get-shit-pretty 0.1.0
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- package/README.md +111 -0
- package/agents/gsp-accessibility-auditor.md +52 -0
- package/agents/gsp-brand-strategist.md +45 -0
- package/agents/gsp-campaign-director.md +48 -0
- package/agents/gsp-critic.md +55 -0
- package/agents/gsp-design-engineer.md +53 -0
- package/agents/gsp-researcher.md +47 -0
- package/agents/gsp-spec-engineer.md +45 -0
- package/agents/gsp-system-architect.md +56 -0
- package/agents/gsp-ui-designer.md +51 -0
- package/bin/install.js +1084 -0
- package/commands/gsp/brand.md +68 -0
- package/commands/gsp/build.md +75 -0
- package/commands/gsp/design.md +71 -0
- package/commands/gsp/help.md +68 -0
- package/commands/gsp/launch.md +75 -0
- package/commands/gsp/new-project.md +91 -0
- package/commands/gsp/progress.md +61 -0
- package/commands/gsp/research.md +71 -0
- package/commands/gsp/review.md +99 -0
- package/commands/gsp/spec.md +68 -0
- package/commands/gsp/system.md +72 -0
- package/package.json +38 -0
- package/prompts/01-design-system-architect.md +29 -0
- package/prompts/02-brand-identity-creator.md +29 -0
- package/prompts/03-ui-ux-pattern-master.md +31 -0
- package/prompts/04-marketing-asset-factory.md +27 -0
- package/prompts/05-figma-auto-layout-expert.md +27 -0
- package/prompts/06-design-critique-partner.md +30 -0
- package/prompts/07-design-trend-synthesizer.md +30 -0
- package/prompts/08-accessibility-auditor.md +29 -0
- package/prompts/09-design-to-code-translator.md +31 -0
- package/references/apple-hig-patterns.md +141 -0
- package/references/design-tokens.md +182 -0
- package/references/nielsen-heuristics.md +141 -0
- package/references/questioning.md +73 -0
- package/references/wcag-checklist.md +111 -0
- package/scripts/gsp-statusline.js +132 -0
- package/templates/config.json +25 -0
- package/templates/phases/brand.md +60 -0
- package/templates/phases/build.md +59 -0
- package/templates/phases/design.md +48 -0
- package/templates/phases/launch.md +62 -0
- package/templates/phases/research.md +53 -0
- package/templates/phases/review.md +88 -0
- package/templates/phases/spec.md +60 -0
- package/templates/phases/system.md +84 -0
- package/templates/project.md +48 -0
- package/templates/roadmap.md +62 -0
- package/templates/state.md +33 -0
|
@@ -0,0 +1,68 @@
|
|
|
1
|
+
---
|
|
2
|
+
name: gsp:brand
|
|
3
|
+
description: Create brand identity — strategy, logo directions, color, typography
|
|
4
|
+
allowed-tools:
|
|
5
|
+
- Read
|
|
6
|
+
- Write
|
|
7
|
+
- Bash
|
|
8
|
+
- Task
|
|
9
|
+
- WebSearch
|
|
10
|
+
- WebFetch
|
|
11
|
+
---
|
|
12
|
+
<context>
|
|
13
|
+
Phase 2 of the GSP design pipeline. Uses the Brand Identity Creator prompt to develop a complete brand identity from strategy to visual applications.
|
|
14
|
+
</context>
|
|
15
|
+
|
|
16
|
+
<objective>
|
|
17
|
+
Create a complete brand identity for the project.
|
|
18
|
+
|
|
19
|
+
**Input:** `.design/BRIEF.md` + `.design/research/TRENDS.md`
|
|
20
|
+
**Output:** `.design/brand/IDENTITY.md`
|
|
21
|
+
**Agent:** `gsp-brand-strategist`
|
|
22
|
+
</objective>
|
|
23
|
+
|
|
24
|
+
<execution_context>
|
|
25
|
+
@/Users/jubs/.claude/get-shit-pretty/prompts/02-brand-identity-creator.md
|
|
26
|
+
@/Users/jubs/.claude/get-shit-pretty/templates/phases/brand.md
|
|
27
|
+
</execution_context>
|
|
28
|
+
|
|
29
|
+
<process>
|
|
30
|
+
## Step 1: Load context
|
|
31
|
+
|
|
32
|
+
Read:
|
|
33
|
+
- `.design/BRIEF.md` — project context, audience, personality
|
|
34
|
+
- `.design/research/TRENDS.md` — trend insights (if available; proceed without if not)
|
|
35
|
+
|
|
36
|
+
If BRIEF.md doesn't exist, tell the user to run `/gsp:new-project` first.
|
|
37
|
+
|
|
38
|
+
## Step 2: Spawn brand strategist
|
|
39
|
+
|
|
40
|
+
Spawn the `gsp-brand-strategist` agent with:
|
|
41
|
+
- The full BRIEF.md and TRENDS.md content
|
|
42
|
+
- The Brand Identity Creator prompt (02)
|
|
43
|
+
- The brand output template
|
|
44
|
+
|
|
45
|
+
The agent should deliver:
|
|
46
|
+
1. Brand strategy (story, archetype, voice matrix, messaging hierarchy)
|
|
47
|
+
2. 3 logo directions with variations and usage rules
|
|
48
|
+
3. Complete color system (Hex, RGB, Pantone, CMYK + rationale)
|
|
49
|
+
4. Typography system
|
|
50
|
+
5. Imagery and photography style guide
|
|
51
|
+
6. Brand applications
|
|
52
|
+
7. Brand book structure
|
|
53
|
+
|
|
54
|
+
## Step 3: Write output
|
|
55
|
+
|
|
56
|
+
Write the completed identity to `.design/brand/IDENTITY.md`.
|
|
57
|
+
|
|
58
|
+
## Step 4: Update state
|
|
59
|
+
|
|
60
|
+
Update `.design/STATE.md`:
|
|
61
|
+
- Set Phase 2 (Brand) status to `complete`
|
|
62
|
+
- Record completion date
|
|
63
|
+
|
|
64
|
+
## Step 5: Route next
|
|
65
|
+
|
|
66
|
+
Display brand identity summary and end with:
|
|
67
|
+
"Run `/gsp:system` to build the design system from this identity."
|
|
68
|
+
</process>
|
|
@@ -0,0 +1,75 @@
|
|
|
1
|
+
---
|
|
2
|
+
name: gsp:build
|
|
3
|
+
description: Translate designs to production-ready frontend code
|
|
4
|
+
allowed-tools:
|
|
5
|
+
- Read
|
|
6
|
+
- Write
|
|
7
|
+
- Bash
|
|
8
|
+
- Task
|
|
9
|
+
---
|
|
10
|
+
<context>
|
|
11
|
+
Phase 7 of the GSP design pipeline. Uses the Design-to-Code Translator prompt to convert Figma specs and design system into production-ready frontend components.
|
|
12
|
+
</context>
|
|
13
|
+
|
|
14
|
+
<objective>
|
|
15
|
+
Translate designs into production-ready frontend code.
|
|
16
|
+
|
|
17
|
+
**Input:** `.design/specs/FIGMA-SPECS.md` + `.design/system/SYSTEM.md`
|
|
18
|
+
**Output:** `.design/build/CODE.md` + `.design/build/components/`
|
|
19
|
+
**Agent:** `gsp-design-engineer`
|
|
20
|
+
</objective>
|
|
21
|
+
|
|
22
|
+
<execution_context>
|
|
23
|
+
@/Users/jubs/.claude/get-shit-pretty/prompts/09-design-to-code-translator.md
|
|
24
|
+
@/Users/jubs/.claude/get-shit-pretty/templates/phases/build.md
|
|
25
|
+
</execution_context>
|
|
26
|
+
|
|
27
|
+
<process>
|
|
28
|
+
## Step 1: Load context
|
|
29
|
+
|
|
30
|
+
Read:
|
|
31
|
+
- `.design/specs/FIGMA-SPECS.md` — component specs
|
|
32
|
+
- `.design/system/SYSTEM.md` — design system
|
|
33
|
+
- `.design/system/tokens.json` — design tokens
|
|
34
|
+
- `.design/BRIEF.md` — tech stack preference
|
|
35
|
+
- `.design/review/CRITIQUE.md` — any fixes to incorporate
|
|
36
|
+
|
|
37
|
+
If FIGMA-SPECS.md doesn't exist, tell the user to run `/gsp:spec` first.
|
|
38
|
+
|
|
39
|
+
Determine tech stack from BRIEF.md config (default: React + Tailwind).
|
|
40
|
+
|
|
41
|
+
## Step 2: Spawn design engineer
|
|
42
|
+
|
|
43
|
+
Spawn the `gsp-design-engineer` agent with:
|
|
44
|
+
- All specs, system, and token files
|
|
45
|
+
- The Design-to-Code Translator prompt (09)
|
|
46
|
+
- The build output template
|
|
47
|
+
- The target tech stack
|
|
48
|
+
|
|
49
|
+
The agent should deliver:
|
|
50
|
+
1. Component hierarchy with props, state, data flow
|
|
51
|
+
2. Production-ready code for each component
|
|
52
|
+
3. Responsive layout implementation
|
|
53
|
+
4. ARIA and accessibility implementation
|
|
54
|
+
5. Error, loading, and empty states
|
|
55
|
+
6. Animations and transitions
|
|
56
|
+
7. Styling with design tokens (dark mode, breakpoints)
|
|
57
|
+
8. Performance optimization notes
|
|
58
|
+
9. Testing strategy
|
|
59
|
+
|
|
60
|
+
## Step 3: Write output
|
|
61
|
+
|
|
62
|
+
1. Write implementation guide to `.design/build/CODE.md`
|
|
63
|
+
2. Write individual components to `.design/build/components/`
|
|
64
|
+
|
|
65
|
+
## Step 4: Update state
|
|
66
|
+
|
|
67
|
+
Update `.design/STATE.md`:
|
|
68
|
+
- Set Phase 7 (Build) status to `complete`
|
|
69
|
+
- Record completion date
|
|
70
|
+
|
|
71
|
+
## Step 5: Route next
|
|
72
|
+
|
|
73
|
+
Display component summary and end with:
|
|
74
|
+
"Run `/gsp:launch` to create marketing campaign assets."
|
|
75
|
+
</process>
|
|
@@ -0,0 +1,71 @@
|
|
|
1
|
+
---
|
|
2
|
+
name: gsp:design
|
|
3
|
+
description: Design UI/UX screens, flows, and interaction patterns
|
|
4
|
+
allowed-tools:
|
|
5
|
+
- Read
|
|
6
|
+
- Write
|
|
7
|
+
- Bash
|
|
8
|
+
- Task
|
|
9
|
+
---
|
|
10
|
+
<context>
|
|
11
|
+
Phase 4 of the GSP design pipeline. Uses the UI/UX Pattern Master prompt to design core screens following Apple HIG and the project's design system.
|
|
12
|
+
</context>
|
|
13
|
+
|
|
14
|
+
<objective>
|
|
15
|
+
Design core UI/UX screens and interaction flows.
|
|
16
|
+
|
|
17
|
+
**Input:** `.design/system/SYSTEM.md` + `.design/BRIEF.md`
|
|
18
|
+
**Output:** `.design/screens/SCREENS.md`
|
|
19
|
+
**Agent:** `gsp-ui-designer`
|
|
20
|
+
</objective>
|
|
21
|
+
|
|
22
|
+
<execution_context>
|
|
23
|
+
@/Users/jubs/.claude/get-shit-pretty/prompts/03-ui-ux-pattern-master.md
|
|
24
|
+
@/Users/jubs/.claude/get-shit-pretty/templates/phases/design.md
|
|
25
|
+
@/Users/jubs/.claude/get-shit-pretty/references/apple-hig-patterns.md
|
|
26
|
+
</execution_context>
|
|
27
|
+
|
|
28
|
+
<process>
|
|
29
|
+
## Step 1: Load context
|
|
30
|
+
|
|
31
|
+
Read:
|
|
32
|
+
- `.design/BRIEF.md` — app type, audience, goals
|
|
33
|
+
- `.design/system/SYSTEM.md` — design system to use
|
|
34
|
+
- `.design/brand/IDENTITY.md` — brand personality
|
|
35
|
+
|
|
36
|
+
If SYSTEM.md doesn't exist, tell the user to run `/gsp:system` first.
|
|
37
|
+
|
|
38
|
+
## Step 2: Spawn UI designer
|
|
39
|
+
|
|
40
|
+
Spawn the `gsp-ui-designer` agent with:
|
|
41
|
+
- All prior artifacts
|
|
42
|
+
- The UI/UX Pattern Master prompt (03)
|
|
43
|
+
- The design output template
|
|
44
|
+
- The Apple HIG patterns reference
|
|
45
|
+
|
|
46
|
+
The agent should deliver:
|
|
47
|
+
1. User personas with goals and pain points
|
|
48
|
+
2. Information architecture
|
|
49
|
+
3. Navigation pattern and gesture definitions
|
|
50
|
+
4. 8 core screens with wireframes, components, interactions
|
|
51
|
+
5. All states: empty, error, loading
|
|
52
|
+
6. Accessibility specs (WCAG, VoiceOver, Dynamic Type)
|
|
53
|
+
7. Micro-interactions and animations
|
|
54
|
+
8. Responsive behavior across breakpoints
|
|
55
|
+
9. Designer's notes
|
|
56
|
+
|
|
57
|
+
## Step 3: Write output
|
|
58
|
+
|
|
59
|
+
Write screens to `.design/screens/SCREENS.md`.
|
|
60
|
+
|
|
61
|
+
## Step 4: Update state
|
|
62
|
+
|
|
63
|
+
Update `.design/STATE.md`:
|
|
64
|
+
- Set Phase 4 (Design) status to `complete`
|
|
65
|
+
- Record completion date
|
|
66
|
+
|
|
67
|
+
## Step 5: Route next
|
|
68
|
+
|
|
69
|
+
Display screen summary (screen count, key flows) and end with:
|
|
70
|
+
"Run `/gsp:spec` to generate Figma-ready specifications."
|
|
71
|
+
</process>
|
|
@@ -0,0 +1,68 @@
|
|
|
1
|
+
---
|
|
2
|
+
name: gsp:help
|
|
3
|
+
description: Show GSP command reference
|
|
4
|
+
---
|
|
5
|
+
<objective>
|
|
6
|
+
Display the complete GSP command reference.
|
|
7
|
+
|
|
8
|
+
Output ONLY the reference content below. Do NOT add:
|
|
9
|
+
- Project-specific analysis
|
|
10
|
+
- File context or git status
|
|
11
|
+
- Next-step suggestions
|
|
12
|
+
- Any commentary beyond the reference
|
|
13
|
+
</objective>
|
|
14
|
+
|
|
15
|
+
<process>
|
|
16
|
+
Output the following reference exactly:
|
|
17
|
+
|
|
18
|
+
```
|
|
19
|
+
🎨 GSP — Get Shit Pretty
|
|
20
|
+
Command Reference
|
|
21
|
+
═══════════════════════════════════════
|
|
22
|
+
|
|
23
|
+
PROJECT SETUP
|
|
24
|
+
/gsp:new-project Initialize design brief through guided Q&A
|
|
25
|
+
/gsp:help Show this command reference
|
|
26
|
+
/gsp:progress Check project status — "How pretty are we?"
|
|
27
|
+
|
|
28
|
+
DESIGN PIPELINE (run in order)
|
|
29
|
+
/gsp:research Phase 1 — Trend analysis & competitive landscape
|
|
30
|
+
/gsp:brand Phase 2 — Brand identity (strategy, logo, color, type)
|
|
31
|
+
/gsp:system Phase 3 — Design system foundations + tokens
|
|
32
|
+
/gsp:design Phase 4 — UI/UX screens & interaction flows
|
|
33
|
+
/gsp:spec Phase 5 — Figma-ready specifications
|
|
34
|
+
/gsp:review Phase 6 — Design critique + accessibility audit
|
|
35
|
+
/gsp:build Phase 7 — Design-to-code translation
|
|
36
|
+
/gsp:launch Phase 8 — Marketing campaign assets
|
|
37
|
+
|
|
38
|
+
PIPELINE FLOW
|
|
39
|
+
new-project → research → brand → system → design → spec → review → build → launch
|
|
40
|
+
↑ |
|
|
41
|
+
└── loop back ───┘
|
|
42
|
+
(if critical issues)
|
|
43
|
+
|
|
44
|
+
PROJECT STRUCTURE
|
|
45
|
+
.design/
|
|
46
|
+
├── BRIEF.md Design brief
|
|
47
|
+
├── ROADMAP.md Phase plan
|
|
48
|
+
├── STATE.md Progress tracking
|
|
49
|
+
├── config.json Preferences
|
|
50
|
+
├── research/TRENDS.md Trend analysis
|
|
51
|
+
├── brand/IDENTITY.md Brand identity
|
|
52
|
+
├── system/SYSTEM.md Design system
|
|
53
|
+
├── system/tokens.json Design tokens
|
|
54
|
+
├── screens/SCREENS.md UI/UX screens
|
|
55
|
+
├── specs/FIGMA-SPECS.md Figma specifications
|
|
56
|
+
├── review/CRITIQUE.md Design critique
|
|
57
|
+
├── review/ACCESSIBILITY.md WCAG audit
|
|
58
|
+
├── build/CODE.md Implementation guide
|
|
59
|
+
├── build/components/ Code components
|
|
60
|
+
└── launch/CAMPAIGN.md Marketing assets
|
|
61
|
+
|
|
62
|
+
TIPS
|
|
63
|
+
• Run phases in order — each builds on the previous
|
|
64
|
+
• /gsp:review loops back if critical issues are found
|
|
65
|
+
• /gsp:progress shows your prettiness meter
|
|
66
|
+
• You can skip phases if you already have the artifacts
|
|
67
|
+
```
|
|
68
|
+
</process>
|
|
@@ -0,0 +1,75 @@
|
|
|
1
|
+
---
|
|
2
|
+
name: gsp:launch
|
|
3
|
+
description: Create marketing campaign assets for launch
|
|
4
|
+
allowed-tools:
|
|
5
|
+
- Read
|
|
6
|
+
- Write
|
|
7
|
+
- Bash
|
|
8
|
+
- Task
|
|
9
|
+
- WebSearch
|
|
10
|
+
---
|
|
11
|
+
<context>
|
|
12
|
+
Phase 8 of the GSP design pipeline. Uses the Marketing Asset Factory prompt to create a full campaign asset library across channels.
|
|
13
|
+
</context>
|
|
14
|
+
|
|
15
|
+
<objective>
|
|
16
|
+
Create marketing campaign assets for product launch.
|
|
17
|
+
|
|
18
|
+
**Input:** `.design/brand/IDENTITY.md` + `.design/screens/SCREENS.md`
|
|
19
|
+
**Output:** `.design/launch/CAMPAIGN.md`
|
|
20
|
+
**Agent:** `gsp-campaign-director`
|
|
21
|
+
</objective>
|
|
22
|
+
|
|
23
|
+
<execution_context>
|
|
24
|
+
@/Users/jubs/.claude/get-shit-pretty/prompts/04-marketing-asset-factory.md
|
|
25
|
+
@/Users/jubs/.claude/get-shit-pretty/templates/phases/launch.md
|
|
26
|
+
</execution_context>
|
|
27
|
+
|
|
28
|
+
<process>
|
|
29
|
+
## Step 1: Load context
|
|
30
|
+
|
|
31
|
+
Read:
|
|
32
|
+
- `.design/brand/IDENTITY.md` — brand voice, visual identity
|
|
33
|
+
- `.design/screens/SCREENS.md` — product screens for showcase
|
|
34
|
+
- `.design/BRIEF.md` — audience, goals, messaging
|
|
35
|
+
- `.design/system/SYSTEM.md` — design system for consistency
|
|
36
|
+
|
|
37
|
+
If IDENTITY.md doesn't exist, tell the user to run `/gsp:brand` first.
|
|
38
|
+
|
|
39
|
+
## Step 2: Spawn campaign director
|
|
40
|
+
|
|
41
|
+
Spawn the `gsp-campaign-director` agent with:
|
|
42
|
+
- Brand identity, screens, and brief
|
|
43
|
+
- The Marketing Asset Factory prompt (04)
|
|
44
|
+
- The launch output template
|
|
45
|
+
|
|
46
|
+
The agent should deliver:
|
|
47
|
+
1. Campaign strategy (objective, audience, key message, tone)
|
|
48
|
+
2. Google Ads (copy, visual direction, CTAs, A/B variants)
|
|
49
|
+
3. Meta/Instagram and TikTok ads
|
|
50
|
+
4. Email sequences (welcome, promo, nurture, re-engagement)
|
|
51
|
+
5. Landing page copy and structure
|
|
52
|
+
6. Social media posts
|
|
53
|
+
7. Sales enablement materials
|
|
54
|
+
8. Content marketing outlines
|
|
55
|
+
|
|
56
|
+
## Step 3: Write output
|
|
57
|
+
|
|
58
|
+
Write campaign to `.design/launch/CAMPAIGN.md`.
|
|
59
|
+
|
|
60
|
+
## Step 4: Update state
|
|
61
|
+
|
|
62
|
+
Update `.design/STATE.md`:
|
|
63
|
+
- Set Phase 8 (Launch) status to `complete`
|
|
64
|
+
- Record completion date
|
|
65
|
+
- Set Prettiness Level to 100%
|
|
66
|
+
|
|
67
|
+
## Step 5: Celebrate
|
|
68
|
+
|
|
69
|
+
Display campaign summary and:
|
|
70
|
+
```
|
|
71
|
+
🎨 Project is fully pretty! All 8 phases complete.
|
|
72
|
+
|
|
73
|
+
Run /gsp:progress to see the full journey.
|
|
74
|
+
```
|
|
75
|
+
</process>
|
|
@@ -0,0 +1,91 @@
|
|
|
1
|
+
---
|
|
2
|
+
name: gsp:new-project
|
|
3
|
+
description: Initialize a new design project with a design brief, roadmap, and state tracking
|
|
4
|
+
allowed-tools:
|
|
5
|
+
- Read
|
|
6
|
+
- Write
|
|
7
|
+
- Bash
|
|
8
|
+
- AskUserQuestion
|
|
9
|
+
---
|
|
10
|
+
<context>
|
|
11
|
+
You are the GSP (Get Shit Pretty) project initializer. Your job is to gather design requirements through conversational Q&A and produce a complete design brief.
|
|
12
|
+
</context>
|
|
13
|
+
|
|
14
|
+
<objective>
|
|
15
|
+
Initialize a new design project through guided questioning → brief → roadmap → state.
|
|
16
|
+
|
|
17
|
+
**Creates:**
|
|
18
|
+
- `.design/BRIEF.md` — design brief (brand, audience, goals, constraints)
|
|
19
|
+
- `.design/ROADMAP.md` — phase plan (8 phases)
|
|
20
|
+
- `.design/STATE.md` — progress tracking
|
|
21
|
+
- `.design/config.json` — project preferences
|
|
22
|
+
|
|
23
|
+
**After this command:** Run `/gsp:research` to begin the design pipeline.
|
|
24
|
+
</objective>
|
|
25
|
+
|
|
26
|
+
<execution_context>
|
|
27
|
+
@/Users/jubs/.claude/get-shit-pretty/references/questioning.md
|
|
28
|
+
@/Users/jubs/.claude/get-shit-pretty/templates/project.md
|
|
29
|
+
@/Users/jubs/.claude/get-shit-pretty/templates/roadmap.md
|
|
30
|
+
@/Users/jubs/.claude/get-shit-pretty/templates/state.md
|
|
31
|
+
@/Users/jubs/.claude/get-shit-pretty/templates/config.json
|
|
32
|
+
</execution_context>
|
|
33
|
+
|
|
34
|
+
<process>
|
|
35
|
+
## Step 1: Greet and orient
|
|
36
|
+
|
|
37
|
+
Display:
|
|
38
|
+
```
|
|
39
|
+
🎨 GSP — Get Shit Pretty
|
|
40
|
+
Starting new design project...
|
|
41
|
+
```
|
|
42
|
+
|
|
43
|
+
## Step 2: Gather the brief
|
|
44
|
+
|
|
45
|
+
Use the questioning techniques from the reference. Ask in 3 conversational rounds:
|
|
46
|
+
|
|
47
|
+
**Round 1 — Context & Brand:**
|
|
48
|
+
- What is this project? (app, website, rebrand, campaign)
|
|
49
|
+
- Who is it for? (audience)
|
|
50
|
+
- Why does it exist? (business goal, user need)
|
|
51
|
+
- Existing brand? Brand personality?
|
|
52
|
+
- Brands admired / styles to avoid?
|
|
53
|
+
|
|
54
|
+
**Round 2 — Scope & Constraints:**
|
|
55
|
+
- Platforms (web, iOS, Android)?
|
|
56
|
+
- Tech stack preferences?
|
|
57
|
+
- Key screens/pages needed?
|
|
58
|
+
- Accessibility level (default: WCAG 2.2 AA)?
|
|
59
|
+
- Timeline and constraints?
|
|
60
|
+
|
|
61
|
+
**Round 3 — Success & Gaps:**
|
|
62
|
+
- What does success look like?
|
|
63
|
+
- Any remaining questions based on gaps in rounds 1-2
|
|
64
|
+
|
|
65
|
+
Use inference over interrogation — state assumptions and let them correct. Offer concrete options when answers are vague.
|
|
66
|
+
|
|
67
|
+
## Step 3: Create project structure
|
|
68
|
+
|
|
69
|
+
```bash
|
|
70
|
+
mkdir -p .design/{research,brand,system,screens,specs,review,build,launch}
|
|
71
|
+
```
|
|
72
|
+
|
|
73
|
+
## Step 4: Write artifacts
|
|
74
|
+
|
|
75
|
+
Using the templates, fill in the gathered information:
|
|
76
|
+
1. Write `.design/BRIEF.md` from the project template with all gathered info
|
|
77
|
+
2. Write `.design/ROADMAP.md` from the roadmap template with project name and date
|
|
78
|
+
3. Write `.design/STATE.md` from the state template
|
|
79
|
+
4. Write `.design/config.json` from the config template with preferences
|
|
80
|
+
|
|
81
|
+
## Step 5: Confirm and route
|
|
82
|
+
|
|
83
|
+
Display a summary of the brief and confirm with the user. Show:
|
|
84
|
+
- Project name and type
|
|
85
|
+
- Brand personality
|
|
86
|
+
- Target audience
|
|
87
|
+
- Key deliverables
|
|
88
|
+
- Phase roadmap overview
|
|
89
|
+
|
|
90
|
+
End with: "Run `/gsp:research` to begin trend analysis, or `/gsp:brand` to jump straight to brand identity."
|
|
91
|
+
</process>
|
|
@@ -0,0 +1,61 @@
|
|
|
1
|
+
---
|
|
2
|
+
name: gsp:progress
|
|
3
|
+
description: Check design project progress — "How pretty are we?"
|
|
4
|
+
allowed-tools:
|
|
5
|
+
- Read
|
|
6
|
+
- Glob
|
|
7
|
+
---
|
|
8
|
+
<context>
|
|
9
|
+
Status check for GSP design projects. Reads STATE.md and shows phase progress with a prettiness meter.
|
|
10
|
+
</context>
|
|
11
|
+
|
|
12
|
+
<objective>
|
|
13
|
+
Display current project progress and route to next action.
|
|
14
|
+
</objective>
|
|
15
|
+
|
|
16
|
+
<process>
|
|
17
|
+
## Step 1: Find project
|
|
18
|
+
|
|
19
|
+
Check for `.design/STATE.md` in the current directory.
|
|
20
|
+
|
|
21
|
+
If not found, display:
|
|
22
|
+
```
|
|
23
|
+
No GSP project found. Run /gsp:new-project to start one.
|
|
24
|
+
```
|
|
25
|
+
|
|
26
|
+
## Step 2: Read state
|
|
27
|
+
|
|
28
|
+
Read `.design/STATE.md` and `.design/BRIEF.md`.
|
|
29
|
+
|
|
30
|
+
## Step 3: Calculate prettiness
|
|
31
|
+
|
|
32
|
+
Count completed phases out of 8. Calculate percentage.
|
|
33
|
+
|
|
34
|
+
## Step 4: Display progress
|
|
35
|
+
|
|
36
|
+
```
|
|
37
|
+
🎨 GSP — How Pretty Are We?
|
|
38
|
+
|
|
39
|
+
Project: {PROJECT_NAME}
|
|
40
|
+
Prettiness: ██████░░░░░░░░░░ 37.5% (3/8 phases)
|
|
41
|
+
|
|
42
|
+
Phase Progress:
|
|
43
|
+
✅ 1. Research — complete
|
|
44
|
+
✅ 2. Brand — complete
|
|
45
|
+
✅ 3. System — complete
|
|
46
|
+
⬜ 4. Design — pending ← you are here
|
|
47
|
+
⬜ 5. Spec — pending
|
|
48
|
+
⬜ 6. Review — pending
|
|
49
|
+
⬜ 7. Build — pending
|
|
50
|
+
⬜ 8. Launch — pending
|
|
51
|
+
|
|
52
|
+
Review Loops: {COUNT}
|
|
53
|
+
```
|
|
54
|
+
|
|
55
|
+
Use ✅ for complete, 🔄 for in-progress/needs-revision, ⬜ for pending.
|
|
56
|
+
|
|
57
|
+
## Step 5: Route next
|
|
58
|
+
|
|
59
|
+
Identify the next pending phase and suggest the command:
|
|
60
|
+
"Next up: Run `/gsp:{next_phase}` to continue."
|
|
61
|
+
</process>
|
|
@@ -0,0 +1,71 @@
|
|
|
1
|
+
---
|
|
2
|
+
name: gsp:research
|
|
3
|
+
description: Analyze design trends and competitive landscape for the project
|
|
4
|
+
allowed-tools:
|
|
5
|
+
- Read
|
|
6
|
+
- Write
|
|
7
|
+
- Bash
|
|
8
|
+
- Task
|
|
9
|
+
- WebSearch
|
|
10
|
+
- WebFetch
|
|
11
|
+
---
|
|
12
|
+
<context>
|
|
13
|
+
Phase 1 of the GSP design pipeline. Uses the Design Trend Synthesizer prompt to analyze industry trends, competitor positioning, and user expectation shifts.
|
|
14
|
+
</context>
|
|
15
|
+
|
|
16
|
+
<objective>
|
|
17
|
+
Research design trends for the project's industry and produce a comprehensive trends report.
|
|
18
|
+
|
|
19
|
+
**Input:** `.design/BRIEF.md`
|
|
20
|
+
**Output:** `.design/research/TRENDS.md`
|
|
21
|
+
**Agent:** `gsp-researcher`
|
|
22
|
+
</objective>
|
|
23
|
+
|
|
24
|
+
<execution_context>
|
|
25
|
+
@/Users/jubs/.claude/get-shit-pretty/prompts/07-design-trend-synthesizer.md
|
|
26
|
+
@/Users/jubs/.claude/get-shit-pretty/templates/phases/research.md
|
|
27
|
+
</execution_context>
|
|
28
|
+
|
|
29
|
+
<process>
|
|
30
|
+
## Step 1: Load context
|
|
31
|
+
|
|
32
|
+
Read `.design/BRIEF.md` to understand:
|
|
33
|
+
- Industry / sector
|
|
34
|
+
- Target audience
|
|
35
|
+
- Brand personality and positioning
|
|
36
|
+
- Competitive landscape (if mentioned)
|
|
37
|
+
|
|
38
|
+
If BRIEF.md doesn't exist, tell the user to run `/gsp:new-project` first.
|
|
39
|
+
|
|
40
|
+
## Step 2: Spawn researcher
|
|
41
|
+
|
|
42
|
+
Spawn the `gsp-researcher` agent with:
|
|
43
|
+
- The full BRIEF.md content
|
|
44
|
+
- The Design Trend Synthesizer prompt (07)
|
|
45
|
+
- The research output template
|
|
46
|
+
- Instruction to focus on the project's specific industry
|
|
47
|
+
|
|
48
|
+
The agent should:
|
|
49
|
+
1. Research current design trends for the industry
|
|
50
|
+
2. Identify 5 macro trends with real brand examples
|
|
51
|
+
3. Map competitor positioning (2x2)
|
|
52
|
+
4. Analyze user expectation shifts
|
|
53
|
+
5. Assess platform evolution (iOS, Material, Web)
|
|
54
|
+
6. Provide strategic recommendations
|
|
55
|
+
7. Define mood board direction
|
|
56
|
+
|
|
57
|
+
## Step 3: Write output
|
|
58
|
+
|
|
59
|
+
Write the completed research to `.design/research/TRENDS.md`.
|
|
60
|
+
|
|
61
|
+
## Step 4: Update state
|
|
62
|
+
|
|
63
|
+
Update `.design/STATE.md`:
|
|
64
|
+
- Set Phase 1 (Research) status to `complete`
|
|
65
|
+
- Record completion date
|
|
66
|
+
|
|
67
|
+
## Step 5: Route next
|
|
68
|
+
|
|
69
|
+
Display a summary of key findings and end with:
|
|
70
|
+
"Run `/gsp:brand` to create brand identity based on these insights."
|
|
71
|
+
</process>
|
|
@@ -0,0 +1,99 @@
|
|
|
1
|
+
---
|
|
2
|
+
name: gsp:review
|
|
3
|
+
description: Design critique (Nielsen's heuristics) + WCAG accessibility audit
|
|
4
|
+
allowed-tools:
|
|
5
|
+
- Read
|
|
6
|
+
- Write
|
|
7
|
+
- Bash
|
|
8
|
+
- Task
|
|
9
|
+
---
|
|
10
|
+
<context>
|
|
11
|
+
Phase 6 of the GSP design pipeline. Runs two agents in parallel: the Design Critique Partner (Nielsen's 10 heuristics) and the Accessibility Auditor (WCAG 2.2 AA). If critical issues are found, loops back to design/spec phases.
|
|
12
|
+
</context>
|
|
13
|
+
|
|
14
|
+
<objective>
|
|
15
|
+
Critique design quality and audit accessibility compliance.
|
|
16
|
+
|
|
17
|
+
**Input:** All prior artifacts
|
|
18
|
+
**Output:** `.design/review/CRITIQUE.md` + `.design/review/ACCESSIBILITY.md`
|
|
19
|
+
**Agents:** `gsp-critic` + `gsp-accessibility-auditor`
|
|
20
|
+
</objective>
|
|
21
|
+
|
|
22
|
+
<execution_context>
|
|
23
|
+
@/Users/jubs/.claude/get-shit-pretty/prompts/06-design-critique-partner.md
|
|
24
|
+
@/Users/jubs/.claude/get-shit-pretty/prompts/08-accessibility-auditor.md
|
|
25
|
+
@/Users/jubs/.claude/get-shit-pretty/templates/phases/review.md
|
|
26
|
+
@/Users/jubs/.claude/get-shit-pretty/references/nielsen-heuristics.md
|
|
27
|
+
@/Users/jubs/.claude/get-shit-pretty/references/wcag-checklist.md
|
|
28
|
+
</execution_context>
|
|
29
|
+
|
|
30
|
+
<process>
|
|
31
|
+
## Step 1: Load context
|
|
32
|
+
|
|
33
|
+
Read all prior artifacts:
|
|
34
|
+
- `.design/BRIEF.md`
|
|
35
|
+
- `.design/brand/IDENTITY.md`
|
|
36
|
+
- `.design/system/SYSTEM.md`
|
|
37
|
+
- `.design/screens/SCREENS.md`
|
|
38
|
+
- `.design/specs/FIGMA-SPECS.md`
|
|
39
|
+
|
|
40
|
+
If SCREENS.md or FIGMA-SPECS.md don't exist, tell the user to complete design and spec phases first.
|
|
41
|
+
|
|
42
|
+
## Step 2: Spawn critics (parallel)
|
|
43
|
+
|
|
44
|
+
**Agent 1: gsp-critic** — Design critique using:
|
|
45
|
+
- The Design Critique Partner prompt (06)
|
|
46
|
+
- Nielsen's 10 Heuristics reference
|
|
47
|
+
- All design artifacts
|
|
48
|
+
|
|
49
|
+
Should deliver:
|
|
50
|
+
1. Nielsen's 10 heuristics scored 1-5 with specific examples
|
|
51
|
+
2. Visual hierarchy analysis
|
|
52
|
+
3. Typography and color assessment
|
|
53
|
+
4. Usability evaluation
|
|
54
|
+
5. Cognitive load analysis
|
|
55
|
+
6. Prioritized fixes (Critical / Important / Polish)
|
|
56
|
+
7. 2 alternative redesign directions
|
|
57
|
+
|
|
58
|
+
**Agent 2: gsp-accessibility-auditor** — WCAG audit using:
|
|
59
|
+
- The Accessibility Auditor prompt (08)
|
|
60
|
+
- WCAG 2.2 AA checklist
|
|
61
|
+
- All design artifacts
|
|
62
|
+
|
|
63
|
+
Should deliver:
|
|
64
|
+
1. Perceivable, Operable, Understandable, Robust checklists
|
|
65
|
+
2. Mobile and cognitive accessibility checks
|
|
66
|
+
3. Violations list with severity and WCAG criteria
|
|
67
|
+
4. Remediation steps
|
|
68
|
+
5. Accessibility statement
|
|
69
|
+
|
|
70
|
+
## Step 3: Write output
|
|
71
|
+
|
|
72
|
+
1. Write critique to `.design/review/CRITIQUE.md`
|
|
73
|
+
2. Write audit to `.design/review/ACCESSIBILITY.md`
|
|
74
|
+
|
|
75
|
+
## Step 4: Assess results
|
|
76
|
+
|
|
77
|
+
Count critical and important issues across both reports.
|
|
78
|
+
|
|
79
|
+
**If critical issues found:**
|
|
80
|
+
- Display issues clearly
|
|
81
|
+
- Recommend looping back: "Run `/gsp:design` or `/gsp:spec` to address critical issues, then `/gsp:review` again."
|
|
82
|
+
- Update STATE.md review loop tracking
|
|
83
|
+
|
|
84
|
+
**If no critical issues:**
|
|
85
|
+
- Display positive summary with any polish items
|
|
86
|
+
- Proceed to build
|
|
87
|
+
|
|
88
|
+
## Step 5: Update state
|
|
89
|
+
|
|
90
|
+
Update `.design/STATE.md`:
|
|
91
|
+
- Set Phase 6 (Review) status to `complete` (or `needs-revision` if critical issues)
|
|
92
|
+
- Record review loop count
|
|
93
|
+
- Record completion date
|
|
94
|
+
|
|
95
|
+
## Step 6: Route next
|
|
96
|
+
|
|
97
|
+
If clean: "Run `/gsp:build` to translate designs to code."
|
|
98
|
+
If issues: "Address the critical issues above, then run `/gsp:review` again."
|
|
99
|
+
</process>
|