get-shit-pretty 0.6.0 → 0.6.2

package/gsp/skills/gsp-brand-strategy/SKILL.md

@@ -2,6 +2,8 @@
 name: brand-strategy
 description: Define positioning, voice, and messaging
 user-invocable: true
+model: opus
+effort: high
 allowed-tools:
   - Read
   - Write
@@ -32,6 +34,8 @@ Define brand strategy and voice through interactive creative direction, then pro
 
 <rules>
 - Always use `AskUserQuestion` for user-facing questions — never raw text prompts
+- One decision per question — never batch multiple questions in a single message
+- Never re-ask what the user already answered in a prior phase — read BRIEF.md and build on it
 - Push opinionated recommendations but let the user decide
 - Quality gate: if you could swap in a competitor's name and it still works, it's too generic
 - Artifacts must balance human readability with agent consumption for downstream phases
@@ -61,9 +65,11 @@ Frame as: "Here's where this brand can win." Keep it to 4-6 lines.
 
 ## Step 3: Archetype selection
 
-Use `AskUserQuestion` with 2-3 archetype candidates. Each option:
+Read the personality direction from BRIEF.md (gathered during `/gsp:start`). Use it as the starting point — don't re-ask for personality. Deepen it into a structural archetype.
+
+Use `AskUserQuestion` with 2-3 archetype candidates that align with the chosen personality direction. Each option:
 - **Label:** archetype name
-- **Description:** strategic reasoning — why it fits the personas and gaps
+- **Description:** strategic reasoning — why it fits the personas and gaps, and how it builds on the personality direction from the brief
 - **Preview:** example sentence in that archetype's voice
 
 Push a recommendation. Let user choose, adjust, or blend.

package/gsp/skills/gsp-brand-sync/SKILL.md

@@ -2,6 +2,8 @@
 name: brand-sync
 description: Sync brand to match a project's shipped state — tokens, voice, visual patterns, personality
 user-invocable: true
+model: opus
+effort: high
 allowed-tools:
   - Read
   - Write
@@ -37,6 +39,7 @@ Compare a project's shipped state against its source brand across all dimensions
 
 <rules>
 - Always use `AskUserQuestion` for user interaction — never prompt via plain text
+- One decision per question — never batch multiple questions in a single message
 - Never update the brand without explicit user confirmation
 - Show before/after for every change — no silent updates
 - Only update dimensions that actually diverged — don't regenerate the entire brand

package/gsp/skills/gsp-design-system/SKILL.md

@@ -2,6 +2,7 @@
 name: design-system
 description: Scan and document the existing design system state
 user-invocable: true
+model: sonnet
 allowed-tools:
   - Read
   - Write

package/gsp/skills/gsp-doctor/SKILL.md

@@ -2,6 +2,7 @@
 name: doctor
 description: Check project health
 user-invocable: true
+model: sonnet
 allowed-tools:
   - Read
   - Glob
@@ -239,6 +240,22 @@ Glob for all SKILL.md files in the skills directory (`{runtime-dir}/skills/*/SKI
 - All present → PASS
 - Missing → WARN: "Skills missing `user-invocable: true`: {list}. They won't appear in the slash-command menu. Re-run the installer or add the field manually."
 
+**Check I2: Skill directories are complete (not just SKILL.md)**
+For each gsp-* skill directory, check if `SKILL.md` references sibling files via `${CLAUDE_SKILL_DIR}/` paths (e.g. `styles/INDEX.yml`). If it does, verify those files/dirs exist in the installed skill directory.
+- All referenced siblings present → PASS
+- Missing siblings → FAIL: "Skill {name} references {path} but it's missing. Re-run the installer: `npx get-shit-pretty`"
+
+**Check I3: Bundle directories accessible**
+Check that the runtime bundle directories exist (`{runtime-dir}/prompts/`, `{runtime-dir}/templates/`, `{runtime-dir}/references/`). Skills reference these via `${CLAUDE_SKILL_DIR}/../../`.
+- All present → PASS
+- Missing → FAIL: "Bundle directory {dir} missing. Re-run the installer: `npx get-shit-pretty`"
+
+**Check I4: VERSION file present**
+Check `{runtime-dir}/VERSION` exists and contains a valid semver string.
+- Present and valid → PASS (show version)
+- Missing → WARN: "VERSION file missing. Re-run the installer."
+- Mismatched with source → INFO: "Installed version {installed} differs from source {source}."
+
 ### Cross-Instance Checks
 
 **Check X1: Multiple projects, same brand**
@@ -287,6 +304,9 @@ Overall Health: {SCORE}/100 {emoji}
 
 ─── Installation Health ───────────────
   ✅ I1. Skills invocable ........ PASS
+  ✅ I2. Skill completeness ...... PASS
+  ✅ I3. Bundle directories ...... PASS
+  ✅ I4. VERSION file ............ PASS
 
 ─── Cross-Instance ────────────────────
   ✅ X1. Brand Consistency ...... PASS

package/gsp/skills/gsp-help/SKILL.md

@@ -2,6 +2,7 @@
 name: help
 description: Show all skills
 user-invocable: true
+model: sonnet
 ---
 <objective>
 Display the complete GSP skill reference as plain text.

package/gsp/skills/gsp-launch/SKILL.md

@@ -2,6 +2,9 @@
 name: launch
 description: Create launch and marketing assets
 user-invocable: true
+model: opus
+effort: high
+context: fork
 allowed-tools:
   - Read
   - Write
@@ -63,9 +66,15 @@ Read `{PROJECT_PATH}/BRIEF.md` — audience, goals.
 
 ## Step 2: Spawn campaign director
 
-Spawn the `gsp-campaign-director` agent with identity chunks, strategy voice-and-tone + messaging chunks, screen chunks, and brief. The Marketing Asset Factory prompt (04). The launch output template.
+Spawn the `gsp-campaign-director` agent. **Inline all content** — the agent should not need to read any input files.
 
-**Output path:** `{PROJECT_PATH}/launch/`
+Pass in the agent prompt:
+- **Content of** all identity chunks (loaded in Step 1)
+- **Content of** strategy voice-and-tone.md + messaging.md (loaded in Step 1)
+- **Content of** all design screen chunks (loaded in Step 1)
+- **Content of** BRIEF.md (loaded in Step 1)
+- The Marketing Asset Factory prompt (04), launch output template (from execution_context)
+- **Output path:** `{PROJECT_PATH}/launch/`
 
 The agent writes chunks directly:
 - `launch/campaign-strategy.md`

package/gsp/skills/gsp-palette/SKILL.md

@@ -2,6 +2,7 @@
 name: palette
 description: Generate OKLCH color palettes — standalone or as a building block for identity
 user-invocable: true
+model: sonnet
 allowed-tools:
   - Read
   - Write
@@ -31,6 +32,7 @@ Generate production-ready OKLCH color palettes from hex colors.
 
 <rules>
 - Always use `AskUserQuestion` for user interaction — never prompt via plain text
+- One decision per question — never batch multiple questions in a single message
 - Palettes are deterministic — same hex input always produces the same OKLCH scales
 - Foundation chunks follow `references/chunk-format.md` format exactly
 - Every palette gets the full 11-stop scale: 50, 100, 200, 300, 400, 500, 600, 700, 800, 900, 950

package/gsp/skills/gsp-pretty/SKILL.md

@@ -2,6 +2,7 @@
 name: pretty
 description: "Surprise ASCII art in the terminal"
 user-invocable: true
+model: sonnet
 allowed-tools:
   - Read
   - Bash

package/gsp/skills/gsp-progress/SKILL.md

@@ -2,6 +2,7 @@
 name: progress
 description: How pretty are we?
 user-invocable: true
+model: sonnet
 allowed-tools:
   - Read
   - Glob

package/gsp/skills/gsp-project-brief/SKILL.md

@@ -2,6 +2,7 @@
 name: project-brief
 description: Scope what you're building
 user-invocable: true
+model: sonnet
 allowed-tools:
   - Read
   - Write
@@ -79,15 +80,15 @@ If the project scope feels large, suggest breaking it into multiple bounded issu
 
 ## Step 2: Spawn scoper
 
-Spawn the `gsp-scoper` agent with:
-- Brand system chunks (or fallback)
-- tokens.json
-- BRIEF.md
-- config.json preferences
-- `.design/system/STACK.md`, `COMPONENTS.md` (when exist)
-- CHANGELOG.md + relevant MANIFEST.md files — for overlap detection and provenance
-- The Project Scoper prompt (10)
-- The brief output template
+Spawn the `gsp-scoper` agent. **Inline all content** — the agent should not need to read any input files.
+
+Pass in the agent prompt:
+- **Content of** brand patterns foundation + component chunks (loaded in Step 1)
+- **Content of** tokens.json (loaded in Step 1)
+- **Content of** BRIEF.md (loaded in Step 1)
+- **Content of** `.design/system/STACK.md`, `COMPONENTS.md` (when loaded in Step 1)
+- **Content of** CHANGELOG.md + relevant MANIFEST.md files (loaded in Step 1)
+- The Project Scoper prompt (10), brief output template (from execution_context)
 - `implementation_target`, `design_scope`, `codebase_type`
 - **Output path:** `{PROJECT_PATH}/brief/`
 

package/gsp/skills/gsp-project-build/SKILL.md

@@ -2,6 +2,8 @@
 name: project-build
 description: Translate designs to code
 user-invocable: true
+model: opus
+effort: high
 allowed-tools:
   - Read
   - Write
@@ -49,9 +51,6 @@ Implement designs as production-ready code in the codebase via phased pipeline w
 <execution_context>
 @${CLAUDE_SKILL_DIR}/../../prompts/09-design-to-code-translator.md
 @${CLAUDE_SKILL_DIR}/../../templates/phases/build.md
-@${CLAUDE_SKILL_DIR}/../../references/visual-effects.md
-@${CLAUDE_SKILL_DIR}/../../references/block-patterns.md
-@${CLAUDE_SKILL_DIR}/../../references/anti-patterns.md
 </execution_context>
 
 <process>
@@ -110,6 +109,15 @@ After scaffold completes, verify `{PROJECT_PATH}/build/SCAFFOLD-LOG.md` exists.
 
 **Gate:** If scaffold reports build failure, stop and surface the error. Do not proceed to foundations with a broken build.
 
+## Step 2.5: Load build references
+
+Read these reference files (relative to skill dir `${CLAUDE_SKILL_DIR}/../../references/`):
+- `visual-effects.md`
+- `block-patterns.md`
+- `anti-patterns.md`
+
+Hold their content for inlining into agent prompts in Steps 3 and 5.
+
 ## Step 3: Phase 2 — FOUNDATIONS
 
 Spawn `gsp-builder` agent with **execution_mode: foundations**.
@@ -125,7 +133,8 @@ Spawn `gsp-builder` agent with **execution_mode: foundations**.
 | `.design/system/CONVENTIONS.md` | Codebase conventions (if exists) |
 | `.design/system/COMPONENTS.md` | Existing components (if exists) |
 | `{PROJECT_PATH}/config.json` | Tech stack, target |
-| Design-to-Code Translator prompt (09) | Translation methodology |
+| Design-to-Code Translator prompt (09, from execution_context) | Translation methodology |
+| Visual effects, block patterns, anti-patterns refs (loaded in Step 2.5) | Design patterns + constraints |
 
 ### Agent instructions:
 
@@ -222,7 +231,8 @@ Build screens sequentially. For each screen in `SCREENS`:
 | `{PROJECT_PATH}/brief/target-adaptations.md` | Component adaptations |
 | `{PROJECT_PATH}/research/reference-specs.md` (if exists) | Technical specs |
 | `{PROJECT_PATH}/critique/prioritized-fixes.md` (if exists) | Critique fixes relevant to this screen |
-| Design-to-Code Translator prompt (09) | Translation methodology |
+| Design-to-Code Translator prompt (09, from execution_context) | Translation methodology |
+| Visual effects, block patterns, anti-patterns refs (loaded in Step 2.5) | Design patterns + constraints |
 
 **Does NOT receive:** other screen chunks, tokens.json (already in codebase), full brand system, research monoliths.
 

package/gsp/skills/gsp-project-critique/SKILL.md

@@ -2,6 +2,9 @@
 name: project-critique
 description: Critique your designs + accessibility audit
 user-invocable: true
+model: opus
+effort: high
+context: fork
 allowed-tools:
   - Read
   - Write
@@ -26,11 +29,6 @@ Critique design quality and audit accessibility compliance.
 @${CLAUDE_SKILL_DIR}/../../prompts/06-design-critique-partner.md
 @${CLAUDE_SKILL_DIR}/../../prompts/08-accessibility-auditor.md
 @${CLAUDE_SKILL_DIR}/../../templates/phases/critique.md
-@${CLAUDE_SKILL_DIR}/../../references/nielsen-heuristics.md
-@${CLAUDE_SKILL_DIR}/../../references/visual-taste.md
-@${CLAUDE_SKILL_DIR}/../../references/anti-patterns.md
-@${CLAUDE_SKILL_DIR}/../../references/wcag-checklist.md
-@${CLAUDE_SKILL_DIR}/../../references/color-composition.md
 </execution_context>
 
 <process>
@@ -72,13 +70,39 @@ Read `{PROJECT_PATH}/config.json` to get `implementation_target`, `design_scope`
 
 **Otherwise:** If design chunks don't exist and scope is not `tokens`, tell the user to complete the design phase first.
 
-## Step 2: Spawn critics (parallel)
+## Step 1.8: Load critique references
+
+Read these reference files (relative to skill dir `${CLAUDE_SKILL_DIR}/../../references/`):
+- `nielsen-heuristics.md`
+- `visual-taste.md`
+- `anti-patterns.md`
+- `wcag-checklist.md`
+- `color-composition.md`
 
-**Agent 1: gsp-critic** — Design critique using Nielsen's 10 Heuristics reference + all design chunks.
-Output path: `{PROJECT_PATH}/critique/`
+Hold their content for inlining into agent prompts in Step 2.
+
+## Step 2: Spawn critics (parallel)
 
-**Agent 2: gsp-accessibility-auditor** — Check if `{PROJECT_PATH}/critique/accessibility-audit.md` already exists from a prior `/gsp:accessibility` run. If yes, skip spawning the accessibility auditor — reuse the existing output. If no, spawn `gsp-accessibility-auditor` with WCAG 2.2 audit using WCAG checklist + all design chunks. Pass `accessibility_level` from config (defaults to "WCAG 2.2 AA") so the auditor adapts its criteria (AA vs AAA).
-Output path: `{PROJECT_PATH}/critique/`
+**Inline all content** — agents should not need to read any input files.
+
+**Agent 1: gsp-critic** — Pass in the agent prompt:
+- **Content of** all design chunks (loaded in Step 1)
+- **Content of** all identity chunks (loaded in Step 1)
+- **Content of** all patterns chunks (loaded in Step 1)
+- **Content of** brief chunks (loaded in Step 1)
+- **Content of** research recommendations.md (loaded in Step 1)
+- **Content of** BRIEF.md
+- **Content of** Nielsen's 10 Heuristics, visual taste, anti-patterns, color composition references (loaded in Step 1.8)
+- The Design Critique Partner prompt (06), critique output template (from execution_context)
+- Output path: `{PROJECT_PATH}/critique/`
+
+**Agent 2: gsp-accessibility-auditor** — Check if `{PROJECT_PATH}/critique/accessibility-audit.md` already exists from a prior `/gsp:accessibility` run. If yes, skip spawning the accessibility auditor — reuse the existing output. If no, pass in the agent prompt:
+- **Content of** all design chunks (loaded in Step 1)
+- **Content of** identity color-system.md and typography.md (loaded in Step 1)
+- **Content of** patterns tokens chunks (loaded in Step 1)
+- **Content of** WCAG checklist reference (loaded in Step 1.8), Accessibility Auditor prompt (08, from execution_context)
+- `accessibility_level` from config (defaults to "WCAG 2.2 AA")
+- Output path: `{PROJECT_PATH}/critique/`
 
 ## Step 3: Write critique INDEX.md
 

package/gsp/skills/gsp-project-design/SKILL.md

@@ -2,6 +2,9 @@
 name: project-design
 description: Design screens and interaction flows
 user-invocable: true
+model: opus
+effort: high
+context: fork
 allowed-tools:
   - Read
   - Write
@@ -27,10 +30,6 @@ Design core UI/UX screens and interaction flows.
 <execution_context>
 @${CLAUDE_SKILL_DIR}/../../prompts/03-ui-ux-pattern-master.md
 @${CLAUDE_SKILL_DIR}/../../templates/phases/design.md
-@${CLAUDE_SKILL_DIR}/../../references/apple-hig-patterns.md
-@${CLAUDE_SKILL_DIR}/../../references/visual-effects.md
-@${CLAUDE_SKILL_DIR}/../../references/block-patterns.md
-@${CLAUDE_SKILL_DIR}/../../references/anti-patterns.md
 </execution_context>
 
 <process>
@@ -111,11 +110,35 @@ When `implementation_target` is not `figma`:
 - **If `.design/system/COMPONENTS.md` and `.design/system/TOKENS.md` exist**, read them. Pass to the agent.
 - **If not**, fall back to scanning the codebase.
 
-## Step 3: Spawn designer
+## Step 2.5: Load design references
+
+Read these reference files (relative to skill dir `${CLAUDE_SKILL_DIR}/../../references/`):
+- `apple-hig-patterns.md`
+- `visual-effects.md`
+- `block-patterns.md`
+- `anti-patterns.md`
 
-Spawn the `gsp-designer` agent with all prior artifacts, the UI/UX Pattern Master prompt (03), design output template, Apple HIG patterns reference, brand style prompt ({brand-name}.md when available), implementation_target, design_scope, codebase_type, target screens (when partial), existing components inventory, custom references (when available), and critique fixes (when in revision mode).
+Hold their content for inlining into the agent prompt in Step 3.
+
+## Step 3: Spawn designer
 
-**Output path:** `{PROJECT_PATH}/design/`
+Spawn the `gsp-designer` agent. **Inline all content** — the agent should not need to read any input files.
+
+Pass in the agent prompt:
+- **Content of** all brand patterns foundation chunks + selective component chunks (loaded in Step 1)
+- **Content of** brand identity chunks: color-system.md, typography.md, imagery-style.md (loaded in Step 1)
+- **Content of** brand style prompt ({brand-name}.md) when available
+- **Content of** brief chunks: scope.md, target-adaptations.md (loaded in Step 1)
+- **Content of** research chunks: ux-patterns.md, recommendations.md, reference-specs.md (loaded in Step 1)
+- **Content of** BRIEF.md
+- **Content of** `.design/system/COMPONENTS.md`, `TOKENS.md` (when loaded in Step 2)
+- **Content of** custom references (when loaded in Step 1)
+- **Content of** critique fixes: prioritized-fixes.md, accessibility-fixes.md (when in revision mode)
+- The UI/UX Pattern Master prompt (03), design output template (from execution_context)
+- **Content of** Apple HIG patterns, visual effects, block patterns, anti-patterns references (loaded in Step 2.5)
+- `implementation_target`, `design_scope`, `codebase_type`
+- Target screens (when partial)
+- **Output path:** `{PROJECT_PATH}/design/`
 
 The agent writes chunks directly:
 - `design/screen-{NN}-{name}.md` (one per screen)

package/gsp/skills/gsp-project-research/SKILL.md

@@ -2,6 +2,8 @@
 name: project-research
 description: Research UX patterns and technical approaches
 user-invocable: true
+model: sonnet
+effort: high
 allowed-tools:
   - Read
   - Write
@@ -80,16 +82,16 @@ If competitor URLs or reference sites are mentioned in BRIEF.md or `{PROJECT_PAT
 
 ## Step 2: Spawn project researcher
 
-Include any pre-fetched reference content in the agent context.
+Spawn the `gsp-project-researcher` agent. **Inline all content** — the agent should not need to read any input files.
 
-Spawn the `gsp-project-researcher` agent with:
-- Brief scope chunks
-- Brand system foundation chunks
-- Brand discovery chunks (competitive audit, trends — to build on, not duplicate)
-- BRIEF.md
-- config.json preferences
-- The Project Researcher prompt (12)
-- The research output template
+Pass in the agent prompt:
+- **Content of** brief scope chunks: scope.md, target-adaptations.md (loaded in Step 1)
+- **Content of** brand patterns foundation chunks (loaded in Step 1)
+- **Content of** brand discovery chunks: competitive-audit.md, trend-analysis.md (loaded in Step 1)
+- **Content of** custom references (loaded in Step 1)
+- **Content of** BRIEF.md (loaded in Step 1)
+- Any pre-fetched reference content (from Step 1.75)
+- The Project Researcher prompt (12), research output template (from execution_context)
 - `implementation_target`, `platform`, `tech_stack`
 - **Output path:** `{PROJECT_PATH}/research/`
 

package/gsp/skills/gsp-project-review/SKILL.md

@@ -2,6 +2,9 @@
 name: project-review
 description: QA review — validate implementation against designs
 user-invocable: true
+model: opus
+effort: high
+context: fork
 allowed-tools:
   - Read
   - Write

package/gsp/skills/gsp-scaffold/SKILL.md

@@ -2,6 +2,7 @@
 name: scaffold
 description: Deterministic stack setup — install deps, create configs, verify build compiles
 user-invocable: true
+model: sonnet
 allowed-tools:
   - Read
   - Write