aiblueprint-cli 1.4.82 → 1.4.84

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.
Files changed (32) hide show
  1. package/README.md +4 -1
  2. package/agents-config/skills/apex/SKILL.md +61 -0
  3. package/agents-config/skills/app-icon/SKILL.md +161 -0
  4. package/agents-config/skills/appstore-connect/SKILL.md +2 -2
  5. package/agents-config/skills/{appstore-connect-setup/SKILL.md → appstore-connect/references/setup.md} +3 -6
  6. package/agents-config/skills/ios-testflight/SKILL.md +2 -2
  7. package/dist/cli.js +39 -38
  8. package/package.json +1 -1
  9. package/agents-config/skills/create-vitejs-app/SKILL.md +0 -273
  10. package/agents-config/skills/nextjs-add-prisma-db/SKILL.md +0 -137
  11. package/agents-config/skills/nextjs-setup-better-auth/SKILL.md +0 -174
  12. package/agents-config/skills/nextjs-setup-project/SKILL.md +0 -201
  13. package/agents-config/skills/saas-challenge-idea/SKILL.md +0 -136
  14. package/agents-config/skills/saas-create-architecture/SKILL.md +0 -243
  15. package/agents-config/skills/saas-create-headline/SKILL.md +0 -133
  16. package/agents-config/skills/saas-create-landing-copywritting/SKILL.md +0 -268
  17. package/agents-config/skills/saas-create-legals-docs/SKILL.md +0 -177
  18. package/agents-config/skills/saas-create-logos/SKILL.md +0 -241
  19. package/agents-config/skills/saas-create-prd/SKILL.md +0 -196
  20. package/agents-config/skills/saas-create-tasks/SKILL.md +0 -241
  21. package/agents-config/skills/saas-define-pricing/SKILL.md +0 -294
  22. package/agents-config/skills/saas-find-domain-name/SKILL.md +0 -191
  23. package/agents-config/skills/saas-implement-landing-page/SKILL.md +0 -258
  24. package/agents-config/skills/setup-tmux/SKILL.md +0 -165
  25. /package/agents-config/skills/{agents-managers → agents-manager}/SKILL.md +0 -0
  26. /package/agents-config/skills/{agents-managers → agents-manager}/references/agents.md +0 -0
  27. /package/agents-config/skills/{agents-managers → agents-manager}/references/context-management.md +0 -0
  28. /package/agents-config/skills/{agents-managers → agents-manager}/references/debugging-agents.md +0 -0
  29. /package/agents-config/skills/{agents-managers → agents-manager}/references/error-handling-and-recovery.md +0 -0
  30. /package/agents-config/skills/{agents-managers → agents-manager}/references/evaluation-and-testing.md +0 -0
  31. /package/agents-config/skills/{agents-managers → agents-manager}/references/orchestration-patterns.md +0 -0
  32. /package/agents-config/skills/{agents-managers → agents-manager}/references/writing-agent-prompts.md +0 -0
@@ -1,201 +0,0 @@
1
- ---
2
- name: nextjs-setup-project
3
- description: Setup a complete Next.js project with TypeScript, Tailwind, shadcn/ui, TanStack Query, and theme support
4
- ---
5
-
6
- <objective>
7
- Setup a production-ready Next.js project in the current directory with the complete AIBlueprint stack.
8
-
9
- This creates a Next.js 15 App Router project with TypeScript, Tailwind CSS, shadcn/ui components, TanStack Query, dark/light theme toggle, dialog manager, server toast, and a landing page with navbar.
10
- </objective>
11
-
12
- <context>
13
- Current directory: !`pwd`
14
- Directory contents: !`ls -la`
15
- Existing package.json: !`cat package.json 2>/dev/null | head -10 || echo "No package.json"`
16
- </context>
17
-
18
- <process>
19
- ## Phase 1: Project Initialization
20
-
21
- 1. **Create Next.js project**:
22
- ```bash
23
- npx create-next-app@latest . --typescript --tailwind --eslint --app --no-src-dir --import-alias "@/*" --yes
24
- ```
25
-
26
- 2. **Create src directory structure**:
27
- ```bash
28
- mkdir -p src/components src/lib src/features/landing
29
- ```
30
-
31
- 3. **Update tsconfig.json** - Configure path aliases:
32
- ```json
33
- {
34
- "compilerOptions": {
35
- "baseUrl": ".",
36
- "paths": {
37
- "@/*": ["./src/*"]
38
- }
39
- }
40
- }
41
- ```
42
-
43
- 4. **Clean default files** - Replace `app/page.tsx` with simple placeholder
44
-
45
- ## Phase 2: Core Libraries
46
-
47
- 5. **Install TanStack Query**:
48
- ```bash
49
- pnpm add @tanstack/react-query
50
- ```
51
-
52
- 6. **Create providers** - `app/providers.tsx`:
53
- ```typescript
54
- "use client";
55
-
56
- import { QueryClient, QueryClientProvider } from "@tanstack/react-query";
57
- import { ThemeProvider } from "next-themes";
58
- import { useState } from "react";
59
-
60
- export function Providers({ children }: { children: React.ReactNode }) {
61
- const [queryClient] = useState(
62
- () => new QueryClient({
63
- defaultOptions: {
64
- queries: { staleTime: 60 * 1000 },
65
- },
66
- })
67
- );
68
-
69
- return (
70
- <QueryClientProvider client={queryClient}>
71
- <ThemeProvider attribute="class" defaultTheme="system" enableSystem>
72
- {children}
73
- </ThemeProvider>
74
- </QueryClientProvider>
75
- );
76
- }
77
- ```
78
-
79
- 7. **Wrap layout with providers** - Update `app/layout.tsx`
80
-
81
- ## Phase 3: UI Components
82
-
83
- 8. **Initialize shadcn/ui**:
84
- ```bash
85
- pnpm dlx shadcn@latest init
86
- ```
87
- Configure paths: Components `@/components`, Utils `@/lib/utils`
88
-
89
- 9. **Install base components**:
90
- ```bash
91
- pnpm dlx shadcn@latest add button input card
92
- ```
93
-
94
- 10. **Install next-themes**:
95
- ```bash
96
- pnpm add next-themes
97
- ```
98
-
99
- 11. **Create theme toggle** - `src/components/theme-toggle.tsx`:
100
- - Use `useTheme` from next-themes
101
- - Handle mounted state for hydration
102
- - Toggle between dark and light
103
-
104
- ## Phase 4: UI Utilities
105
-
106
- 12. **Install dialog manager**:
107
- ```bash
108
- pnpm dlx shadcn@latest add https://ui.nowts.app/r/dialog-manager.json
109
- ```
110
-
111
- 13. **Install server toast**:
112
- ```bash
113
- pnpm dlx shadcn@latest add https://ui.nowts.app/r/server-toast.json
114
- ```
115
-
116
- 14. **Update layout** - Add to `app/layout.tsx`:
117
- ```typescript
118
- import { DialogManagerRenderer } from "@/lib/dialog-manager/dialog-manager-renderer";
119
- import { ServerToaster } from "@/components/server-toast/server-toast.server";
120
- import { Suspense } from "react";
121
-
122
- // Inside body after Providers:
123
- <DialogManagerRenderer />
124
- <Suspense>
125
- <ServerToaster />
126
- </Suspense>
127
- ```
128
-
129
- ## Phase 5: Landing Page
130
-
131
- 15. **Install navbar**:
132
- ```bash
133
- pnpm dlx shadcn@latest add https://coss.com/origin/r/comp-577.json
134
- ```
135
- Rename to `navbar.tsx` with semantic component name
136
-
137
- 16. **Install landing sections**:
138
- ```bash
139
- pnpm dlx shadcn@latest add https://shadcn-ui-blocks.akashmoradiya.com/r/hero-03.json
140
- pnpm dlx shadcn@latest add https://shadcn-ui-blocks.akashmoradiya.com/r/footer-05.json
141
- pnpm dlx shadcn@latest add https://shadcn-ui-blocks.akashmoradiya.com/r/faq-02.json
142
- ```
143
-
144
- 17. **Organize landing components**:
145
- - Move to `src/features/landing/`
146
- - Rename: `hero.tsx`, `footer.tsx`, `faq.tsx`
147
- - Update component names inside files
148
-
149
- 18. **Assemble landing page** - Update `app/page.tsx`:
150
- - Import all landing components
151
- - Navbar with logo, signin link, theme toggle
152
- - Hero, FAQ, Footer sections
153
-
154
- ## Phase 6: Quality & Documentation
155
-
156
- 19. **Add quality scripts** to `package.json`:
157
- ```json
158
- {
159
- "lint": "eslint .",
160
- "ts": "tsc --noEmit"
161
- }
162
- ```
163
-
164
- 20. **Run quality checks**:
165
- ```bash
166
- pnpm lint
167
- pnpm ts
168
- ```
169
- Fix any errors before completing
170
-
171
- 21. **Create CLAUDE.md** with:
172
- - Stack overview (Next.js 15, TanStack Query, shadcn/ui)
173
- - Directory structure explanation
174
- - UI Utilities section (dialog manager, server toast usage)
175
- - Project structure section
176
-
177
- 22. **Create AGENTS.md** as symlink:
178
- ```bash
179
- ln -s CLAUDE.md AGENTS.md
180
- ```
181
- </process>
182
-
183
- <output>
184
- **Files created**:
185
- - `app/` - Next.js App Router pages and layouts
186
- - `src/components/` - UI components including shadcn/ui
187
- - `src/lib/` - Utilities and helpers
188
- - `src/features/landing/` - Landing page components
189
- - `CLAUDE.md` - Project documentation
190
- - `AGENTS.md` - Symlink to CLAUDE.md
191
- </output>
192
-
193
- <success_criteria>
194
- - `pnpm dev` starts without errors
195
- - `pnpm lint` passes
196
- - `pnpm ts` passes (no TypeScript errors)
197
- - Landing page renders with navbar, hero, FAQ, footer
198
- - Theme toggle switches between dark/light mode
199
- - Dialog manager and server toast are integrated
200
- - CLAUDE.md contains complete stack documentation
201
- </success_criteria>
@@ -1,136 +0,0 @@
1
- ---
2
- name: saas-challenge-idea
3
- description: Challenge and validate a business idea through competitive research and brutally honest market analysis
4
- ---
5
-
6
- <objective>
7
- Challenge and validate a business idea through competitive research and market analysis.
8
-
9
- Provide brutally honest, evidence-based feedback on whether the idea has genuine market potential or should be abandoned/pivoted. Save founders from wasting time on bad ideas.
10
- </objective>
11
-
12
- <process>
13
- ## Phase 1: Gather the Idea
14
-
15
- 1. **If idea not provided**, ask the user:
16
- > "Tell me about your idea:
17
- > - What problem does it solve?
18
- > - Who is it for?
19
- > - What's your proposed solution?
20
- > - What makes it different?"
21
-
22
- 2. **Do NOT proceed** until you understand:
23
- - The specific problem being solved
24
- - The target user/customer
25
- - The proposed solution
26
- - Any claimed differentiation
27
-
28
- ## Phase 2: Research Competitors
29
-
30
- 3. **Use web search** to find existing solutions:
31
- - Search: "[problem] tool", "[solution] software", "[target user] platform"
32
- - Check: Product Hunt, Y Combinator portfolio, Crunchbase
33
- - Search: Reddit discussions, Alternative.to
34
-
35
- 4. **For each competitor**, document:
36
- - Name and URL
37
- - What they do (brief description)
38
- - Pricing model (free/paid/freemium)
39
- - Revenue signals: funding, team size, visible traction
40
- - Market positioning
41
-
42
- 5. **Find 5-10 competitors minimum** - both direct and indirect:
43
- - **Direct**: Same problem, same solution approach
44
- - **Indirect**: Same problem, different solution approach
45
-
46
- ## Phase 3: Analyze the Market
47
-
48
- 6. **Assess market saturation**:
49
- - 10+ competitors = Crowded (high risk, need strong differentiation)
50
- - 3-9 competitors = Moderate (validated demand, need positioning)
51
- - 1-2 competitors = Sparse (early market or weak demand)
52
- - 0 competitors = Usually no market (rarely means opportunity)
53
-
54
- 7. **Evaluate revenue signals**:
55
- - **Strong**: Paid tiers, funding raised, growing teams
56
- - **Weak**: Free only, abandoned projects, no monetization
57
-
58
- 8. **Check opportunity**:
59
- - Is there proof people PAY for this?
60
- - Is there room for another player?
61
- - What would truly differentiate this idea?
62
-
63
- ## Phase 4: Deliver Honest Feedback
64
-
65
- 9. **Generate validation report** in this format:
66
-
67
- ```markdown
68
- # Validation: [Idea Name]
69
-
70
- ## Quick Summary
71
- [1 sentence - what their idea is]
72
-
73
- ## What I Found
74
-
75
- **Competitors**:
76
- 1. [Name] - [What they do, pricing, traction signals]
77
- 2. [Name] - [What they do, pricing, traction signals]
78
- 3. [Continue for 3-5 key competitors]
79
-
80
- **Market**: [Crowded/Moderate/Sparse] - [1 sentence why]
81
- **Revenue proof**: [Yes/No] - [Evidence people pay for this]
82
-
83
- ## Viability Score: [X/10]
84
-
85
- [2-3 sentences explaining the score - be direct, no fluff]
86
-
87
- ## The Hard Truth
88
-
89
- **What's actually unique**: [What truly differentiates this, if anything]
90
- **What's NOT unique**: [What they claim is unique but competitors already do]
91
-
92
- **Biggest problems**:
93
- - [Problem 1]
94
- - [Problem 2]
95
- - [Problem 3]
96
-
97
- ## My Take
98
-
99
- **Would I build this?** [Yes/No/Maybe]
100
-
101
- **Why**: [Honest opinion in 1-2 sentences - don't hold back]
102
-
103
- **If you proceed**: [Specific advice on how to differentiate]
104
- **If you pivot**: [Specific pivot suggestion based on research]
105
-
106
- **Bottom line**: [Final direct recommendation - pursue, pivot, or kill]
107
- ```
108
- </process>
109
-
110
- <constraints>
111
- **BE BRUTALLY HONEST**:
112
- - Don't sugarcoat bad ideas
113
- - Be rude if necessary - founders need truth
114
- - Call out red flags directly
115
- - Use evidence from research, not opinions
116
-
117
- **ALWAYS RESEARCH FIRST**:
118
- - Never give feedback without finding real competitors
119
- - Challenge claims of "no competitors exist"
120
- - Focus on whether people actually PAY for solutions
121
-
122
- **RED FLAGS TO CALL OUT**:
123
- - "Uber for X" without unique insight
124
- - Solution looking for a problem
125
- - Targeting "everyone" (no clear user)
126
- - Ignoring dominant market leaders
127
- - Claims no competitors exist (almost always false)
128
- </constraints>
129
-
130
- <success_criteria>
131
- - Found and documented 5+ real competitors with web research
132
- - Provided specific pricing and revenue signals for competitors
133
- - Gave honest viability score with clear reasoning
134
- - Called out what's NOT unique about the idea
135
- - Provided actionable recommendation: pursue, pivot, or kill
136
- </success_criteria>
@@ -1,243 +0,0 @@
1
- ---
2
- name: saas-create-architecture
3
- description: Design technical architecture for a Next.js SaaS application based on PRD requirements
4
- ---
5
-
6
- <objective>
7
- Design the complete technical architecture for a Next.js application based on the Product Requirements Document.
8
-
9
- Define tool selection, technology stack decisions, infrastructure choices, and architecture patterns. Challenge every decision to ensure the best fit for specific needs. This comes AFTER the PRD - the PRD defines WHAT to build, this defines HOW to build it.
10
- </objective>
11
-
12
- <context>
13
- Tools reference: @https://github.com/Melvynx/aiblueprint/blob/main/ai-coding/tools.md
14
- </context>
15
-
16
- <process>
17
- ## Phase 1: Read PRD
18
-
19
- 1. **Ask for PRD location**:
20
- > "Where is your PRD? Please provide the file path or paste the content."
21
-
22
- 2. **Extract from PRD**:
23
- - **Core Features**: What features drive architecture decisions
24
- - **User Personas**: Who uses this (impacts willingness to pay for tools)
25
- - **Success Metrics**: Performance/scale targets
26
- - **Timeline**: MVP timeline (impacts complexity decisions)
27
- - **Out of Scope**: What WON'T be built in v1
28
-
29
- 3. **Map technical implications**:
30
- - Real-time features → Need WebSocket/real-time database
31
- - File uploads → Need storage solution
32
- - AI features → Need LLM integration
33
- - Multi-user/teams → Need robust auth and permissions
34
- - High traffic → Need caching and performance optimization
35
-
36
- ## Phase 2: Read Tools Reference
37
-
38
- 4. **Fetch tools.md** - Read the complete tool reference at:
39
- https://github.com/Melvynx/aiblueprint/blob/main/ai-coding/tools.md
40
-
41
- 5. **Use as primary source** for tool recommendations
42
-
43
- ## Phase 3: Information Gathering
44
-
45
- 6. **Ask questions progressively** (don't ask all at once):
46
-
47
- **Frontend Requirements**:
48
- - UI component strategy? (shadcn/ui, custom, other)
49
- - Client-side state management beyond React?
50
- - URL state requirements? (filters, search, pagination)
51
- - Data fetching pattern? (Server Components, TanStack Query, mix)
52
-
53
- **Backend & API**:
54
- - API pattern preference? (Server Actions, API Routes, both)
55
- - External API access needed?
56
- - Validation/security layer? (Zod, rate limiting)
57
- - Third-party API integrations?
58
-
59
- **Authentication & Authorization**:
60
- - Authentication method? (email/password, OAuth, magic link, OTP)
61
- - Organizations/team features?
62
- - Permission model? (simple roles, complex RBAC, none)
63
-
64
- **Database & Data**:
65
- - Data type? (relational, document, real-time, files)
66
- - SQL vs NoSQL preference?
67
- - Real-time features? (live updates, collaborative editing)
68
- - Expected data volume and query patterns?
69
- - ORM preference? (Prisma, Drizzle)
70
-
71
- **AI & LLM** (if applicable):
72
- - AI features needed? (chat, completion, embeddings, vision)
73
- - LLM providers? (OpenAI, Anthropic, multiple)
74
- - Streaming responses needed?
75
-
76
- **Infrastructure**:
77
- - Deployment platform? (default: Vercel)
78
- - Expected traffic/scale?
79
- - Background jobs or cron tasks?
80
- - Budget constraints?
81
-
82
- **Additional**:
83
- - Email sending? (transactional, marketing)
84
- - File uploads/storage?
85
- - Payment processing?
86
- - Analytics/monitoring?
87
-
88
- ## Phase 4: Challenge Every Decision
89
-
90
- 7. **For EVERY tool choice**, challenge with:
91
-
92
- **"Why This?" Challenge**:
93
- - Why this tool specifically?
94
- - What are the alternatives?
95
- - What's the trade-off?
96
-
97
- **"Do You Really Need It?" Challenge**:
98
- - Can you ship without this?
99
- - Is this solving a real problem or hypothetical future problem?
100
- - What's the simplest solution that could work?
101
-
102
- **"What's The Cost?" Challenge**:
103
- - Learning curve for the team?
104
- - Monetary cost at scale?
105
- - Maintenance burden?
106
-
107
- ## Phase 5: Generate Architecture Document
108
-
109
- 8. **Only after gathering ALL information**, create ARCHI.md:
110
-
111
- ```markdown
112
- # Technical Architecture: [Project Name]
113
-
114
- ## Architecture Overview
115
-
116
- **Philosophy**: [2-3 sentences on approach and principles]
117
-
118
- **Tech Stack Summary**:
119
- - Framework: [e.g., Next.js 15 with App Router]
120
- - Deployment: [e.g., Vercel serverless]
121
- - Database: [e.g., Neon PostgreSQL with Prisma]
122
- - Authentication: [e.g., Better Auth with email/password]
123
-
124
- ## Frontend Architecture
125
-
126
- ### Core Stack
127
- - **Framework**: [Tool + version]
128
- - **Why**: [Specific reason]
129
- - **Trade-off**: [What you gain vs sacrifice]
130
-
131
- - **UI Components**: [Tool]
132
- - **Why**: [Reason]
133
- - **Trade-off**: [Analysis]
134
-
135
- ### State Management
136
- - **Global State**: [Tool or "React hooks only"]
137
- - **Server State**: [Tool + pattern]
138
- - **URL State**: [Tool or "Not needed"]
139
-
140
- ### Data Fetching Strategy
141
- [Describe pattern - Server Components first, Client when needed, etc.]
142
-
143
- ## Backend Architecture
144
-
145
- ### API Layer
146
- - **Pattern**: [Server Actions / API Routes / Both]
147
- - **Validation**: [Tool]
148
- - **Security**: [Rate limiting, CORS config]
149
-
150
- ### Authentication & Authorization
151
- - **Provider**: [Tool] - **Why**: [Reason]
152
- - **Method**: [Email/password, OAuth, etc.]
153
- - **Authorization**: [Permission model description]
154
-
155
- ### Database & Data Layer
156
- - **Database**: [Tool + type] - **Why**: [Reason]
157
- - **ORM**: [Tool] - **Why**: [Justification]
158
- - **Real-time**: [Tool/approach if needed]
159
-
160
- ### AI Integration (if applicable)
161
- - **SDK**: [Tool]
162
- - **Providers**: [Primary + fallback]
163
- - **Use Cases**: [Feature → model mapping]
164
-
165
- ## Infrastructure & Deployment
166
-
167
- - **Platform**: [Platform] - **Why**: [Reason]
168
- - **Region Strategy**: [Single/multi/edge]
169
- - **Background Jobs**: [Tool or "Not needed"]
170
- - **Monitoring**: [Tools for app monitoring, errors, analytics]
171
-
172
- ## Additional Services
173
-
174
- - **Email**: [Provider or "Not needed"]
175
- - **Storage**: [Provider or "Not needed"]
176
- - **Payments**: [Provider or "Not needed"]
177
-
178
- ## Architecture Decision Records
179
-
180
- ### ADR-001: [Key Decision Title]
181
- - **Context**: [Why decision was needed]
182
- - **Decision**: [What was decided]
183
- - **Alternatives**: [What else was evaluated]
184
- - **Rationale**: [Why this choice]
185
- - **Consequences**: [Trade-offs]
186
-
187
- ## Folder Structure
188
- [Project structure diagram]
189
-
190
- ## Cost Estimation
191
-
192
- ### Monthly at Scale
193
- - Database: $[amount]
194
- - Hosting: $[amount]
195
- - [Other services]
196
- - **Total**: ~$[total]/month at [expected scale]
197
-
198
- ### Free Tier Limits
199
- - [Service]: [Limits]
200
-
201
- ## Implementation Priority
202
-
203
- ### Phase 1: Foundation
204
- 1. [Setup task]
205
- 2. [Setup task]
206
-
207
- ### Phase 2: Core Features
208
- 1. [Feature]
209
- 2. [Feature]
210
-
211
- ### Phase 3: Enhancement
212
- 1. [Enhancement]
213
- ```
214
- </process>
215
-
216
- <constraints>
217
- **MANDATORY BEFORE WRITING**:
218
- - PRD read completely
219
- - Tools reference read
220
- - ALL 8 information areas answered
221
- - Every tool choice challenged
222
-
223
- **CHECKLIST**:
224
- - ✅ PRD understanding complete
225
- - ✅ Frontend stack decided (aligned with PRD UI needs)
226
- - ✅ Backend stack decided (supporting PRD features)
227
- - ✅ Auth strategy decided (matching PRD user personas)
228
- - ✅ Database choice made (based on PRD data requirements)
229
- - ✅ AI requirements specified (from PRD AI features)
230
- - ✅ Infrastructure decided (based on PRD timeline/metrics)
231
- - ✅ Additional services identified (driven by PRD features)
232
-
233
- **IF ANY MISSING**: Ask more questions first
234
- </constraints>
235
-
236
- <success_criteria>
237
- - Every technology choice has clear "Why" and "Trade-off"
238
- - Architecture aligns with ALL PRD features
239
- - Cost estimation included
240
- - Implementation phases defined
241
- - Architecture Decision Records for key choices
242
- - ARCHI.md saved in same directory as PRD
243
- </success_criteria>
@@ -1,133 +0,0 @@
1
- ---
2
- name: saas-create-headline
3
- description: Generate clickbait headlines that sound too good to be true for your landing page
4
- arguments:
5
- - name: product-description
6
- description: Brief description of your product and what it does
7
- required: true
8
- ---
9
-
10
- <objective>
11
- Generate 10+ headline options for a landing page that make visitors say "Wait… what?"
12
-
13
- Headlines should sound slightly clickbait - bold enough to spark curiosity and open a loop in the reader's brain.
14
- </objective>
15
-
16
- <process>
17
- ## Phase 1: Understand the Product
18
-
19
- **Product**: #$ARGUMENTS.product-description
20
-
21
- 1. **If project exists**, find and read:
22
- - `**/PRD.md` or `**/*prd*.md`
23
- - `**/*marketing*/**/*.md`
24
- - Existing landing page files
25
-
26
- 2. **Extract**:
27
- - Core transformation (before → after)
28
- - Main pain point
29
- - Target audience
30
- - Key differentiator
31
-
32
- ## Phase 2: Generate Headlines
33
-
34
- 4. **Apply the TechCrunch test**: If TechCrunch wrote an article about this product, what would the headline be?
35
-
36
- 5. **Generate 10+ headlines** using these patterns:
37
-
38
- **Result + Timeline**:
39
-
40
- - "Launch your SaaS in days, not months"
41
- - "Learn to code in weeks, not years"
42
-
43
- **Transformation Promise**:
44
-
45
- - "From zero to shipped in one weekend"
46
- - "Turn your idea into a live product tonight"
47
-
48
- **Pain Elimination**:
49
-
50
- - "Never write boilerplate code again"
51
- - "Stop wasting months building the same features"
52
-
53
- **Social Proof + Promise**:
54
-
55
- - "Join 7,000+ developers who ship faster"
56
- - "The secret 10,000 entrepreneurs use to launch"
57
-
58
- **Specific Number**:
59
-
60
- - "Get 10x more done with half the work"
61
- - "Build a $10K/month SaaS in 30 days"
62
-
63
- **Question Format**:
64
-
65
- - "What if you could launch this weekend?"
66
- - "Ready to ship your startup in 48 hours?"
67
-
68
- **Too-Good-To-Be-True**:
69
-
70
- - "All the code you need. None of the headaches."
71
- - "Build like a senior dev on day one."
72
-
73
- ## Phase 3: Output
74
-
75
- 6. **Create HEADLINE.md** with this exact format:
76
-
77
- ```markdown
78
- # Headline Options for [Product Name]
79
-
80
- ## Recommended
81
-
82
- **[Best headline]**
83
- _Why: [One sentence explaining why this works]_
84
-
85
- ## All Options
86
-
87
- 1. [Headline]
88
- 2. [Headline]
89
- 3. [Headline]
90
- 4. [Headline]
91
- 5. [Headline]
92
- 6. [Headline]
93
- 7. [Headline]
94
- 8. [Headline]
95
- 9. [Headline]
96
- 10. [Headline]
97
-
98
- ## Testing Advice
99
-
100
- Send to 5 friends. After 24h, ask which they remember. Pick that one.
101
- ```
102
-
103
- </process>
104
-
105
- <constraints>
106
- **GOOD HEADLINES**:
107
- - Sound too good to be true
108
- - Open a curiosity loop
109
- - Focus on results, not features
110
- - Are specific (numbers, timeframes)
111
- - Make people scroll to learn more
112
-
113
- **BAD HEADLINES**:
114
-
115
- - Generic: "The best solution for X"
116
- - Feature-focused: "We have AI-powered..."
117
- - Boring: "Welcome to our product"
118
- - Safe: "A tool for developers"
119
- - Vague: "Build better software"
120
-
121
- **NEVER**:
122
-
123
- - Write "clean" or "safe" copy that nobody remembers
124
- - Use corporate jargon
125
- - Lead with features
126
- - Be forgettable
127
- </constraints>
128
-
129
- <output>
130
- **File**: HEADLINE.md in same directory as PRD (or current directory)
131
- **Content**: 10+ headline options with clear recommendation
132
- **Format**: Simple list, no fluff
133
- </output>