulator the way the platform reference's Verifying the build section describes. When the harness reports the user's actual viewport (an in-app browser's size, a named resolution), add that width to the set: the width that breaks is the one the user sees first. On anything interactive, the round drives every interaction and reachable state (open, closed, hover, focus, error, empty, overlong content) and checks each for overlap and clipping, not only the resting screen. Critique the render against the user's request and the direction contract, fix material gaps, and confirm with one final round; two rounds is the ceiling, and fixes batch between them rather than earning per-tweak screenshots. The ceiling bounds inspection and never closes an open build-phase gate: a gate stays open until it passes or the route it names is taken. On a comp-led build, run `.dsh/skills/impeccable/scripts/impeccable comp-diff --comp <approved comp> --build .impeccable/review/desktop.png --spec .impeccable/build/spec.json --out-dir .impeccable/review/diff/final` and read its region rows and paired crops as the critique: the side-by-side is the view the build thread never has on its own, and a region it scores missing or contradicted is a fix whatever the page looks like from memory. Never judge fidelity from one full-page thumbnail; it hides exactly the failures that matter. On a Persuade surface, verify the mode did its job: a first-time visitor should know what this is, why it matters, and what to do within seconds, in the form's own vocabulary.
A capture is evidence only when it is valid, and you validate before you send. Settle or disable entrance motion first: an element hidden by animation timing reads as a missing element and gets fixed into a regression. Capture full-page shots from the document top. Capture the comp comparison at the comp's own pixel dimensions. Then open every file once and confirm it shows what its name claims: no black or blank regions, no wrong section behind a right filename, no half-loaded state. A malformed capture sent onward costs the whole round; the reviewer answers it with `disposition: recapture` and nothing it reviewed binds.
diff --git a/.dsh/skills/impeccable/reference/overdrive.md b/.dsh/skills/impeccable/reference/overdrive.md
index bdd7bc438..bdc17d162 100644
--- a/.dsh/skills/impeccable/reference/overdrive.md
+++ b/.dsh/skills/impeccable/reference/overdrive.md
@@ -21,7 +21,7 @@ Skipping this step risks building something embarrassing that needs to be thrown
### Verify in the Browser
-Ambitious effects rarely work on the first try, so verify them in the browser rather than from code: an effect that technically runs can still stutter, ease wrong, or land on the wrong frame. SKILL.md's bounded passes apply: one batched inspection, one fix batch, at most one confirming round. Point them at the effect's timing, easing, and smoothness, then stop and name what still falls short instead of opening another round.
+Ambitious effects rarely work on the first try, so verify them in the browser rather than from code: an effect that technically runs can still stutter, ease wrong, or land on the wrong frame. Aim the checks at timing, easing, and smoothness, batch the fixes between looks, and keep refining until the effect is extraordinary.
---
diff --git a/.gemini/skills/impeccable/SKILL.md b/.gemini/skills/impeccable/SKILL.md
index 7f22169fc..f667d6448 100644
--- a/.gemini/skills/impeccable/SKILL.md
+++ b/.gemini/skills/impeccable/SKILL.md
@@ -9,7 +9,7 @@ This skill gives you the tools and permission to make design that earns the name
Core principles:
- Finish the job. No hedging, no shortcuts: the deliverable is complete, except assets the user must provide.
- Scale ambition to the surface's mode (see Modes). Persuade and Experience get full commitment: bold, distinct work with a point of view. On Operate and Read, ambition shows as precision and restraint, and patterns the visitor already knows earn their place.
-- Verify in bounded passes, not a loop, and the ceiling covers the whole cycle: screenshots, defect scans, micro-edits, and rebuilds alike. Build fully, inspect once with a batched round (desktop and mobile together on the web; the shipped device classes on a native platform), fix everything it shows in one batch, confirm with at most one more round, and stop polishing. Open-ended self-QA burns the user's money doing worse what the finish handoffs do better.
+- Verify by looking at the rendered result at the sizes it ships at, and batch the checks: fix what one look shows before looking again, rather than screenshotting after every tweak.
## Setup
diff --git a/.gemini/skills/impeccable/reference/craft-floor.md b/.gemini/skills/impeccable/reference/craft-floor.md
index 9653b9808..64a195eba 100644
--- a/.gemini/skills/impeccable/reference/craft-floor.md
+++ b/.gemini/skills/impeccable/reference/craft-floor.md
@@ -4,7 +4,7 @@ Load this after the direction is settled, and build without announcing the check
## 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.
+Each of these is a check on the built result, not an intention. Run them together in each batched look at the render, not as separate screenshot trips; the checks share one render.
- **Contrast:** body and placeholder text ≥4.5:1, large text ≥3:1. On colored surfaces tint secondary text from that hue or the foreground; never gray.
- **Depth:** shadows carry an offset and a soft blur. A zero-offset colored halo is decoration.
diff --git a/.gemini/skills/impeccable/reference/new-work.md b/.gemini/skills/impeccable/reference/new-work.md
index 859dcc1b2..3098c150d 100644
--- a/.gemini/skills/impeccable/reference/new-work.md
+++ b/.gemini/skills/impeccable/reference/new-work.md
@@ -133,7 +133,7 @@ Preserve semantics, accessibility, performance, responsiveness, project conventi
## 7. Inspect and finish
-Inspect the surface's target sizes in one batched screenshot round: desktop and mobile on the web; on a native platform (`ios` / `android` / `adaptive`), the shipped device classes per OS, captured from the simulator or emulator the way the platform reference's Verifying the build section describes. When the harness reports the user's actual viewport (an in-app browser's size, a named resolution), add that width to the set: the width that breaks is the one the user sees first. Critique the render against the user's request and the direction contract, fix material gaps, and confirm with one final round; two rounds is the ceiling, and fixes batch between them rather than earning per-tweak screenshots. The ceiling bounds inspection and never closes an open build-phase gate: a gate stays open until it passes or the route it names is taken. On a comp-led build, run `.gemini/skills/impeccable/scripts/impeccable comp-diff --comp <approved comp> --build .impeccable/review/desktop.png --spec .impeccable/build/spec.json --out-dir .impeccable/review/diff/final` and read its region rows and paired crops as the critique: the side-by-side is the view the build thread never has on its own, and a region it scores missing or contradicted is a fix whatever the page looks like from memory. Never judge fidelity from one full-page thumbnail; it hides exactly the failures that matter. On a Persuade surface, verify the mode did its job: a first-time visitor should know what this is, why it matters, and what to do within seconds, in the form's own vocabulary.
+Inspect the surface's target sizes in one batched screenshot round: desktop and mobile on the web; on a native platform (`ios` / `android` / `adaptive`), the shipped device classes per OS, captured from the simulator or emulator the way the platform reference's Verifying the build section describes. When the harness reports the user's actual viewport (an in-app browser's size, a named resolution), add that width to the set: the width that breaks is the one the user sees first. On anything interactive, the round drives every interaction and reachable state (open, closed, hover, focus, error, empty, overlong content) and checks each for overlap and clipping, not only the resting screen. Critique the render against the user's request and the direction contract, fix material gaps, and confirm with one final round; two rounds is the ceiling, and fixes batch between them rather than earning per-tweak screenshots. The ceiling bounds inspection and never closes an open build-phase gate: a gate stays open until it passes or the route it names is taken. On a comp-led build, run `.gemini/skills/impeccable/scripts/impeccable comp-diff --comp <approved comp> --build .impeccable/review/desktop.png --spec .impeccable/build/spec.json --out-dir .impeccable/review/diff/final` and read its region rows and paired crops as the critique: the side-by-side is the view the build thread never has on its own, and a region it scores missing or contradicted is a fix whatever the page looks like from memory. Never judge fidelity from one full-page thumbnail; it hides exactly the failures that matter. On a Persuade surface, verify the mode did its job: a first-time visitor should know what this is, why it matters, and what to do within seconds, in the form's own vocabulary.
A capture is evidence only when it is valid, and you validate before you send. Settle or disable entrance motion first: an element hidden by animation timing reads as a missing element and gets fixed into a regression. Capture full-page shots from the document top. Capture the comp comparison at the comp's own pixel dimensions. Then open every file once and confirm it shows what its name claims: no black or blank regions, no wrong section behind a right filename, no half-loaded state. A malformed capture sent onward costs the whole round; the reviewer answers it with `disposition: recapture` and nothing it reviewed binds.
diff --git a/.gemini/skills/impeccable/reference/overdrive.md b/.gemini/skills/impeccable/reference/overdrive.md
index 89be08b5c..b3dfb708c 100644
--- a/.gemini/skills/impeccable/reference/overdrive.md
+++ b/.gemini/skills/impeccable/reference/overdrive.md
@@ -21,7 +21,7 @@ Skipping this step risks building something embarrassing that needs to be thrown
### Verify in the Browser
-Ambitious effects rarely work on the first try, so verify them in the browser rather than from code: an effect that technically runs can still stutter, ease wrong, or land on the wrong frame. SKILL.md's bounded passes apply: one batched inspection, one fix batch, at most one confirming round. Point them at the effect's timing, easing, and smoothness, then stop and name what still falls short instead of opening another round.
+Ambitious effects rarely work on the first try, so verify them in the browser rather than from code: an effect that technically runs can still stutter, ease wrong, or land on the wrong frame. Aim the checks at timing, easing, and smoothness, batch the fixes between looks, and keep refining until the effect is extraordinary.
---
diff --git a/.github/skills/impeccable/SKILL.md b/.github/skills/impeccable/SKILL.md
index a19155bd1..80afecd42 100644
--- a/.github/skills/impeccable/SKILL.md
+++ b/.github/skills/impeccable/SKILL.md
@@ -12,7 +12,7 @@ This skill gives you the tools and permission to make design that earns the name
Core principles:
- Finish the job. No hedging, no shortcuts: the deliverable is complete, except assets the user must provide.
- Scale ambition to the surface's mode (see Modes). Persuade and Experience get full commitment: bold, distinct work with a point of view. On Operate and Read, ambition shows as precision and restraint, and patterns the visitor already knows earn their place.
-- Verify in bounded passes, not a loop, and the ceiling covers the whole cycle: screenshots, defect scans, micro-edits, and rebuilds alike. Build fully, inspect once with a batched round (desktop and mobile together on the web; the shipped device classes on a native platform), fix everything it shows in one batch, confirm with at most one more round, and stop polishing. Open-ended self-QA burns the user's money doing worse what the finish handoffs do better.
+- Verify by looking at the rendered result at the sizes it ships at, and batch the checks: fix what one look shows before looking again, rather than screenshotting after every tweak.
## Setup
diff --git a/.github/skills/impeccable/reference/craft-floor.md b/.github/skills/impeccable/reference/craft-floor.md
index 9653b9808..64a195eba 100644
--- a/.github/skills/impeccable/reference/craft-floor.md
+++ b/.github/skills/impeccable/reference/craft-floor.md
@@ -4,7 +4,7 @@ Load this after the direction is settled, and build without announcing the check
## 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.
+Each of these is a check on the built result, not an intention. Run them together in each batched look at the render, not as separate screenshot trips; the checks share one render.
- **Contrast:** body and placeholder text ≥4.5:1, large text ≥3:1. On colored surfaces tint secondary text from that hue or the foreground; never gray.
- **Depth:** shadows carry an offset and a soft blur. A zero-offset colored halo is decoration.
diff --git a/.github/skills/impeccable/reference/new-work.md b/.github/skills/impeccable/reference/new-work.md
index 774708d2d..ba6622e4f 100644
--- a/.github/skills/impeccable/reference/new-work.md
+++ b/.github/skills/impeccable/reference/new-work.md
@@ -133,7 +133,7 @@ Preserve semantics, accessibility, performance, responsiveness, project conventi
## 7. Inspect and finish
-Inspect the surface's target sizes in one batched screenshot round: desktop and mobile on the web; on a native platform (`ios` / `android` / `adaptive`), the shipped device classes per OS, captured from the simulator or emulator the way the platform reference's Verifying the build section describes. When the harness reports the user's actual viewport (an in-app browser's size, a named resolution), add that width to the set: the width that breaks is the one the user sees first. Critique the render against the user's request and the direction contract, fix material gaps, and confirm with one final round; two rounds is the ceiling, and fixes batch between them rather than earning per-tweak screenshots. The ceiling bounds inspection and never closes an open build-phase gate: a gate stays open until it passes or the route it names is taken. On a comp-led build, run `.github/skills/impeccable/scripts/impeccable comp-diff --comp <approved comp> --build .impeccable/review/desktop.png --spec .impeccable/build/spec.json --out-dir .impeccable/review/diff/final` and read its region rows and paired crops as the critique: the side-by-side is the view the build thread never has on its own, and a region it scores missing or contradicted is a fix whatever the page looks like from memory. Never judge fidelity from one full-page thumbnail; it hides exactly the failures that matter. On a Persuade surface, verify the mode did its job: a first-time visitor should know what this is, why it matters, and what to do within seconds, in the form's own vocabulary.
+Inspect the surface's target sizes in one batched screenshot round: desktop and mobile on the web; on a native platform (`ios` / `android` / `adaptive`), the shipped device classes per OS, captured from the simulator or emulator the way the platform reference's Verifying the build section describes. When the harness reports the user's actual viewport (an in-app browser's size, a named resolution), add that width to the set: the width that breaks is the one the user sees first. On anything interactive, the round drives every interaction and reachable state (open, closed, hover, focus, error, empty, overlong content) and checks each for overlap and clipping, not only the resting screen. Critique the render against the user's request and the direction contract, fix material gaps, and confirm with one final round; two rounds is the ceiling, and fixes batch between them rather than earning per-tweak screenshots. The ceiling bounds inspection and never closes an open build-phase gate: a gate stays open until it passes or the route it names is taken. On a comp-led build, run `.github/skills/impeccable/scripts/impeccable comp-diff --comp <approved comp> --build .impeccable/review/desktop.png --spec .impeccable/build/spec.json --out-dir .impeccable/review/diff/final` and read its region rows and paired crops as the critique: the side-by-side is the view the build thread never has on its own, and a region it scores missing or contradicted is a fix whatever the page looks like from memory. Never judge fidelity from one full-page thumbnail; it hides exactly the failures that matter. On a Persuade surface, verify the mode did its job: a first-time visitor should know what this is, why it matters, and what to do within seconds, in the form's own vocabulary.
A capture is evidence only when it is valid, and you validate before you send. Settle or disable entrance motion first: an element hidden by animation timing reads as a missing element and gets fixed into a regression. Capture full-page shots from the document top. Capture the comp comparison at the comp's own pixel dimensions. Then open every file once and confirm it shows what its name claims: no black or blank regions, no wrong section behind a right filename, no half-loaded state. A malformed capture sent onward costs the whole round; the reviewer answers it with `disposition: recapture` and nothing it reviewed binds.
diff --git a/.github/skills/impeccable/reference/overdrive.md b/.github/skills/impeccable/reference/overdrive.md
index 89be08b5c..b3dfb708c 100644
--- a/.github/skills/impeccable/reference/overdrive.md
+++ b/.github/skills/impeccable/reference/overdrive.md
@@ -21,7 +21,7 @@ Skipping this step risks building something embarrassing that needs to be thrown
### Verify in the Browser
-Ambitious effects rarely work on the first try, so verify them in the browser rather than from code: an effect that technically runs can still stutter, ease wrong, or land on the wrong frame. SKILL.md's bounded passes apply: one batched inspection, one fix batch, at most one confirming round. Point them at the effect's timing, easing, and smoothness, then stop and name what still falls short instead of opening another round.
+Ambitious effects rarely work on the first try, so verify them in the browser rather than from code: an effect that technically runs can still stutter, ease wrong, or land on the wrong frame. Aim the checks at timing, easing, and smoothness, batch the fixes between looks, and keep refining until the effect is extraordinary.
---
diff --git a/.grok/skills/impeccable/SKILL.md b/.grok/skills/impeccable/SKILL.md
index 0dd48f30c..aea48a70d 100644
--- a/.grok/skills/impeccable/SKILL.md
+++ b/.grok/skills/impeccable/SKILL.md
@@ -15,7 +15,7 @@ This skill gives you the tools and permission to make design that earns the name
Core principles:
- Finish the job. No hedging, no shortcuts: the deliverable is complete, except assets the user must provide.
- Scale ambition to the surface's mode (see Modes). Persuade and Experience get full commitment: bold, distinct work with a point of view. On Operate and Read, ambition shows as precision and restraint, and patterns the visitor already knows earn their place.
-- Verify in bounded passes, not a loop, and the ceiling covers the whole cycle: screenshots, defect scans, micro-edits, and rebuilds alike. Build fully, inspect once with a batched round (desktop and mobile together on the web; the shipped device classes on a native platform), fix everything it shows in one batch, confirm with at most one more round, and stop polishing. Open-ended self-QA burns the user's money doing worse what the finish handoffs do better.
+- Verify by looking at the rendered result at the sizes it ships at, and batch the checks: fix what one look shows before looking again, rather than screenshotting after every tweak.
## Setup
diff --git a/.grok/skills/impeccable/reference/craft-floor.md b/.grok/skills/impeccable/reference/craft-floor.md
index 9653b9808..64a195eba 100644
--- a/.grok/skills/impeccable/reference/craft-floor.md
+++ b/.grok/skills/impeccable/reference/craft-floor.md
@@ -4,7 +4,7 @@ Load this after the direction is settled, and build without announcing the check
## 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.
+Each of these is a check on the built result, not an intention. Run them together in each batched look at the render, not as separate screenshot trips; the checks share one render.
- **Contrast:** body and placeholder text ≥4.5:1, large text ≥3:1. On colored surfaces tint secondary text from that hue or the foreground; never gray.
- **Depth:** shadows carry an offset and a soft blur. A zero-offset colored halo is decoration.
diff --git a/.grok/skills/impeccable/reference/new-work.md b/.grok/skills/impeccable/reference/new-work.md
index 38ac0ec31..d2c8ea78c 100644
--- a/.grok/skills/impeccable/reference/new-work.md
+++ b/.grok/skills/impeccable/reference/new-work.md
@@ -133,7 +133,7 @@ Preserve semantics, accessibility, performance, responsiveness, project conventi
## 7. Inspect and finish
-Inspect the surface's target sizes in one batched screenshot round: desktop and mobile on the web; on a native platform (`ios` / `android` / `adaptive`), the shipped device classes per OS, captured from the simulator or emulator the way the platform reference's Verifying the build section describes. When the harness reports the user's actual viewport (an in-app browser's size, a named resolution), add that width to the set: the width that breaks is the one the user sees first. Critique the render against the user's request and the direction contract, fix material gaps, and confirm with one final round; two rounds is the ceiling, and fixes batch between them rather than earning per-tweak screenshots. The ceiling bounds inspection and never closes an open build-phase gate: a gate stays open until it passes or the route it names is taken. On a comp-led build, run `.grok/skills/impeccable/scripts/impeccable comp-diff --comp <approved comp> --build .impeccable/review/desktop.png --spec .impeccable/build/spec.json --out-dir .impeccable/review/diff/final` and read its region rows and paired crops as the critique: the side-by-side is the view the build thread never has on its own, and a region it scores missing or contradicted is a fix whatever the page looks like from memory. Never judge fidelity from one full-page thumbnail; it hides exactly the failures that matter. On a Persuade surface, verify the mode did its job: a first-time visitor should know what this is, why it matters, and what to do within seconds, in the form's own vocabulary.
+Inspect the surface's target sizes in one batched screenshot round: desktop and mobile on the web; on a native platform (`ios` / `android` / `adaptive`), the shipped device classes per OS, captured from the simulator or emulator the way the platform reference's Verifying the build section describes. When the harness reports the user's actual viewport (an in-app browser's size, a named resolution), add that width to the set: the width that breaks is the one the user sees first. On anything interactive, the round drives every interaction and reachable state (open, closed, hover, focus, error, empty, overlong content) and checks each for overlap and clipping, not only the resting screen. Critique the render against the user's request and the direction contract, fix material gaps, and confirm with one final round; two rounds is the ceiling, and fixes batch between them rather than earning per-tweak screenshots. The ceiling bounds inspection and never closes an open build-phase gate: a gate stays open until it passes or the route it names is taken. On a comp-led build, run `.grok/skills/impeccable/scripts/impeccable comp-diff --comp <approved comp> --build .impeccable/review/desktop.png --spec .impeccable/build/spec.json --out-dir .impeccable/review/diff/final` and read its region rows and paired crops as the critique: the side-by-side is the view the build thread never has on its own, and a region it scores missing or contradicted is a fix whatever the page looks like from memory. Never judge fidelity from one full-page thumbnail; it hides exactly the failures that matter. On a Persuade surface, verify the mode did its job: a first-time visitor should know what this is, why it matters, and what to do within seconds, in the form's own vocabulary.
A capture is evidence only when it is valid, and you validate before you send. Settle or disable entrance motion first: an element hidden by animation timing reads as a missing element and gets fixed into a regression. Capture full-page shots from the document top. Capture the comp comparison at the comp's own pixel dimensions. Then open every file once and confirm it shows what its name claims: no black or blank regions, no wrong section behind a right filename, no half-loaded state. A malformed capture sent onward costs the whole round; the reviewer answers it with `disposition: recapture` and nothing it reviewed binds.
diff --git a/.grok/skills/impeccable/reference/overdrive.md b/.grok/skills/impeccable/reference/overdrive.md
index fd8f544f3..01a1cb131 100644
--- a/.grok/skills/impeccable/reference/overdrive.md
+++ b/.grok/skills/impeccable/reference/overdrive.md
@@ -21,7 +21,7 @@ Skipping this step risks building something embarrassing that needs to be thrown
### Verify in the Browser
-Ambitious effects rarely work on the first try, so verify them in the browser rather than from code: an effect that technically runs can still stutter, ease wrong, or land on the wrong frame. SKILL.md's bounded passes apply: one batched inspection, one fix batch, at most one confirming round. Point them at the effect's timing, easing, and smoothness, then stop and name what still falls short instead of opening another round.
+Ambitious effects rarely work on the first try, so verify them in the browser rather than from code: an effect that technically runs can still stutter, ease wrong, or land on the wrong frame. Aim the checks at timing, easing, and smoothness, batch the fixes between looks, and keep refining until the effect is extraordinary.
---
diff --git a/.hermes/skills/impeccable/SKILL.md b/.hermes/skills/impeccable/SKILL.md
index ac2ccc7b8..111c509e1 100644
--- a/.hermes/skills/impeccable/SKILL.md
+++ b/.hermes/skills/impeccable/SKILL.md
@@ -10,7 +10,7 @@ This skill gives you the tools and permission to make design that earns the name
Core principles:
- Finish the job. No hedging, no shortcuts: the deliverable is complete, except assets the user must provide.
- Scale ambition to the surface's mode (see Modes). Persuade and Experience get full commitment: bold, distinct work with a point of view. On Operate and Read, ambition shows as precision and restraint, and patterns the visitor already knows earn their place.
-- Verify in bounded passes, not a loop, and the ceiling covers the whole cycle: screenshots, defect scans, micro-edits, and rebuilds alike. Build fully, inspect once with a batched round (desktop and mobile together on the web; the shipped device classes on a native platform), fix everything it shows in one batch, confirm with at most one more round, and stop polishing. Open-ended self-QA burns the user's money doing worse what the finish handoffs do better.
+- Verify by looking at the rendered result at the sizes it ships at, and batch the checks: fix what one look shows before looking again, rather than screenshotting after every tweak.
## Setup
diff --git a/.hermes/skills/impeccable/reference/craft-floor.md b/.hermes/skills/impeccable/reference/craft-floor.md
index 9653b9808..64a195eba 100644
--- a/.hermes/skills/impeccable/reference/craft-floor.md
+++ b/.hermes/skills/impeccable/reference/craft-floor.md
@@ -4,7 +4,7 @@ Load this after the direction is settled, and build without announcing the check
## 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.
+Each of these is a check on the built result, not an intention. Run them together in each batched look at the render, not as separate screenshot trips; the checks share one render.
- **Contrast:** body and placeholder text ≥4.5:1, large text ≥3:1. On colored surfaces tint secondary text from that hue or the foreground; never gray.
- **Depth:** shadows carry an offset and a soft blur. A zero-offset colored halo is decoration.
diff --git a/.hermes/skills/impeccable/reference/new-work.md b/.hermes/skills/impeccable/reference/new-work.md
index ba4d53061..8c8d33870 100644
--- a/.hermes/skills/impeccable/reference/new-work.md
+++ b/.hermes/skills/impeccable/reference/new-work.md
@@ -133,7 +133,7 @@ Preserve semantics, accessibility, performance, responsiveness, project conventi
## 7. Inspect and finish
-Inspect the surface's target sizes in one batched screenshot round: desktop and mobile on the web; on a native platform (`ios` / `android` / `adaptive`), the shipped device classes per OS, captured from the simulator or emulator the way the platform reference's Verifying the build section describes. When the harness reports the user's actual viewport (an in-app browser's size, a named resolution), add that width to the set: the width that breaks is the one the user sees first. Critique the render against the user's request and the direction contract, fix material gaps, and confirm with one final round; two rounds is the ceiling, and fixes batch between them rather than earning per-tweak screenshots. The ceiling bounds inspection and never closes an open build-phase gate: a gate stays open until it passes or the route it names is taken. On a comp-led build, run `.hermes/skills/impeccable/scripts/impeccable comp-diff --comp <approved comp> --build .impeccable/review/desktop.png --spec .impeccable/build/spec.json --out-dir .impeccable/review/diff/final` and read its region rows and paired crops as the critique: the side-by-side is the view the build thread never has on its own, and a region it scores missing or contradicted is a fix whatever the page looks like from memory. Never judge fidelity from one full-page thumbnail; it hides exactly the failures that matter. On a Persuade surface, verify the mode did its job: a first-time visitor should know what this is, why it matters, and what to do within seconds, in the form's own vocabulary.
+Inspect the surface's target sizes in one batched screenshot round: desktop and mobile on the web; on a native platform (`ios` / `android` / `adaptive`), the shipped device classes per OS, captured from the simulator or emulator the way the platform reference's Verifying the build section describes. When the harness reports the user's actual viewport (an in-app browser's size, a named resolution), add that width to the set: the width that breaks is the one the user sees first. On anything interactive, the round drives every interaction and reachable state (open, closed, hover, focus, error, empty, overlong content) and checks each for overlap and clipping, not only the resting screen. Critique the render against the user's request and the direction contract, fix material gaps, and confirm with one final round; two rounds is the ceiling, and fixes batch between them rather than earning per-tweak screenshots. The ceiling bounds inspection and never closes an open build-phase gate: a gate stays open until it passes or the route it names is taken. On a comp-led build, run `.hermes/skills/impeccable/scripts/impeccable comp-diff --comp <approved comp> --build .impeccable/review/desktop.png --spec .impeccable/build/spec.json --out-dir .impeccable/review/diff/final` and read its region rows and paired crops as the critique: the side-by-side is the view the build thread never has on its own, and a region it scores missing or contradicted is a fix whatever the page looks like from memory. Never judge fidelity from one full-page thumbnail; it hides exactly the failures that matter. On a Persuade surface, verify the mode did its job: a first-time visitor should know what this is, why it matters, and what to do within seconds, in the form's own vocabulary.
A capture is evidence only when it is valid, and you validate before you send. Settle or disable entrance motion first: an element hidden by animation timing reads as a missing element and gets fixed into a regression. Capture full-page shots from the document top. Capture the comp comparison at the comp's own pixel dimensions. Then open every file once and confirm it shows what its name claims: no black or blank regions, no wrong section behind a right filename, no half-loaded state. A malformed capture sent onward costs the whole round; the reviewer answers it with `disposition: recapture` and nothing it reviewed binds.
diff --git a/.hermes/skills/impeccable/reference/overdrive.md b/.hermes/skills/impeccable/reference/overdrive.md
index 89be08b5c..b3dfb708c 100644
--- a/.hermes/skills/impeccable/reference/overdrive.md
+++ b/.hermes/skills/impeccable/reference/overdrive.md
@@ -21,7 +21,7 @@ Skipping this step risks building something embarrassing that needs to be thrown
### Verify in the Browser
-Ambitious effects rarely work on the first try, so verify them in the browser rather than from code: an effect that technically runs can still stutter, ease wrong, or land on the wrong frame. SKILL.md's bounded passes apply: one batched inspection, one fix batch, at most one confirming round. Point them at the effect's timing, easing, and smoothness, then stop and name what still falls short instead of opening another round.
+Ambitious effects rarely work on the first try, so verify them in the browser rather than from code: an effect that technically runs can still stutter, ease wrong, or land on the wrong frame. Aim the checks at timing, easing, and smoothness, batch the fixes between looks, and keep refining until the effect is extraordinary.
---
diff --git a/.kiro/skills/impeccable/SKILL.md b/.kiro/skills/impeccable/SKILL.md
index ddfb3928c..be3846578 100644
--- a/.kiro/skills/impeccable/SKILL.md
+++ b/.kiro/skills/impeccable/SKILL.md
@@ -10,7 +10,7 @@ This skill gives you the tools and permission to make design that earns the name
Core principles:
- Finish the job. No hedging, no shortcuts: the deliverable is complete, except assets the user must provide.
- Scale ambition to the surface's mode (see Modes). Persuade and Experience get full commitment: bold, distinct work with a point of view. On Operate and Read, ambition shows as precision and restraint, and patterns the visitor already knows earn their place.
-- Verify in bounded passes, not a loop, and the ceiling covers the whole cycle: screenshots, defect scans, micro-edits, and rebuilds alike. Build fully, inspect once with a batched round (desktop and mobile together on the web; the shipped device classes on a native platform), fix everything it shows in one batch, confirm with at most one more round, and stop polishing. Open-ended self-QA burns the user's money doing worse what the finish handoffs do better.
+- Verify by looking at the rendered result at the sizes it ships at, and batch the checks: fix what one look shows before looking again, rather than screenshotting after every tweak.
## Setup
diff --git a/.kiro/skills/impeccable/reference/craft-floor.md b/.kiro/skills/impeccable/reference/craft-floor.md
index 9653b9808..64a195eba 100644
--- a/.kiro/skills/impeccable/reference/craft-floor.md
+++ b/.kiro/skills/impeccable/reference/craft-floor.md
@@ -4,7 +4,7 @@ Load this after the direction is settled, and build without announcing the check
## 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.
+Each of these is a check on the built result, not an intention. Run them together in each batched look at the render, not as separate screenshot trips; the checks share one render.
- **Contrast:** body and placeholder text ≥4.5:1, large text ≥3:1. On colored surfaces tint secondary text from that hue or the foreground; never gray.
- **Depth:** shadows carry an offset and a soft blur. A zero-offset colored halo is decoration.
diff --git a/.kiro/skills/impeccable/reference/new-work.md b/.kiro/skills/impeccable/reference/new-work.md
index e5fe06eb1..58d258f63 100644
--- a/.kiro/skills/impeccable/reference/new-work.md
+++ b/.kiro/skills/impeccable/reference/new-work.md
@@ -133,7 +133,7 @@ Preserve semantics, accessibility, performance, responsiveness, project conventi
## 7. Inspect and finish
-Inspect the surface's target sizes in one batched screenshot round: desktop and mobile on the web; on a native platform (`ios` / `android` / `adaptive`), the shipped device classes per OS, captured from the simulator or emulator the way the platform reference's Verifying the build section describes. When the harness reports the user's actual viewport (an in-app browser's size, a named resolution), add that width to the set: the width that breaks is the one the user sees first. Critique the render against the user's request and the direction contract, fix material gaps, and confirm with one final round; two rounds is the ceiling, and fixes batch between them rather than earning per-tweak screenshots. The ceiling bounds inspection and never closes an open build-phase gate: a gate stays open until it passes or the route it names is taken. On a comp-led build, run `.kiro/skills/impeccable/scripts/impeccable comp-diff --comp <approved comp> --build .impeccable/review/desktop.png --spec .impeccable/build/spec.json --out-dir .impeccable/review/diff/final` and read its region rows and paired crops as the critique: the side-by-side is the view the build thread never has on its own, and a region it scores missing or contradicted is a fix whatever the page looks like from memory. Never judge fidelity from one full-page thumbnail; it hides exactly the failures that matter. On a Persuade surface, verify the mode did its job: a first-time visitor should know what this is, why it matters, and what to do within seconds, in the form's own vocabulary.
+Inspect the surface's target sizes in one batched screenshot round: desktop and mobile on the web; on a native platform (`ios` / `android` / `adaptive`), the shipped device classes per OS, captured from the simulator or emulator the way the platform reference's Verifying the build section describes. When the harness reports the user's actual viewport (an in-app browser's size, a named resolution), add that width to the set: the width that breaks is the one the user sees first. On anything interactive, the round drives every interaction and reachable state (open, closed, hover, focus, error, empty, overlong content) and checks each for overlap and clipping, not only the resting screen. Critique the render against the user's request and the direction contract, fix material gaps, and confirm with one final round; two rounds is the ceiling, and fixes batch between them rather than earning per-tweak screenshots. The ceiling bounds inspection and never closes an open build-phase gate: a gate stays open until it passes or the route it names is taken. On a comp-led build, run `.kiro/skills/impeccable/scripts/impeccable comp-diff --comp <approved comp> --build .impeccable/review/desktop.png --spec .impeccable/build/spec.json --out-dir .impeccable/review/diff/final` and read its region rows and paired crops as the critique: the side-by-side is the view the build thread never has on its own, and a region it scores missing or contradicted is a fix whatever the page looks like from memory. Never judge fidelity from one full-page thumbnail; it hides exactly the failures that matter. On a Persuade surface, verify the mode did its job: a first-time visitor should know what this is, why it matters, and what to do within seconds, in the form's own vocabulary.
A capture is evidence only when it is valid, and you validate before you send. Settle or disable entrance motion first: an element hidden by animation timing reads as a missing element and gets fixed into a regression. Capture full-page shots from the document top. Capture the comp comparison at the comp's own pixel dimensions. Then open every file once and confirm it shows what its name claims: no black or blank regions, no wrong section behind a right filename, no half-loaded state. A malformed capture sent onward costs the whole round; the reviewer answers it with `disposition: recapture` and nothing it reviewed binds.
diff --git a/.kiro/skills/impeccable/reference/overdrive.md b/.kiro/skills/impeccable/reference/overdrive.md
index 89be08b5c..b3dfb708c 100644
--- a/.kiro/skills/impeccable/reference/overdrive.md
+++ b/.kiro/skills/impeccable/reference/overdrive.md
@@ -21,7 +21,7 @@ Skipping this step risks building something embarrassing that needs to be thrown
### Verify in the Browser
-Ambitious effects rarely work on the first try, so verify them in the browser rather than from code: an effect that technically runs can still stutter, ease wrong, or land on the wrong frame. SKILL.md's bounded passes apply: one batched inspection, one fix batch, at most one confirming round. Point them at the effect's timing, easing, and smoothness, then stop and name what still falls short instead of opening another round.
+Ambitious effects rarely work on the first try, so verify them in the browser rather than from code: an effect that technically runs can still stutter, ease wrong, or land on the wrong frame. Aim the checks at timing, easing, and smoothness, batch the fixes between looks, and keep refining until the effect is extraordinary.
---
diff --git a/.omp/skills/impeccable/SKILL.md b/.omp/skills/impeccable/SKILL.md
index 188a30517..3f6207dde 100644
--- a/.omp/skills/impeccable/SKILL.md
+++ b/.omp/skills/impeccable/SKILL.md
@@ -13,7 +13,7 @@ This skill gives you the tools and permission to make design that earns the name
Core principles:
- Finish the job. No hedging, no shortcuts: the deliverable is complete, except assets the user must provide.
- Scale ambition to the surface's mode (see Modes). Persuade and Experience get full commitment: bold, distinct work with a point of view. On Operate and Read, ambition shows as precision and restraint, and patterns the visitor already knows earn their place.
-- Verify in bounded passes, not a loop, and the ceiling covers the whole cycle: screenshots, defect scans, micro-edits, and rebuilds alike. Build fully, inspect once with a batched round (desktop and mobile together on the web; the shipped device classes on a native platform), fix everything it shows in one batch, confirm with at most one more round, and stop polishing. Open-ended self-QA burns the user's money doing worse what the finish handoffs do better.
+- Verify by looking at the rendered result at the sizes it ships at, and batch the checks: fix what one look shows before looking again, rather than screenshotting after every tweak.
## Setup
diff --git a/.omp/skills/impeccable/reference/craft-floor.md b/.omp/skills/impeccable/reference/craft-floor.md
index 9653b9808..64a195eba 100644
--- a/.omp/skills/impeccable/reference/craft-floor.md
+++ b/.omp/skills/impeccable/reference/craft-floor.md
@@ -4,7 +4,7 @@ Load this after the direction is settled, and build without announcing the check
## 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.
+Each of these is a check on the built result, not an intention. Run them together in each batched look at the render, not as separate screenshot trips; the checks share one render.
- **Contrast:** body and placeholder text ≥4.5:1, large text ≥3:1. On colored surfaces tint secondary text from that hue or the foreground; never gray.
- **Depth:** shadows carry an offset and a soft blur. A zero-offset colored halo is decoration.
diff --git a/.omp/skills/impeccable/reference/new-work.md b/.omp/skills/impeccable/reference/new-work.md
index 7c9a095a1..cff490d64 100644
--- a/.omp/skills/impeccable/reference/new-work.md
+++ b/.omp/skills/impeccable/reference/new-work.md
@@ -133,7 +133,7 @@ Preserve semantics, accessibility, performance, responsiveness, project conventi
## 7. Inspect and finish
-Inspect the surface's target sizes in one batched screenshot round: desktop and mobile on the web; on a native platform (`ios` / `android` / `adaptive`), the shipped device classes per OS, captured from the simulator or emulator the way the platform reference's Verifying the build section describes. When the harness reports the user's actual viewport (an in-app browser's size, a named resolution), add that width to the set: the width that breaks is the one the user sees first. Critique the render against the user's request and the direction contract, fix material gaps, and confirm with one final round; two rounds is the ceiling, and fixes batch between them rather than earning per-tweak screenshots. The ceiling bounds inspection and never closes an open build-phase gate: a gate stays open until it passes or the route it names is taken. On a comp-led build, run `.omp/skills/impeccable/scripts/impeccable comp-diff --comp <approved comp> --build .impeccable/review/desktop.png --spec .impeccable/build/spec.json --out-dir .impeccable/review/diff/final` and read its region rows and paired crops as the critique: the side-by-side is the view the build thread never has on its own, and a region it scores missing or contradicted is a fix whatever the page looks like from memory. Never judge fidelity from one full-page thumbnail; it hides exactly the failures that matter. On a Persuade surface, verify the mode did its job: a first-time visitor should know what this is, why it matters, and what to do within seconds, in the form's own vocabulary.
+Inspect the surface's target sizes in one batched screenshot round: desktop and mobile on the web; on a native platform (`ios` / `android` / `adaptive`), the shipped device classes per OS, captured from the simulator or emulator the way the platform reference's Verifying the build section describes. When the harness reports the user's actual viewport (an in-app browser's size, a named resolution), add that width to the set: the width that breaks is the one the user sees first. On anything interactive, the round drives every interaction and reachable state (open, closed, hover, focus, error, empty, overlong content) and checks each for overlap and clipping, not only the resting screen. Critique the render against the user's request and the direction contract, fix material gaps, and confirm with one final round; two rounds is the ceiling, and fixes batch between them rather than earning per-tweak screenshots. The ceiling bounds inspection and never closes an open build-phase gate: a gate stays open until it passes or the route it names is taken. On a comp-led build, run `.omp/skills/impeccable/scripts/impeccable comp-diff --comp <approved comp> --build .impeccable/review/desktop.png --spec .impeccable/build/spec.json --out-dir .impeccable/review/diff/final` and read its region rows and paired crops as the critique: the side-by-side is the view the build thread never has on its own, and a region it scores missing or contradicted is a fix whatever the page looks like from memory. Never judge fidelity from one full-page thumbnail; it hides exactly the failures that matter. On a Persuade surface, verify the mode did its job: a first-time visitor should know what this is, why it matters, and what to do within seconds, in the form's own vocabulary.
A capture is evidence only when it is valid, and you validate before you send. Settle or disable entrance motion first: an element hidden by animation timing reads as a missing element and gets fixed into a regression. Capture full-page shots from the document top. Capture the comp comparison at the comp's own pixel dimensions. Then open every file once and confirm it shows what its name claims: no black or blank regions, no wrong section behind a right filename, no half-loaded state. A malformed capture sent onward costs the whole round; the reviewer answers it with `disposition: recapture` and nothing it reviewed binds.
diff --git a/.omp/skills/impeccable/reference/overdrive.md b/.omp/skills/impeccable/reference/overdrive.md
index 89be08b5c..b3dfb708c 100644
--- a/.omp/skills/impeccable/reference/overdrive.md
+++ b/.omp/skills/impeccable/reference/overdrive.md
@@ -21,7 +21,7 @@ Skipping this step risks building something embarrassing that needs to be thrown
### Verify in the Browser
-Ambitious effects rarely work on the first try, so verify them in the browser rather than from code: an effect that technically runs can still stutter, ease wrong, or land on the wrong frame. SKILL.md's bounded passes apply: one batched inspection, one fix batch, at most one confirming round. Point them at the effect's timing, easing, and smoothness, then stop and name what still falls short instead of opening another round.
+Ambitious effects rarely work on the first try, so verify them in the browser rather than from code: an effect that technically runs can still stutter, ease wrong, or land on the wrong frame. Aim the checks at timing, easing, and smoothness, batch the fixes between looks, and keep refining until the effect is extraordinary.
---
diff --git a/.opencode/skills/impeccable/SKILL.md b/.opencode/skills/impeccable/SKILL.md
index ae3470be7..455d5d863 100644
--- a/.opencode/skills/impeccable/SKILL.md
+++ b/.opencode/skills/impeccable/SKILL.md
@@ -15,7 +15,7 @@ This skill gives you the tools and permission to make design that earns the name
Core principles:
- Finish the job. No hedging, no shortcuts: the deliverable is complete, except assets the user must provide.
- Scale ambition to the surface's mode (see Modes). Persuade and Experience get full commitment: bold, distinct work with a point of view. On Operate and Read, ambition shows as precision and restraint, and patterns the visitor already knows earn their place.
-- Verify in bounded passes, not a loop, and the ceiling covers the whole cycle: screenshots, defect scans, micro-edits, and rebuilds alike. Build fully, inspect once with a batched round (desktop and mobile together on the web; the shipped device classes on a native platform), fix everything it shows in one batch, confirm with at most one more round, and stop polishing. Open-ended self-QA burns the user's money doing worse what the finish handoffs do better.
+- Verify by looking at the rendered result at the sizes it ships at, and batch the checks: fix what one look shows before looking again, rather than screenshotting after every tweak.
## Setup
diff --git a/.opencode/skills/impeccable/reference/craft-floor.md b/.opencode/skills/impeccable/reference/craft-floor.md
index 9653b9808..64a195eba 100644
--- a/.opencode/skills/impeccable/reference/craft-floor.md
+++ b/.opencode/skills/impeccable/reference/craft-floor.md
@@ -4,7 +4,7 @@ Load this after the direction is settled, and build without announcing the check
## 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.
+Each of these is a check on the built result, not an intention. Run them together in each batched look at the render, not as separate screenshot trips; the checks share one render.
- **Contrast:** body and placeholder text ≥4.5:1, large text ≥3:1. On colored surfaces tint secondary text from that hue or the foreground; never gray.
- **Depth:** shadows carry an offset and a soft blur. A zero-offset colored halo is decoration.
diff --git a/.opencode/skills/impeccable/reference/new-work.md b/.opencode/skills/impeccable/reference/new-work.md
index abd134668..1c4c97466 100644
--- a/.opencode/skills/impeccable/reference/new-work.md
+++ b/.opencode/skills/impeccable/reference/new-work.md
@@ -133,7 +133,7 @@ Preserve semantics, accessibility, performance, responsiveness, project conventi
## 7. Inspect and finish
-Inspect the surface's target sizes in one batched screenshot round: desktop and mobile on the web; on a native platform (`ios` / `android` / `adaptive`), the shipped device classes per OS, captured from the simulator or emulator the way the platform reference's Verifying the build section describes. When the harness reports the user's actual viewport (an in-app browser's size, a named resolution), add that width to the set: the width that breaks is the one the user sees first. Critique the render against the user's request and the direction contract, fix material gaps, and confirm with one final round; two rounds is the ceiling, and fixes batch between them rather than earning per-tweak screenshots. The ceiling bounds inspection and never closes an open build-phase gate: a gate stays open until it passes or the route it names is taken. On a comp-led build, run `.opencode/skills/impeccable/scripts/impeccable comp-diff --comp <approved comp> --build .impeccable/review/desktop.png --spec .impeccable/build/spec.json --out-dir .impeccable/review/diff/final` and read its region rows and paired crops as the critique: the side-by-side is the view the build thread never has on its own, and a region it scores missing or contradicted is a fix whatever the page looks like from memory. Never judge fidelity from one full-page thumbnail; it hides exactly the failures that matter. On a Persuade surface, verify the mode did its job: a first-time visitor should know what this is, why it matters, and what to do within seconds, in the form's own vocabulary.
+Inspect the surface's target sizes in one batched screenshot round: desktop and mobile on the web; on a native platform (`ios` / `android` / `adaptive`), the shipped device classes per OS, captured from the simulator or emulator the way the platform reference's Verifying the build section describes. When the harness reports the user's actual viewport (an in-app browser's size, a named resolution), add that width to the set: the width that breaks is the one the user sees first. On anything interactive, the round drives every interaction and reachable state (open, closed, hover, focus, error, empty, overlong content) and checks each for overlap and clipping, not only the resting screen. Critique the render against the user's request and the direction contract, fix material gaps, and confirm with one final round; two rounds is the ceiling, and fixes batch between them rather than earning per-tweak screenshots. The ceiling bounds inspection and never closes an open build-phase gate: a gate stays open until it passes or the route it names is taken. On a comp-led build, run `.opencode/skills/impeccable/scripts/impeccable comp-diff --comp <approved comp> --build .impeccable/review/desktop.png --spec .impeccable/build/spec.json --out-dir .impeccable/review/diff/final` and read its region rows and paired crops as the critique: the side-by-side is the view the build thread never has on its own, and a region it scores missing or contradicted is a fix whatever the page looks like from memory. Never judge fidelity from one full-page thumbnail; it hides exactly the failures that matter. On a Persuade surface, verify the mode did its job: a first-time visitor should know what this is, why it matters, and what to do within seconds, in the form's own vocabulary.
A capture is evidence only when it is valid, and you validate before you send. Settle or disable entrance motion first: an element hidden by animation timing reads as a missing element and gets fixed into a regression. Capture full-page shots from the document top. Capture the comp comparison at the comp's own pixel dimensions. Then open every file once and confirm it shows what its name claims: no black or blank regions, no wrong section behind a right filename, no half-loaded state. A malformed capture sent onward costs the whole round; the reviewer answers it with `disposition: recapture` and nothing it reviewed binds.
diff --git a/.opencode/skills/impeccable/reference/overdrive.md b/.opencode/skills/impeccable/reference/overdrive.md
index 1df958279..2880e9370 100644
--- a/.opencode/skills/impeccable/reference/overdrive.md
+++ b/.opencode/skills/impeccable/reference/overdrive.md
@@ -21,7 +21,7 @@ Skipping this step risks building something embarrassing that needs to be thrown
### Verify in the Browser
-Ambitious effects rarely work on the first try, so verify them in the browser rather than from code: an effect that technically runs can still stutter, ease wrong, or land on the wrong frame. SKILL.md's bounded passes apply: one batched inspection, one fix batch, at most one confirming round. Point them at the effect's timing, easing, and smoothness, then stop and name what still falls short instead of opening another round.
+Ambitious effects rarely work on the first try, so verify them in the browser rather than from code: an effect that technically runs can still stutter, ease wrong, or land on the wrong frame. Aim the checks at timing, easing, and smoothness, batch the fixes between looks, and keep refining until the effect is extraordinary.
---
diff --git a/.pi/skills/impeccable/SKILL.md b/.pi/skills/impeccable/SKILL.md
index 904d99fc9..d82480aa1 100644
--- a/.pi/skills/impeccable/SKILL.md
+++ b/.pi/skills/impeccable/SKILL.md
@@ -13,7 +13,7 @@ This skill gives you the tools and permission to make design that earns the name
Core principles:
- Finish the job. No hedging, no shortcuts: the deliverable is complete, except assets the user must provide.
- Scale ambition to the surface's mode (see Modes). Persuade and Experience get full commitment: bold, distinct work with a point of view. On Operate and Read, ambition shows as precision and restraint, and patterns the visitor already knows earn their place.
-- Verify in bounded passes, not a loop, and the ceiling covers the whole cycle: screenshots, defect scans, micro-edits, and rebuilds alike. Build fully, inspect once with a batched round (desktop and mobile together on the web; the shipped device classes on a native platform), fix everything it shows in one batch, confirm with at most one more round, and stop polishing. Open-ended self-QA burns the user's money doing worse what the finish handoffs do better.
+- Verify by looking at the rendered result at the sizes it ships at, and batch the checks: fix what one look shows before looking again, rather than screenshotting after every tweak.
## Setup
diff --git a/.pi/skills/impeccable/reference/craft-floor.md b/.pi/skills/impeccable/reference/craft-floor.md
index 9653b9808..64a195eba 100644
--- a/.pi/skills/impeccable/reference/craft-floor.md
+++ b/.pi/skills/impeccable/reference/craft-floor.md
@@ -4,7 +4,7 @@ Load this after the direction is settled, and build without announcing the check
## 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.
+Each of these is a check on the built result, not an intention. Run them together in each batched look at the render, not as separate screenshot trips; the checks share one render.
- **Contrast:** body and placeholder text ≥4.5:1, large text ≥3:1. On colored surfaces tint secondary text from that hue or the foreground; never gray.
- **Depth:** shadows carry an offset and a soft blur. A zero-offset colored halo is decoration.
diff --git a/.pi/skills/impeccable/reference/new-work.md b/.pi/skills/impeccable/reference/new-work.md
index 68adbc657..0ef27cc6b 100644
--- a/.pi/skills/impeccable/reference/new-work.md
+++ b/.pi/skills/impeccable/reference/new-work.md
@@ -133,7 +133,7 @@ Preserve semantics, accessibility, performance, responsiveness, project conventi
## 7. Inspect and finish
-Inspect the surface's target sizes in one batched screenshot round: desktop and mobile on the web; on a native platform (`ios` / `android` / `adaptive`), the shipped device classes per OS, captured from the simulator or emulator the way the platform reference's Verifying the build section describes. When the harness reports the user's actual viewport (an in-app browser's size, a named resolution), add that width to the set: the width that breaks is the one the user sees first. Critique the render against the user's request and the direction contract, fix material gaps, and confirm with one final round; two rounds is the ceiling, and fixes batch between them rather than earning per-tweak screenshots. The ceiling bounds inspection and never closes an open build-phase gate: a gate stays open until it passes or the route it names is taken. On a comp-led build, run `.pi/skills/impeccable/scripts/impeccable comp-diff --comp <approved comp> --build .impeccable/review/desktop.png --spec .impeccable/build/spec.json --out-dir .impeccable/review/diff/final` and read its region rows and paired crops as the critique: the side-by-side is the view the build thread never has on its own, and a region it scores missing or contradicted is a fix whatever the page looks like from memory. Never judge fidelity from one full-page thumbnail; it hides exactly the failures that matter. On a Persuade surface, verify the mode did its job: a first-time visitor should know what this is, why it matters, and what to do within seconds, in the form's own vocabulary.
+Inspect the surface's target sizes in one batched screenshot round: desktop and mobile on the web; on a native platform (`ios` / `android` / `adaptive`), the shipped device classes per OS, captured from the simulator or emulator the way the platform reference's Verifying the build section describes. When the harness reports the user's actual viewport (an in-app browser's size, a named resolution), add that width to the set: the width that breaks is the one the user sees first. On anything interactive, the round drives every interaction and reachable state (open, closed, hover, focus, error, empty, overlong content) and checks each for overlap and clipping, not only the resting screen. Critique the render against the user's request and the direction contract, fix material gaps, and confirm with one final round; two rounds is the ceiling, and fixes batch between them rather than earning per-tweak screenshots. The ceiling bounds inspection and never closes an open build-phase gate: a gate stays open until it passes or the route it names is taken. On a comp-led build, run `.pi/skills/impeccable/scripts/impeccable comp-diff --comp <approved comp> --build .impeccable/review/desktop.png --spec .impeccable/build/spec.json --out-dir .impeccable/review/diff/final` and read its region rows and paired crops as the critique: the side-by-side is the view the build thread never has on its own, and a region it scores missing or contradicted is a fix whatever the page looks like from memory. Never judge fidelity from one full-page thumbnail; it hides exactly the failures that matter. On a Persuade surface, verify the mode did its job: a first-time visitor should know what this is, why it matters, and what to do within seconds, in the form's own vocabulary.
A capture is evidence only when it is valid, and you validate before you send. Settle or disable entrance motion first: an element hidden by animation timing reads as a missing element and gets fixed into a regression. Capture full-page shots from the document top. Capture the comp comparison at the comp's own pixel dimensions. Then open every file once and confirm it shows what its name claims: no black or blank regions, no wrong section behind a right filename, no half-loaded state. A malformed capture sent onward costs the whole round; the reviewer answers it with `disposition: recapture` and nothing it reviewed binds.
diff --git a/.pi/skills/impeccable/reference/overdrive.md b/.pi/skills/impeccable/reference/overdrive.md
index 89be08b5c..b3dfb708c 100644
--- a/.pi/skills/impeccable/reference/overdrive.md
+++ b/.pi/skills/impeccable/reference/overdrive.md
@@ -21,7 +21,7 @@ Skipping this step risks building something embarrassing that needs to be thrown
### Verify in the Browser
-Ambitious effects rarely work on the first try, so verify them in the browser rather than from code: an effect that technically runs can still stutter, ease wrong, or land on the wrong frame. SKILL.md's bounded passes apply: one batched inspection, one fix batch, at most one confirming round. Point them at the effect's timing, easing, and smoothness, then stop and name what still falls short instead of opening another round.
+Ambitious effects rarely work on the first try, so verify them in the browser rather than from code: an effect that technically runs can still stutter, ease wrong, or land on the wrong frame. Aim the checks at timing, easing, and smoothness, batch the fixes between looks, and keep refining until the effect is extraordinary.
---
diff --git a/.qoder/skills/impeccable/SKILL.md b/.qoder/skills/impeccable/SKILL.md
index 9049ffcd0..36c2c6e0b 100644
--- a/.qoder/skills/impeccable/SKILL.md
+++ b/.qoder/skills/impeccable/SKILL.md
@@ -15,7 +15,7 @@ This skill gives you the tools and permission to make design that earns the name
Core principles:
- Finish the job. No hedging, no shortcuts: the deliverable is complete, except assets the user must provide.
- Scale ambition to the surface's mode (see Modes). Persuade and Experience get full commitment: bold, distinct work with a point of view. On Operate and Read, ambition shows as precision and restraint, and patterns the visitor already knows earn their place.
-- Verify in bounded passes, not a loop, and the ceiling covers the whole cycle: screenshots, defect scans, micro-edits, and rebuilds alike. Build fully, inspect once with a batched round (desktop and mobile together on the web; the shipped device classes on a native platform), fix everything it shows in one batch, confirm with at most one more round, and stop polishing. Open-ended self-QA burns the user's money doing worse what the finish handoffs do better.
+- Verify by looking at the rendered result at the sizes it ships at, and batch the checks: fix what one look shows before looking again, rather than screenshotting after every tweak.
## Setup
diff --git a/.qoder/skills/impeccable/reference/craft-floor.md b/.qoder/skills/impeccable/reference/craft-floor.md
index 9653b9808..64a195eba 100644
--- a/.qoder/skills/impeccable/reference/craft-floor.md
+++ b/.qoder/skills/impeccable/reference/craft-floor.md
@@ -4,7 +4,7 @@ Load this after the direction is settled, and build without announcing the check
## 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.
+Each of these is a check on the built result, not an intention. Run them together in each batched look at the render, not as separate screenshot trips; the checks share one render.
- **Contrast:** body and placeholder text ≥4.5:1, large text ≥3:1. On colored surfaces tint secondary text from that hue or the foreground; never gray.
- **Depth:** shadows carry an offset and a soft blur. A zero-offset colored halo is decoration.
diff --git a/.qoder/skills/impeccable/reference/new-work.md b/.qoder/skills/impeccable/reference/new-work.md
index c025eee9e..be1c399b7 100644
--- a/.qoder/skills/impeccable/reference/new-work.md
+++ b/.qoder/skills/impeccable/reference/new-work.md
@@ -133,7 +133,7 @@ Preserve semantics, accessibility, performance, responsiveness, project conventi
## 7. Inspect and finish
-Inspect the surface's target sizes in one batched screenshot round: desktop and mobile on the web; on a native platform (`ios` / `android` / `adaptive`), the shipped device classes per OS, captured from the simulator or emulator the way the platform reference's Verifying the build section describes. When the harness reports the user's actual viewport (an in-app browser's size, a named resolution), add that width to the set: the width that breaks is the one the user sees first. Critique the render against the user's request and the direction contract, fix material gaps, and confirm with one final round; two rounds is the ceiling, and fixes batch between them rather than earning per-tweak screenshots. The ceiling bounds inspection and never closes an open build-phase gate: a gate stays open until it passes or the route it names is taken. On a comp-led build, run `.qoder/skills/impeccable/scripts/impeccable comp-diff --comp <approved comp> --build .impeccable/review/desktop.png --spec .impeccable/build/spec.json --out-dir .impeccable/review/diff/final` and read its region rows and paired crops as the critique: the side-by-side is the view the build thread never has on its own, and a region it scores missing or contradicted is a fix whatever the page looks like from memory. Never judge fidelity from one full-page thumbnail; it hides exactly the failures that matter. On a Persuade surface, verify the mode did its job: a first-time visitor should know what this is, why it matters, and what to do within seconds, in the form's own vocabulary.
+Inspect the surface's target sizes in one batched screenshot round: desktop and mobile on the web; on a native platform (`ios` / `android` / `adaptive`), the shipped device classes per OS, captured from the simulator or emulator the way the platform reference's Verifying the build section describes. When the harness reports the user's actual viewport (an in-app browser's size, a named resolution), add that width to the set: the width that breaks is the one the user sees first. On anything interactive, the round drives every interaction and reachable state (open, closed, hover, focus, error, empty, overlong content) and checks each for overlap and clipping, not only the resting screen. Critique the render against the user's request and the direction contract, fix material gaps, and confirm with one final round; two rounds is the ceiling, and fixes batch between them rather than earning per-tweak screenshots. The ceiling bounds inspection and never closes an open build-phase gate: a gate stays open until it passes or the route it names is taken. On a comp-led build, run `.qoder/skills/impeccable/scripts/impeccable comp-diff --comp <approved comp> --build .impeccable/review/desktop.png --spec .impeccable/build/spec.json --out-dir .impeccable/review/diff/final` and read its region rows and paired crops as the critique: the side-by-side is the view the build thread never has on its own, and a region it scores missing or contradicted is a fix whatever the page looks like from memory. Never judge fidelity from one full-page thumbnail; it hides exactly the failures that matter. On a Persuade surface, verify the mode did its job: a first-time visitor should know what this is, why it matters, and what to do within seconds, in the form's own vocabulary.
A capture is evidence only when it is valid, and you validate before you send. Settle or disable entrance motion first: an element hidden by animation timing reads as a missing element and gets fixed into a regression. Capture full-page shots from the document top. Capture the comp comparison at the comp's own pixel dimensions. Then open every file once and confirm it shows what its name claims: no black or blank regions, no wrong section behind a right filename, no half-loaded state. A malformed capture sent onward costs the whole round; the reviewer answers it with `disposition: recapture` and nothing it reviewed binds.
diff --git a/.qoder/skills/impeccable/reference/overdrive.md b/.qoder/skills/impeccable/reference/overdrive.md
index 89be08b5c..b3dfb708c 100644
--- a/.qoder/skills/impeccable/reference/overdrive.md
+++ b/.qoder/skills/impeccable/reference/overdrive.md
@@ -21,7 +21,7 @@ Skipping this step risks building something embarrassing that needs to be thrown
### Verify in the Browser
-Ambitious effects rarely work on the first try, so verify them in the browser rather than from code: an effect that technically runs can still stutter, ease wrong, or land on the wrong frame. SKILL.md's bounded passes apply: one batched inspection, one fix batch, at most one confirming round. Point them at the effect's timing, easing, and smoothness, then stop and name what still falls short instead of opening another round.
+Ambitious effects rarely work on the first try, so verify them in the browser rather than from code: an effect that technically runs can still stutter, ease wrong, or land on the wrong frame. Aim the checks at timing, easing, and smoothness, batch the fixes between looks, and keep refining until the effect is extraordinary.
---
diff --git a/.rovodev/skills/impeccable/SKILL.md b/.rovodev/skills/impeccable/SKILL.md
index 88f4fa634..97c1a8c24 100644
--- a/.rovodev/skills/impeccable/SKILL.md
+++ b/.rovodev/skills/impeccable/SKILL.md
@@ -15,7 +15,7 @@ This skill gives you the tools and permission to make design that earns the name
Core principles:
- Finish the job. No hedging, no shortcuts: the deliverable is complete, except assets the user must provide.
- Scale ambition to the surface's mode (see Modes). Persuade and Experience get full commitment: bold, distinct work with a point of view. On Operate and Read, ambition shows as precision and restraint, and patterns the visitor already knows earn their place.
-- Verify in bounded passes, not a loop, and the ceiling covers the whole cycle: screenshots, defect scans, micro-edits, and rebuilds alike. Build fully, inspect once with a batched round (desktop and mobile together on the web; the shipped device classes on a native platform), fix everything it shows in one batch, confirm with at most one more round, and stop polishing. Open-ended self-QA burns the user's money doing worse what the finish handoffs do better.
+- Verify by looking at the rendered result at the sizes it ships at, and batch the checks: fix what one look shows before looking again, rather than screenshotting after every tweak.
## Setup
diff --git a/.rovodev/skills/impeccable/reference/craft-floor.md b/.rovodev/skills/impeccable/reference/craft-floor.md
index 9653b9808..64a195eba 100644
--- a/.rovodev/skills/impeccable/reference/craft-floor.md
+++ b/.rovodev/skills/impeccable/reference/craft-floor.md
@@ -4,7 +4,7 @@ Load this after the direction is settled, and build without announcing the check
## 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.
+Each of these is a check on the built result, not an intention. Run them together in each batched look at the render, not as separate screenshot trips; the checks share one render.
- **Contrast:** body and placeholder text ≥4.5:1, large text ≥3:1. On colored surfaces tint secondary text from that hue or the foreground; never gray.
- **Depth:** shadows carry an offset and a soft blur. A zero-offset colored halo is decoration.
diff --git a/.rovodev/skills/impeccable/reference/new-work.md b/.rovodev/skills/impeccable/reference/new-work.md
index 6e43092b7..223ab9445 100644
--- a/.rovodev/skills/impeccable/reference/new-work.md
+++ b/.rovodev/skills/impeccable/reference/new-work.md
@@ -133,7 +133,7 @@ Preserve semantics, accessibility, performance, responsiveness, project conventi
## 7. Inspect and finish
-Inspect the surface's target sizes in one batched screenshot round: desktop and mobile on the web; on a native platform (`ios` / `android` / `adaptive`), the shipped device classes per OS, captured from the simulator or emulator the way the platform reference's Verifying the build section describes. When the harness reports the user's actual viewport (an in-app browser's size, a named resolution), add that width to the set: the width that breaks is the one the user sees first. Critique the render against the user's request and the direction contract, fix material gaps, and confirm with one final round; two rounds is the ceiling, and fixes batch between them rather than earning per-tweak screenshots. The ceiling bounds inspection and never closes an open build-phase gate: a gate stays open until it passes or the route it names is taken. On a comp-led build, run `.rovodev/skills/impeccable/scripts/impeccable comp-diff --comp <approved comp> --build .impeccable/review/desktop.png --spec .impeccable/build/spec.json --out-dir .impeccable/review/diff/final` and read its region rows and paired crops as the critique: the side-by-side is the view the build thread never has on its own, and a region it scores missing or contradicted is a fix whatever the page looks like from memory. Never judge fidelity from one full-page thumbnail; it hides exactly the failures that matter. On a Persuade surface, verify the mode did its job: a first-time visitor should know what this is, why it matters, and what to do within seconds, in the form's own vocabulary.
+Inspect the surface's target sizes in one batched screenshot round: desktop and mobile on the web; on a native platform (`ios` / `android` / `adaptive`), the shipped device classes per OS, captured from the simulator or emulator the way the platform reference's Verifying the build section describes. When the harness reports the user's actual viewport (an in-app browser's size, a named resolution), add that width to the set: the width that breaks is the one the user sees first. On anything interactive, the round drives every interaction and reachable state (open, closed, hover, focus, error, empty, overlong content) and checks each for overlap and clipping, not only the resting screen. Critique the render against the user's request and the direction contract, fix material gaps, and confirm with one final round; two rounds is the ceiling, and fixes batch between them rather than earning per-tweak screenshots. The ceiling bounds inspection and never closes an open build-phase gate: a gate stays open until it passes or the route it names is taken. On a comp-led build, run `.rovodev/skills/impeccable/scripts/impeccable comp-diff --comp <approved comp> --build .impeccable/review/desktop.png --spec .impeccable/build/spec.json --out-dir .impeccable/review/diff/final` and read its region rows and paired crops as the critique: the side-by-side is the view the build thread never has on its own, and a region it scores missing or contradicted is a fix whatever the page looks like from memory. Never judge fidelity from one full-page thumbnail; it hides exactly the failures that matter. On a Persuade surface, verify the mode did its job: a first-time visitor should know what this is, why it matters, and what to do within seconds, in the form's own vocabulary.
A capture is evidence only when it is valid, and you validate before you send. Settle or disable entrance motion first: an element hidden by animation timing reads as a missing element and gets fixed into a regression. Capture full-page shots from the document top. Capture the comp comparison at the comp's own pixel dimensions. Then open every file once and confirm it shows what its name claims: no black or blank regions, no wrong section behind a right filename, no half-loaded state. A malformed capture sent onward costs the whole round; the reviewer answers it with `disposition: recapture` and nothing it reviewed binds.
diff --git a/.rovodev/skills/impeccable/reference/overdrive.md b/.rovodev/skills/impeccable/reference/overdrive.md
index 89be08b5c..b3dfb708c 100644
--- a/.rovodev/skills/impeccable/reference/overdrive.md
+++ b/.rovodev/skills/impeccable/reference/overdrive.md
@@ -21,7 +21,7 @@ Skipping this step risks building something embarrassing that needs to be thrown
### Verify in the Browser
-Ambitious effects rarely work on the first try, so verify them in the browser rather than from code: an effect that technically runs can still stutter, ease wrong, or land on the wrong frame. SKILL.md's bounded passes apply: one batched inspection, one fix batch, at most one confirming round. Point them at the effect's timing, easing, and smoothness, then stop and name what still falls short instead of opening another round.
+Ambitious effects rarely work on the first try, so verify them in the browser rather than from code: an effect that technically runs can still stutter, ease wrong, or land on the wrong frame. Aim the checks at timing, easing, and smoothness, batch the fixes between looks, and keep refining until the effect is extraordinary.
---
diff --git a/.trae-cn/skills/impeccable/SKILL.md b/.trae-cn/skills/impeccable/SKILL.md
index db7039716..3cabbe05e 100644
--- a/.trae-cn/skills/impeccable/SKILL.md
+++ b/.trae-cn/skills/impeccable/SKILL.md
@@ -12,7 +12,7 @@ This skill gives you the tools and permission to make design that earns the name
Core principles:
- Finish the job. No hedging, no shortcuts: the deliverable is complete, except assets the user must provide.
- Scale ambition to the surface's mode (see Modes). Persuade and Experience get full commitment: bold, distinct work with a point of view. On Operate and Read, ambition shows as precision and restraint, and patterns the visitor already knows earn their place.
-- Verify in bounded passes, not a loop, and the ceiling covers the whole cycle: screenshots, defect scans, micro-edits, and rebuilds alike. Build fully, inspect once with a batched round (desktop and mobile together on the web; the shipped device classes on a native platform), fix everything it shows in one batch, confirm with at most one more round, and stop polishing. Open-ended self-QA burns the user's money doing worse what the finish handoffs do better.
+- Verify by looking at the rendered result at the sizes it ships at, and batch the checks: fix what one look shows before looking again, rather than screenshotting after every tweak.
## Setup
diff --git a/.trae-cn/skills/impeccable/reference/craft-floor.md b/.trae-cn/skills/impeccable/reference/craft-floor.md
index 9653b9808..64a195eba 100644
--- a/.trae-cn/skills/impeccable/reference/craft-floor.md
+++ b/.trae-cn/skills/impeccable/reference/craft-floor.md
@@ -4,7 +4,7 @@ Load this after the direction is settled, and build without announcing the check
## 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.
+Each of these is a check on the built result, not an intention. Run them together in each batched look at the render, not as separate screenshot trips; the checks share one render.
- **Contrast:** body and placeholder text ≥4.5:1, large text ≥3:1. On colored surfaces tint secondary text from that hue or the foreground; never gray.
- **Depth:** shadows carry an offset and a soft blur. A zero-offset colored halo is decoration.
diff --git a/.trae-cn/skills/impeccable/reference/new-work.md b/.trae-cn/skills/impeccable/reference/new-work.md
index 7e9d3ef60..43b3b6b3d 100644
--- a/.trae-cn/skills/impeccable/reference/new-work.md
+++ b/.trae-cn/skills/impeccable/reference/new-work.md
@@ -133,7 +133,7 @@ Preserve semantics, accessibility, performance, responsiveness, project conventi
## 7. Inspect and finish
-Inspect the surface's target sizes in one batched screenshot round: desktop and mobile on the web; on a native platform (`ios` / `android` / `adaptive`), the shipped device classes per OS, captured from the simulator or emulator the way the platform reference's Verifying the build section describes. When the harness reports the user's actual viewport (an in-app browser's size, a named resolution), add that width to the set: the width that breaks is the one the user sees first. Critique the render against the user's request and the direction contract, fix material gaps, and confirm with one final round; two rounds is the ceiling, and fixes batch between them rather than earning per-tweak screenshots. The ceiling bounds inspection and never closes an open build-phase gate: a gate stays open until it passes or the route it names is taken. On a comp-led build, run `.trae-cn/skills/impeccable/scripts/impeccable comp-diff --comp <approved comp> --build .impeccable/review/desktop.png --spec .impeccable/build/spec.json --out-dir .impeccable/review/diff/final` and read its region rows and paired crops as the critique: the side-by-side is the view the build thread never has on its own, and a region it scores missing or contradicted is a fix whatever the page looks like from memory. Never judge fidelity from one full-page thumbnail; it hides exactly the failures that matter. On a Persuade surface, verify the mode did its job: a first-time visitor should know what this is, why it matters, and what to do within seconds, in the form's own vocabulary.
+Inspect the surface's target sizes in one batched screenshot round: desktop and mobile on the web; on a native platform (`ios` / `android` / `adaptive`), the shipped device classes per OS, captured from the simulator or emulator the way the platform reference's Verifying the build section describes. When the harness reports the user's actual viewport (an in-app browser's size, a named resolution), add that width to the set: the width that breaks is the one the user sees first. On anything interactive, the round drives every interaction and reachable state (open, closed, hover, focus, error, empty, overlong content) and checks each for overlap and clipping, not only the resting screen. Critique the render against the user's request and the direction contract, fix material gaps, and confirm with one final round; two rounds is the ceiling, and fixes batch between them rather than earning per-tweak screenshots. The ceiling bounds inspection and never closes an open build-phase gate: a gate stays open until it passes or the route it names is taken. On a comp-led build, run `.trae-cn/skills/impeccable/scripts/impeccable comp-diff --comp <approved comp> --build .impeccable/review/desktop.png --spec .impeccable/build/spec.json --out-dir .impeccable/review/diff/final` and read its region rows and paired crops as the critique: the side-by-side is the view the build thread never has on its own, and a region it scores missing or contradicted is a fix whatever the page looks like from memory. Never judge fidelity from one full-page thumbnail; it hides exactly the failures that matter. On a Persuade surface, verify the mode did its job: a first-time visitor should know what this is, why it matters, and what to do within seconds, in the form's own vocabulary.
A capture is evidence only when it is valid, and you validate before you send. Settle or disable entrance motion first: an element hidden by animation timing reads as a missing element and gets fixed into a regression. Capture full-page shots from the document top. Capture the comp comparison at the comp's own pixel dimensions. Then open every file once and confirm it shows what its name claims: no black or blank regions, no wrong section behind a right filename, no half-loaded state. A malformed capture sent onward costs the whole round; the reviewer answers it with `disposition: recapture` and nothing it reviewed binds.
diff --git a/.trae-cn/skills/impeccable/reference/overdrive.md b/.trae-cn/skills/impeccable/reference/overdrive.md
index 89be08b5c..b3dfb708c 100644
--- a/.trae-cn/skills/impeccable/reference/overdrive.md
+++ b/.trae-cn/skills/impeccable/reference/overdrive.md
@@ -21,7 +21,7 @@ Skipping this step risks building something embarrassing that needs to be thrown
### Verify in the Browser
-Ambitious effects rarely work on the first try, so verify them in the browser rather than from code: an effect that technically runs can still stutter, ease wrong, or land on the wrong frame. SKILL.md's bounded passes apply: one batched inspection, one fix batch, at most one confirming round. Point them at the effect's timing, easing, and smoothness, then stop and name what still falls short instead of opening another round.
+Ambitious effects rarely work on the first try, so verify them in the browser rather than from code: an effect that technically runs can still stutter, ease wrong, or land on the wrong frame. Aim the checks at timing, easing, and smoothness, batch the fixes between looks, and keep refining until the effect is extraordinary.
---
diff --git a/.trae/skills/impeccable/SKILL.md b/.trae/skills/impeccable/SKILL.md
index a1948f30d..510be32b1 100644
--- a/.trae/skills/impeccable/SKILL.md
+++ b/.trae/skills/impeccable/SKILL.md
@@ -12,7 +12,7 @@ This skill gives you the tools and permission to make design that earns the name
Core principles:
- Finish the job. No hedging, no shortcuts: the deliverable is complete, except assets the user must provide.
- Scale ambition to the surface's mode (see Modes). Persuade and Experience get full commitment: bold, distinct work with a point of view. On Operate and Read, ambition shows as precision and restraint, and patterns the visitor already knows earn their place.
-- Verify in bounded passes, not a loop, and the ceiling covers the whole cycle: screenshots, defect scans, micro-edits, and rebuilds alike. Build fully, inspect once with a batched round (desktop and mobile together on the web; the shipped device classes on a native platform), fix everything it shows in one batch, confirm with at most one more round, and stop polishing. Open-ended self-QA burns the user's money doing worse what the finish handoffs do better.
+- Verify by looking at the rendered result at the sizes it ships at, and batch the checks: fix what one look shows before looking again, rather than screenshotting after every tweak.
## Setup
diff --git a/.trae/skills/impeccable/reference/craft-floor.md b/.trae/skills/impeccable/reference/craft-floor.md
index 9653b9808..64a195eba 100644
--- a/.trae/skills/impeccable/reference/craft-floor.md
+++ b/.trae/skills/impeccable/reference/craft-floor.md
@@ -4,7 +4,7 @@ Load this after the direction is settled, and build without announcing the check
## 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.
+Each of these is a check on the built result, not an intention. Run them together in each batched look at the render, not as separate screenshot trips; the checks share one render.
- **Contrast:** body and placeholder text ≥4.5:1, large text ≥3:1. On colored surfaces tint secondary text from that hue or the foreground; never gray.
- **Depth:** shadows carry an offset and a soft blur. A zero-offset colored halo is decoration.
diff --git a/.trae/skills/impeccable/reference/new-work.md b/.trae/skills/impeccable/reference/new-work.md
index c8e1db84c..0ce3dc0a5 100644
--- a/.trae/skills/impeccable/reference/new-work.md
+++ b/.trae/skills/impeccable/reference/new-work.md
@@ -133,7 +133,7 @@ Preserve semantics, accessibility, performance, responsiveness, project conventi
## 7. Inspect and finish
-Inspect the surface's target sizes in one batched screenshot round: desktop and mobile on the web; on a native platform (`ios` / `android` / `adaptive`), the shipped device classes per OS, captured from the simulator or emulator the way the platform reference's Verifying the build section describes. When the harness reports the user's actual viewport (an in-app browser's size, a named resolution), add that width to the set: the width that breaks is the one the user sees first. Critique the render against the user's request and the direction contract, fix material gaps, and confirm with one final round; two rounds is the ceiling, and fixes batch between them rather than earning per-tweak screenshots. The ceiling bounds inspection and never closes an open build-phase gate: a gate stays open until it passes or the route it names is taken. On a comp-led build, run `.trae/skills/impeccable/scripts/impeccable comp-diff --comp <approved comp> --build .impeccable/review/desktop.png --spec .impeccable/build/spec.json --out-dir .impeccable/review/diff/final` and read its region rows and paired crops as the critique: the side-by-side is the view the build thread never has on its own, and a region it scores missing or contradicted is a fix whatever the page looks like from memory. Never judge fidelity from one full-page thumbnail; it hides exactly the failures that matter. On a Persuade surface, verify the mode did its job: a first-time visitor should know what this is, why it matters, and what to do within seconds, in the form's own vocabulary.
+Inspect the surface's target sizes in one batched screenshot round: desktop and mobile on the web; on a native platform (`ios` / `android` / `adaptive`), the shipped device classes per OS, captured from the simulator or emulator the way the platform reference's Verifying the build section describes. When the harness reports the user's actual viewport (an in-app browser's size, a named resolution), add that width to the set: the width that breaks is the one the user sees first. On anything interactive, the round drives every interaction and reachable state (open, closed, hover, focus, error, empty, overlong content) and checks each for overlap and clipping, not only the resting screen. Critique the render against the user's request and the direction contract, fix material gaps, and confirm with one final round; two rounds is the ceiling, and fixes batch between them rather than earning per-tweak screenshots. The ceiling bounds inspection and never closes an open build-phase gate: a gate stays open until it passes or the route it names is taken. On a comp-led build, run `.trae/skills/impeccable/scripts/impeccable comp-diff --comp <approved comp> --build .impeccable/review/desktop.png --spec .impeccable/build/spec.json --out-dir .impeccable/review/diff/final` and read its region rows and paired crops as the critique: the side-by-side is the view the build thread never has on its own, and a region it scores missing or contradicted is a fix whatever the page looks like from memory. Never judge fidelity from one full-page thumbnail; it hides exactly the failures that matter. On a Persuade surface, verify the mode did its job: a first-time visitor should know what this is, why it matters, and what to do within seconds, in the form's own vocabulary.
A capture is evidence only when it is valid, and you validate before you send. Settle or disable entrance motion first: an element hidden by animation timing reads as a missing element and gets fixed into a regression. Capture full-page shots from the document top. Capture the comp comparison at the comp's own pixel dimensions. Then open every file once and confirm it shows what its name claims: no black or blank regions, no wrong section behind a right filename, no half-loaded state. A malformed capture sent onward costs the whole round; the reviewer answers it with `disposition: recapture` and nothing it reviewed binds.
diff --git a/.trae/skills/impeccable/reference/overdrive.md b/.trae/skills/impeccable/reference/overdrive.md
index 89be08b5c..b3dfb708c 100644
--- a/.trae/skills/impeccable/reference/overdrive.md
+++ b/.trae/skills/impeccable/reference/overdrive.md
@@ -21,7 +21,7 @@ Skipping this step risks building something embarrassing that needs to be thrown
### Verify in the Browser
-Ambitious effects rarely work on the first try, so verify them in the browser rather than from code: an effect that technically runs can still stutter, ease wrong, or land on the wrong frame. SKILL.md's bounded passes apply: one batched inspection, one fix batch, at most one confirming round. Point them at the effect's timing, easing, and smoothness, then stop and name what still falls short instead of opening another round.
+Ambitious effects rarely work on the first try, so verify them in the browser rather than from code: an effect that technically runs can still stutter, ease wrong, or land on the wrong frame. Aim the checks at timing, easing, and smoothness, batch the fixes between looks, and keep refining until the effect is extraordinary.
---
diff --git a/.veto/skills/impeccable/SKILL.md b/.veto/skills/impeccable/SKILL.md
index 54f86f54e..434c63bb0 100644
--- a/.veto/skills/impeccable/SKILL.md
+++ b/.veto/skills/impeccable/SKILL.md
@@ -10,7 +10,7 @@ This skill gives you the tools and permission to make design that earns the name
Core principles:
- Finish the job. No hedging, no shortcuts: the deliverable is complete, except assets the user must provide.
- Scale ambition to the surface's mode (see Modes). Persuade and Experience get full commitment: bold, distinct work with a point of view. On Operate and Read, ambition shows as precision and restraint, and patterns the visitor already knows earn their place.
-- Verify in bounded passes, not a loop, and the ceiling covers the whole cycle: screenshots, defect scans, micro-edits, and rebuilds alike. Build fully, inspect once with a batched round (desktop and mobile together on the web; the shipped device classes on a native platform), fix everything it shows in one batch, confirm with at most one more round, and stop polishing. Open-ended self-QA burns the user's money doing worse what the finish handoffs do better.
+- Verify by looking at the rendered result at the sizes it ships at, and batch the checks: fix what one look shows before looking again, rather than screenshotting after every tweak.
## Setup
diff --git a/.veto/skills/impeccable/reference/craft-floor.md b/.veto/skills/impeccable/reference/craft-floor.md
index 9653b9808..64a195eba 100644
--- a/.veto/skills/impeccable/reference/craft-floor.md
+++ b/.veto/skills/impeccable/reference/craft-floor.md
@@ -4,7 +4,7 @@ Load this after the direction is settled, and build without announcing the check
## 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.
+Each of these is a check on the built result, not an intention. Run them together in each batched look at the render, not as separate screenshot trips; the checks share one render.
- **Contrast:** body and placeholder text ≥4.5:1, large text ≥3:1. On colored surfaces tint secondary text from that hue or the foreground; never gray.
- **Depth:** shadows carry an offset and a soft blur. A zero-offset colored halo is decoration.
diff --git a/.veto/skills/impeccable/reference/new-work.md b/.veto/skills/impeccable/reference/new-work.md
index 480ac73c1..fc4cf303c 100644
--- a/.veto/skills/impeccable/reference/new-work.md
+++ b/.veto/skills/impeccable/reference/new-work.md
@@ -133,7 +133,7 @@ Preserve semantics, accessibility, performance, responsiveness, project conventi
## 7. Inspect and finish
-Inspect the surface's target sizes in one batched screenshot round: desktop and mobile on the web; on a native platform (`ios` / `android` / `adaptive`), the shipped device classes per OS, captured from the simulator or emulator the way the platform reference's Verifying the build section describes. When the harness reports the user's actual viewport (an in-app browser's size, a named resolution), add that width to the set: the width that breaks is the one the user sees first. Critique the render against the user's request and the direction contract, fix material gaps, and confirm with one final round; two rounds is the ceiling, and fixes batch between them rather than earning per-tweak screenshots. The ceiling bounds inspection and never closes an open build-phase gate: a gate stays open until it passes or the route it names is taken. On a comp-led build, run `.veto/skills/impeccable/scripts/impeccable comp-diff --comp <approved comp> --build .impeccable/review/desktop.png --spec .impeccable/build/spec.json --out-dir .impeccable/review/diff/final` and read its region rows and paired crops as the critique: the side-by-side is the view the build thread never has on its own, and a region it scores missing or contradicted is a fix whatever the page looks like from memory. Never judge fidelity from one full-page thumbnail; it hides exactly the failures that matter. On a Persuade surface, verify the mode did its job: a first-time visitor should know what this is, why it matters, and what to do within seconds, in the form's own vocabulary.
+Inspect the surface's target sizes in one batched screenshot round: desktop and mobile on the web; on a native platform (`ios` / `android` / `adaptive`), the shipped device classes per OS, captured from the simulator or emulator the way the platform reference's Verifying the build section describes. When the harness reports the user's actual viewport (an in-app browser's size, a named resolution), add that width to the set: the width that breaks is the one the user sees first. On anything interactive, the round drives every interaction and reachable state (open, closed, hover, focus, error, empty, overlong content) and checks each for overlap and clipping, not only the resting screen. Critique the render against the user's request and the direction contract, fix material gaps, and confirm with one final round; two rounds is the ceiling, and fixes batch between them rather than earning per-tweak screenshots. The ceiling bounds inspection and never closes an open build-phase gate: a gate stays open until it passes or the route it names is taken. On a comp-led build, run `.veto/skills/impeccable/scripts/impeccable comp-diff --comp <approved comp> --build .impeccable/review/desktop.png --spec .impeccable/build/spec.json --out-dir .impeccable/review/diff/final` and read its region rows and paired crops as the critique: the side-by-side is the view the build thread never has on its own, and a region it scores missing or contradicted is a fix whatever the page looks like from memory. Never judge fidelity from one full-page thumbnail; it hides exactly the failures that matter. On a Persuade surface, verify the mode did its job: a first-time visitor should know what this is, why it matters, and what to do within seconds, in the form's own vocabulary.
A capture is evidence only when it is valid, and you validate before you send. Settle or disable entrance motion first: an element hidden by animation timing reads as a missing element and gets fixed into a regression. Capture full-page shots from the document top. Capture the comp comparison at the comp's own pixel dimensions. Then open every file once and confirm it shows what its name claims: no black or blank regions, no wrong section behind a right filename, no half-loaded state. A malformed capture sent onward costs the whole round; the reviewer answers it with `disposition: recapture` and nothing it reviewed binds.
diff --git a/.veto/skills/impeccable/reference/overdrive.md b/.veto/skills/impeccable/reference/overdrive.md
index 89be08b5c..b3dfb708c 100644
--- a/.veto/skills/impeccable/reference/overdrive.md
+++ b/.veto/skills/impeccable/reference/overdrive.md
@@ -21,7 +21,7 @@ Skipping this step risks building something embarrassing that needs to be thrown
### Verify in the Browser
-Ambitious effects rarely work on the first try, so verify them in the browser rather than from code: an effect that technically runs can still stutter, ease wrong, or land on the wrong frame. SKILL.md's bounded passes apply: one batched inspection, one fix batch, at most one confirming round. Point them at the effect's timing, easing, and smoothness, then stop and name what still falls short instead of opening another round.
+Ambitious effects rarely work on the first try, so verify them in the browser rather than from code: an effect that technically runs can still stutter, ease wrong, or land on the wrong frame. Aim the checks at timing, easing, and smoothness, batch the fixes between looks, and keep refining until the effect is extraordinary.
---
diff --git a/.vibe/skills/impeccable/SKILL.md b/.vibe/skills/impeccable/SKILL.md
index fff597e73..593cd63e1 100644
--- a/.vibe/skills/impeccable/SKILL.md
+++ b/.vibe/skills/impeccable/SKILL.md
@@ -14,7 +14,7 @@ This skill gives you the tools and permission to make design that earns the name
Core principles:
- Finish the job. No hedging, no shortcuts: the deliverable is complete, except assets the user must provide.
- Scale ambition to the surface's mode (see Modes). Persuade and Experience get full commitment: bold, distinct work with a point of view. On Operate and Read, ambition shows as precision and restraint, and patterns the visitor already knows earn their place.
-- Verify in bounded passes, not a loop, and the ceiling covers the whole cycle: screenshots, defect scans, micro-edits, and rebuilds alike. Build fully, inspect once with a batched round (desktop and mobile together on the web; the shipped device classes on a native platform), fix everything it shows in one batch, confirm with at most one more round, and stop polishing. Open-ended self-QA burns the user's money doing worse what the finish handoffs do better.
+- Verify by looking at the rendered result at the sizes it ships at, and batch the checks: fix what one look shows before looking again, rather than screenshotting after every tweak.
## Setup
diff --git a/.vibe/skills/impeccable/reference/craft-floor.md b/.vibe/skills/impeccable/reference/craft-floor.md
index 9653b9808..64a195eba 100644
--- a/.vibe/skills/impeccable/reference/craft-floor.md
+++ b/.vibe/skills/impeccable/reference/craft-floor.md
@@ -4,7 +4,7 @@ Load this after the direction is settled, and build without announcing the check
## 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.
+Each of these is a check on the built result, not an intention. Run them together in each batched look at the render, not as separate screenshot trips; the checks share one render.
- **Contrast:** body and placeholder text ≥4.5:1, large text ≥3:1. On colored surfaces tint secondary text from that hue or the foreground; never gray.
- **Depth:** shadows carry an offset and a soft blur. A zero-offset colored halo is decoration.
diff --git a/.vibe/skills/impeccable/reference/new-work.md b/.vibe/skills/impeccable/reference/new-work.md
index c71cf57b0..fb9c61610 100644
--- a/.vibe/skills/impeccable/reference/new-work.md
+++ b/.vibe/skills/impeccable/reference/new-work.md
@@ -133,7 +133,7 @@ Preserve semantics, accessibility, performance, responsiveness, project conventi
## 7. Inspect and finish
-Inspect the surface's target sizes in one batched screenshot round: desktop and mobile on the web; on a native platform (`ios` / `android` / `adaptive`), the shipped device classes per OS, captured from the simulator or emulator the way the platform reference's Verifying the build section describes. When the harness reports the user's actual viewport (an in-app browser's size, a named resolution), add that width to the set: the width that breaks is the one the user sees first. Critique the render against the user's request and the direction contract, fix material gaps, and confirm with one final round; two rounds is the ceiling, and fixes batch between them rather than earning per-tweak screenshots. The ceiling bounds inspection and never closes an open build-phase gate: a gate stays open until it passes or the route it names is taken. On a comp-led build, run `.vibe/skills/impeccable/scripts/impeccable comp-diff --comp <approved comp> --build .impeccable/review/desktop.png --spec .impeccable/build/spec.json --out-dir .impeccable/review/diff/final` and read its region rows and paired crops as the critique: the side-by-side is the view the build thread never has on its own, and a region it scores missing or contradicted is a fix whatever the page looks like from memory. Never judge fidelity from one full-page thumbnail; it hides exactly the failures that matter. On a Persuade surface, verify the mode did its job: a first-time visitor should know what this is, why it matters, and what to do within seconds, in the form's own vocabulary.
+Inspect the surface's target sizes in one batched screenshot round: desktop and mobile on the web; on a native platform (`ios` / `android` / `adaptive`), the shipped device classes per OS, captured from the simulator or emulator the way the platform reference's Verifying the build section describes. When the harness reports the user's actual viewport (an in-app browser's size, a named resolution), add that width to the set: the width that breaks is the one the user sees first. On anything interactive, the round drives every interaction and reachable state (open, closed, hover, focus, error, empty, overlong content) and checks each for overlap and clipping, not only the resting screen. Critique the render against the user's request and the direction contract, fix material gaps, and confirm with one final round; two rounds is the ceiling, and fixes batch between them rather than earning per-tweak screenshots. The ceiling bounds inspection and never closes an open build-phase gate: a gate stays open until it passes or the route it names is taken. On a comp-led build, run `.vibe/skills/impeccable/scripts/impeccable comp-diff --comp <approved comp> --build .impeccable/review/desktop.png --spec .impeccable/build/spec.json --out-dir .impeccable/review/diff/final` and read its region rows and paired crops as the critique: the side-by-side is the view the build thread never has on its own, and a region it scores missing or contradicted is a fix whatever the page looks like from memory. Never judge fidelity from one full-page thumbnail; it hides exactly the failures that matter. On a Persuade surface, verify the mode did its job: a first-time visitor should know what this is, why it matters, and what to do within seconds, in the form's own vocabulary.
A capture is evidence only when it is valid, and you validate before you send. Settle or disable entrance motion first: an element hidden by animation timing reads as a missing element and gets fixed into a regression. Capture full-page shots from the document top. Capture the comp comparison at the comp's own pixel dimensions. Then open every file once and confirm it shows what its name claims: no black or blank regions, no wrong section behind a right filename, no half-loaded state. A malformed capture sent onward costs the whole round; the reviewer answers it with `disposition: recapture` and nothing it reviewed binds.
diff --git a/.vibe/skills/impeccable/reference/overdrive.md b/.vibe/skills/impeccable/reference/overdrive.md
index 89be08b5c..b3dfb708c 100644
--- a/.vibe/skills/impeccable/reference/overdrive.md
+++ b/.vibe/skills/impeccable/reference/overdrive.md
@@ -21,7 +21,7 @@ Skipping this step risks building something embarrassing that needs to be thrown
### Verify in the Browser
-Ambitious effects rarely work on the first try, so verify them in the browser rather than from code: an effect that technically runs can still stutter, ease wrong, or land on the wrong frame. SKILL.md's bounded passes apply: one batched inspection, one fix batch, at most one confirming round. Point them at the effect's timing, easing, and smoothness, then stop and name what still falls short instead of opening another round.
+Ambitious effects rarely work on the first try, so verify them in the browser rather than from code: an effect that technically runs can still stutter, ease wrong, or land on the wrong frame. Aim the checks at timing, easing, and smoothness, batch the fixes between looks, and keep refining until the effect is extraordinary.
---
diff --git a/cursor-plugin/skills/impeccable/SKILL.md b/cursor-plugin/skills/impeccable/SKILL.md
index 1f4fcdeca..8402ca75f 100644
--- a/cursor-plugin/skills/impeccable/SKILL.md
+++ b/cursor-plugin/skills/impeccable/SKILL.md
@@ -10,7 +10,7 @@ This skill gives you the tools and permission to make design that earns the name
Core principles:
- Finish the job. No hedging, no shortcuts: the deliverable is complete, except assets the user must provide.
- Scale ambition to the surface's mode (see Modes). Persuade and Experience get full commitment: bold, distinct work with a point of view. On Operate and Read, ambition shows as precision and restraint, and patterns the visitor already knows earn their place.
-- Verify in bounded passes, not a loop, and the ceiling covers the whole cycle: screenshots, defect scans, micro-edits, and rebuilds alike. Build fully, inspect once with a batched round (desktop and mobile together on the web; the shipped device classes on a native platform), fix everything it shows in one batch, confirm with at most one more round, and stop polishing. Open-ended self-QA burns the user's money doing worse what the finish handoffs do better.
+- Verify by looking at the rendered result at the sizes it ships at, and batch the checks: fix what one look shows before looking again, rather than screenshotting after every tweak.
## Setup
diff --git a/cursor-plugin/skills/impeccable/reference/craft-floor.md b/cursor-plugin/skills/impeccable/reference/craft-floor.md
index 9653b9808..64a195eba 100644
--- a/cursor-plugin/skills/impeccable/reference/craft-floor.md
+++ b/cursor-plugin/skills/impeccable/reference/craft-floor.md
@@ -4,7 +4,7 @@ Load this after the direction is settled, and build without announcing the check
## 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.
+Each of these is a check on the built result, not an intention. Run them together in each batched look at the render, not as separate screenshot trips; the checks share one render.
- **Contrast:** body and placeholder text ≥4.5:1, large text ≥3:1. On colored surfaces tint secondary text from that hue or the foreground; never gray.
- **Depth:** shadows carry an offset and a soft blur. A zero-offset colored halo is decoration.
diff --git a/cursor-plugin/skills/impeccable/reference/new-work.md b/cursor-plugin/skills/impeccable/reference/new-work.md
index 8917c4f55..4127dd4a4 100644
--- a/cursor-plugin/skills/impeccable/reference/new-work.md
+++ b/cursor-plugin/skills/impeccable/reference/new-work.md
@@ -133,7 +133,7 @@ Preserve semantics, accessibility, performance, responsiveness, project conventi
## 7. Inspect and finish
-Inspect the surface's target sizes in one batched screenshot round: desktop and mobile on the web; on a native platform (`ios` / `android` / `adaptive`), the shipped device classes per OS, captured from the simulator or emulator the way the platform reference's Verifying the build section describes. When the harness reports the user's actual viewport (an in-app browser's size, a named resolution), add that width to the set: the width that breaks is the one the user sees first. Critique the render against the user's request and the direction contract, fix material gaps, and confirm with one final round; two rounds is the ceiling, and fixes batch between them rather than earning per-tweak screenshots. The ceiling bounds inspection and never closes an open build-phase gate: a gate stays open until it passes or the route it names is taken. On a comp-led build, run `"<skill-base-dir>/scripts/impeccable" comp-diff --comp <approved comp> --build .impeccable/review/desktop.png --spec .impeccable/build/spec.json --out-dir .impeccable/review/diff/final` and read its region rows and paired crops as the critique: the side-by-side is the view the build thread never has on its own, and a region it scores missing or contradicted is a fix whatever the page looks like from memory. Never judge fidelity from one full-page thumbnail; it hides exactly the failures that matter. On a Persuade surface, verify the mode did its job: a first-time visitor should know what this is, why it matters, and what to do within seconds, in the form's own vocabulary.
+Inspect the surface's target sizes in one batched screenshot round: desktop and mobile on the web; on a native platform (`ios` / `android` / `adaptive`), the shipped device classes per OS, captured from the simulator or emulator the way the platform reference's Verifying the build section describes. When the harness reports the user's actual viewport (an in-app browser's size, a named resolution), add that width to the set: the width that breaks is the one the user sees first. On anything interactive, the round drives every interaction and reachable state (open, closed, hover, focus, error, empty, overlong content) and checks each for overlap and clipping, not only the resting screen. Critique the render against the user's request and the direction contract, fix material gaps, and confirm with one final round; two rounds is the ceiling, and fixes batch between them rather than earning per-tweak screenshots. The ceiling bounds inspection and never closes an open build-phase gate: a gate stays open until it passes or the route it names is taken. On a comp-led build, run `"<skill-base-dir>/scripts/impeccable" comp-diff --comp <approved comp> --build .impeccable/review/desktop.png --spec .impeccable/build/spec.json --out-dir .impeccable/review/diff/final` and read its region rows and paired crops as the critique: the side-by-side is the view the build thread never has on its own, and a region it scores missing or contradicted is a fix whatever the page looks like from memory. Never judge fidelity from one full-page thumbnail; it hides exactly the failures that matter. On a Persuade surface, verify the mode did its job: a first-time visitor should know what this is, why it matters, and what to do within seconds, in the form's own vocabulary.
A capture is evidence only when it is valid, and you validate before you send. Settle or disable entrance motion first: an element hidden by animation timing reads as a missing element and gets fixed into a regression. Capture full-page shots from the document top. Capture the comp comparison at the comp's own pixel dimensions. Then open every file once and confirm it shows what its name claims: no black or blank regions, no wrong section behind a right filename, no half-loaded state. A malformed capture sent onward costs the whole round; the reviewer answers it with `disposition: recapture` and nothing it reviewed binds.
diff --git a/cursor-plugin/skills/impeccable/reference/overdrive.md b/cursor-plugin/skills/impeccable/reference/overdrive.md
index 89be08b5c..b3dfb708c 100644
--- a/cursor-plugin/skills/impeccable/reference/overdrive.md
+++ b/cursor-plugin/skills/impeccable/reference/overdrive.md
@@ -21,7 +21,7 @@ Skipping this step risks building something embarrassing that needs to be thrown
### Verify in the Browser
-Ambitious effects rarely work on the first try, so verify them in the browser rather than from code: an effect that technically runs can still stutter, ease wrong, or land on the wrong frame. SKILL.md's bounded passes apply: one batched inspection, one fix batch, at most one confirming round. Point them at the effect's timing, easing, and smoothness, then stop and name what still falls short instead of opening another round.
+Ambitious effects rarely work on the first try, so verify them in the browser rather than from code: an effect that technically runs can still stutter, ease wrong, or land on the wrong frame. Aim the checks at timing, easing, and smoothness, batch the fixes between looks, and keep refining until the effect is extraordinary.
---
diff --git a/plugin/skills/impeccable/SKILL.md b/plugin/skills/impeccable/SKILL.md
index fbd396d49..c7297a309 100644
--- a/plugin/skills/impeccable/SKILL.md
+++ b/plugin/skills/impeccable/SKILL.md
@@ -12,7 +12,7 @@ This skill gives you the tools and permission to make design that earns the name
Core principles:
- Finish the job. No hedging, no shortcuts: the deliverable is complete, except assets the user must provide.
- Scale ambition to the surface's mode (see Modes). Persuade and Experience get full commitment: bold, distinct work with a point of view. On Operate and Read, ambition shows as precision and restraint, and patterns the visitor already knows earn their place.
-- Verify in bounded passes, not a loop, and the ceiling covers the whole cycle: screenshots, defect scans, micro-edits, and rebuilds alike. Build fully, inspect once with a batched round (desktop and mobile together on the web; the shipped device classes on a native platform), fix everything it shows in one batch, confirm with at most one more round, and stop polishing. Open-ended self-QA burns the user's money doing worse what the finish handoffs do better.
+- Verify by looking at the rendered result at the sizes it ships at, and batch the checks: fix what one look shows before looking again, rather than screenshotting after every tweak.
## Setup
diff --git a/plugin/skills/impeccable/reference/craft-floor.md b/plugin/skills/impeccable/reference/craft-floor.md
index 9653b9808..64a195eba 100644
--- a/plugin/skills/impeccable/reference/craft-floor.md
+++ b/plugin/skills/impeccable/reference/craft-floor.md
@@ -4,7 +4,7 @@ Load this after the direction is settled, and build without announcing the check
## 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.
+Each of these is a check on the built result, not an intention. Run them together in each batched look at the render, not as separate screenshot trips; the checks share one render.
- **Contrast:** body and placeholder text ≥4.5:1, large text ≥3:1. On colored surfaces tint secondary text from that hue or the foreground; never gray.
- **Depth:** shadows carry an offset and a soft blur. A zero-offset colored halo is decoration.
diff --git a/plugin/skills/impeccable/reference/new-work.md b/plugin/skills/impeccable/reference/new-work.md
index e969e424e..960b7dcfb 100644
--- a/plugin/skills/impeccable/reference/new-work.md
+++ b/plugin/skills/impeccable/reference/new-work.md
@@ -135,7 +135,7 @@ Preserve semantics, accessibility, performance, responsiveness, project conventi
## 7. Inspect and finish
-Inspect the surface's target sizes in one batched screenshot round: desktop and mobile on the web; on a native platform (`ios` / `android` / `adaptive`), the shipped device classes per OS, captured from the simulator or emulator the way the platform reference's Verifying the build section describes. When the harness reports the user's actual viewport (an in-app browser's size, a named resolution), add that width to the set: the width that breaks is the one the user sees first. Critique the render against the user's request and the direction contract, fix material gaps, and confirm with one final round; two rounds is the ceiling, and fixes batch between them rather than earning per-tweak screenshots. The ceiling bounds inspection and never closes an open build-phase gate: a gate stays open until it passes or the route it names is taken. On a comp-led build, run `"<skill-base-dir>/scripts/impeccable" comp-diff --comp <approved comp> --build .impeccable/review/desktop.png --spec .impeccable/build/spec.json --out-dir .impeccable/review/diff/final` and read its region rows and paired crops as the critique: the side-by-side is the view the build thread never has on its own, and a region it scores missing or contradicted is a fix whatever the page looks like from memory. Never judge fidelity from one full-page thumbnail; it hides exactly the failures that matter. On a Persuade surface, verify the mode did its job: a first-time visitor should know what this is, why it matters, and what to do within seconds, in the form's own vocabulary.
+Inspect the surface's target sizes in one batched screenshot round: desktop and mobile on the web; on a native platform (`ios` / `android` / `adaptive`), the shipped device classes per OS, captured from the simulator or emulator the way the platform reference's Verifying the build section describes. When the harness reports the user's actual viewport (an in-app browser's size, a named resolution), add that width to the set: the width that breaks is the one the user sees first. On anything interactive, the round drives every interaction and reachable state (open, closed, hover, focus, error, empty, overlong content) and checks each for overlap and clipping, not only the resting screen. Critique the render against the user's request and the direction contract, fix material gaps, and confirm with one final round; two rounds is the ceiling, and fixes batch between them rather than earning per-tweak screenshots. The ceiling bounds inspection and never closes an open build-phase gate: a gate stays open until it passes or the route it names is taken. On a comp-led build, run `"<skill-base-dir>/scripts/impeccable" comp-diff --comp <approved comp> --build .impeccable/review/desktop.png --spec .impeccable/build/spec.json --out-dir .impeccable/review/diff/final` and read its region rows and paired crops as the critique: the side-by-side is the view the build thread never has on its own, and a region it scores missing or contradicted is a fix whatever the page looks like from memory. Never judge fidelity from one full-page thumbnail; it hides exactly the failures that matter. On a Persuade surface, verify the mode did its job: a first-time visitor should know what this is, why it matters, and what to do within seconds, in the form's own vocabulary.
A capture is evidence only when it is valid, and you validate before you send. Settle or disable entrance motion first: an element hidden by animation timing reads as a missing element and gets fixed into a regression. Capture full-page shots from the document top. Capture the comp comparison at the comp's own pixel dimensions. Then open every file once and confirm it shows what its name claims: no black or blank regions, no wrong section behind a right filename, no half-loaded state. A malformed capture sent onward costs the whole round; the reviewer answers it with `disposition: recapture` and nothing it reviewed binds.
diff --git a/plugin/skills/impeccable/reference/overdrive.md b/plugin/skills/impeccable/reference/overdrive.md
index fd8f544f3..01a1cb131 100644
--- a/plugin/skills/impeccable/reference/overdrive.md
+++ b/plugin/skills/impeccable/reference/overdrive.md
@@ -21,7 +21,7 @@ Skipping this step risks building something embarrassing that needs to be thrown
### Verify in the Browser
-Ambitious effects rarely work on the first try, so verify them in the browser rather than from code: an effect that technically runs can still stutter, ease wrong, or land on the wrong frame. SKILL.md's bounded passes apply: one batched inspection, one fix batch, at most one confirming round. Point them at the effect's timing, easing, and smoothness, then stop and name what still falls short instead of opening another round.
+Ambitious effects rarely work on the first try, so verify them in the browser rather than from code: an effect that technically runs can still stutter, ease wrong, or land on the wrong frame. Aim the checks at timing, easing, and smoothness, batch the fixes between looks, and keep refining until the effect is extraordinary.
---