npm - aiblueprint-cli - Versions diffs - 1.4.59 → 1.4.61 - Mend

aiblueprint-cli 1.4.59 → 1.4.61

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 (184) hide show

package/agents-config/commands/prompts/saas-create-logos.md ADDED Viewed

@@ -0,0 +1,240 @@
+---
+description: Generate 20 abstract SVG logo variations with brainstorming and HTML showcase
+arguments:
+  - name: app-name
+    description: Name of your app/product
+    required: true
+  - name: app-description
+    description: What does your app do? Who is it for?
+    required: true
+---
+<objective>
+Generate 20 unique, minimalist SVG logo variations inspired by iconic logos (Stripe, Google, Claude, ChatGPT, Dub, Excalidraw).
+Output:
+1. `specs/logos/brainstorm.md` - Logo concepts and rationale
+2. `specs/logos/showcase-v1.html` - First batch of 20 logos (increment version for iterations)
+   </objective>
+<context>
+**App Name**: #$ARGUMENTS.app-name
+**Description**: #$ARGUMENTS.app-description
+</context>
+<process>
+## Phase 0: Ask Clarifying Questions
+**BEFORE doing anything**, ask the user these questions using AskUserQuestion:
+1. **Logo Style**:
+   - Lettermark (stylized letter like Stripe's "S")
+   - Abstract symbol (like Claude's icon)
+   - Geometric shape (like Airbnb)
+   - Wordmark (full name stylized)
+2. **Brand Personality**:
+   - Modern & Techy
+   - Playful & Friendly
+   - Professional & Corporate
+   - Bold & Edgy
+   - Minimal & Clean
+3. **Visual Inspiration** (can select multiple):
+   - Stripe (depth, gradients)
+   - Google (simple geometry)
+   - Linear (sharp, modern)
+   - Notion (clean, minimal)
+   - Discord (friendly, rounded)
+Wait for user answers before proceeding.
+## Phase 1: Analyze & Brainstorm
+1. **Study the app name** with user preferences in mind:
+   - Break down letters, sounds, meaning
+   - Find visual metaphors (e.g., "Mail" → envelope, arrow, @ symbol)
+   - Match to selected brand personality
+2. **Apply user's style choices**:
+   - Use selected logo style as primary direction
+   - Incorporate visual inspiration references
+   - Align with brand personality
+3. **Create brainstorm.md**:
+```markdown
+# Logo Brainstorm: [App Name]
+## Brand Analysis
+- **Name meaning**: [What does the name evoke?]
+- **Core function**: [What does the app do?]
+- **Target vibe**: [Modern/Playful/Professional/Bold/Minimal]
+## Visual Concepts
+### Lettermark Ideas
+1. [First letter stylized - describe approach]
+2. [Initials combined - describe approach]
+3. [Full name simplified - describe approach]
+### Symbol Ideas
+1. [Abstract shape representing core function]
+2. [Geometric interpretation of name]
+3. [Metaphor visualization]
+### Style Direction
+- **Colors**: Black/white primary (can add accent later)
+- **Style**: Geometric, clean, SVG-optimized
+- **Inspiration**: [Which logos to draw from]
+## 20 Variations Plan
+1-5: Lettermark variations
+6-10: Abstract symbols
+11-15: Geometric shapes
+16-20: Experimental/unique
+```
+## Phase 2: Generate HTML Showcase
+4. **Check for existing versions**:
+   - Look for `specs/logos/showcase-v*.html`
+   - Increment version number (v1, v2, v3...)
+   - If v1 exists, create v2, etc.
+5. **Create showcase-vX.html** with this structure:
+```html
+<!DOCTYPE html>
+<html lang="en">
+  <head>
+    <meta charset="UTF-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>[App Name] - Logos</title>
+    <style>
+      * {
+        margin: 0;
+        padding: 0;
+        box-sizing: border-box;
+      }
+      body {
+        background: #fff;
+      }
+      .grid {
+        display: grid;
+        grid-template-columns: repeat(5, 1fr);
+      }
+      .cell {
+        aspect-ratio: 1;
+        border: 1px solid #e5e5e5;
+        display: flex;
+        align-items: center;
+        justify-content: center;
+        padding: 24px;
+      }
+      .cell svg {
+        width: 100%;
+        height: 100%;
+        max-width: 80px;
+        max-height: 80px;
+      }
+    </style>
+  </head>
+  <body>
+    <div class="grid">
+      <div class="cell">
+        <svg
+          viewBox="0 0 500 500"
+          fill="none"
+          xmlns="http://www.w3.org/2000/svg"
+        >
+          <!-- SVG content here - use #000 for fills -->
+        </svg>
+      </div>
+      <!-- ... 19 more cells -->
+    </div>
+  </body>
+</html>
+```
+## Phase 3: Design 20 Logos
+6. **Generate each SVG** following these rules:
+   - ViewBox: `0 0 500 500`
+   - Colors: Black (`#000`) on white background
+   - No gradients in base version
+   - Clean paths, minimal complexity
+   - Must work at 16px and 512px
+7. **Logo categories** (adapt based on user's style choice):
+   - **1-5**: Primary style (user's selected logo style)
+   - **6-10**: Secondary variations
+   - **11-15**: Alternative interpretations
+   - **16-20**: Experimental (unique, creative approaches)
+8. **Document in brainstorm.md** (not in HTML):
+   - Each logo's concept and rationale
+   - Why it fits the brand
+## Phase 4: Iteration Support
+9. **After generating**, ask user:
+   > "Which logos do you like? I can create v2 with variations of your favorites."
+10. **For next iteration**: - Note favorites from previous version - Create `showcase-v2.html` with 20 new variations inspired by favorites - Each version builds on user feedback
+    </process>
+<constraints>
+**SVG RULES**:
+- Pure SVG, no raster images
+- Black (`#000`) fills only
+- Simple paths, no complex effects
+- Must scale from favicon to billboard
+**DESIGN RULES**:
+- Inspired by: Stripe, Google, Claude, ChatGPT, Dub
+- Minimalist, geometric, modern
+- Recognizable at small sizes
+- No text (unless lettermark)
+**HTML RULES**:
+- Light mode only (white background)
+- 5x4 grid of logos
+- NO titles, NO descriptions, NO text
+- Just logos in cells separated by borders
+- Clean, minimal presentation
+**OUTPUT RULES**:
+- Create `specs/logos/` directory first
+- Save `brainstorm.md` with concepts and rationale
+- Save `showcase-vX.html` with just the logo grid
+- ONE HTML file per version
+  </constraints>
+<output>
+**Files created**:
+1. `specs/logos/brainstorm.md` - Logo concepts and rationale (all text here)
+2. `specs/logos/showcase-v1.html` - Clean 5x4 grid of logos (no text)
+**HTML structure**:
+- 5x4 grid (20 logos)
+- Each cell: just the SVG, no text
+- Light gray borders between cells
+- White background
+**Iteration flow**:
+- v1: Initial 20 based on user preferences
+- v2+: Refinements based on user favorites
+  </output>

package/agents-config/commands/prompts/saas-create-prd.md ADDED Viewed

@@ -0,0 +1,195 @@
+---
+description: Create a lean Product Requirements Document through interactive discovery conversation
+---
+<objective>
+Create a focused Product Requirements Document (PRD) for an MVP through iterative conversation.
+Guide the user through defining product vision, user personas, core features, and success metrics. Keep the PRD minimal and focused on essential features that solve real problems. This is the FOUNDATION - the PRD comes FIRST, then ARCHI, then implementation.
+</objective>
+<process>
+## Phase 1: Information Discovery
+1. **Do NOT write anything until you have answers to ALL 5 areas below**
+Ask questions progressively (2-3 at a time, not all at once) until you have complete clarity:
+### Area 1: Problem & Vision
+- What problem does this product solve?
+- Who experiences this problem most acutely?
+- What makes this solution unique or different?
+- What does success look like?
+### Area 2: Target Users
+- Who are the primary users? (1-3 personas max)
+- What are their key characteristics? (role, context, pain points)
+- What motivates them to use this product?
+- What would make them stop using it?
+### Area 3: Core Features
+- What is the ONE thing this product must do? (critical feature)
+- What 2-3 features support that core capability?
+- What features are nice-to-have but not critical for MVP?
+- How will users accomplish their primary goal?
+### Area 4: Success Metrics
+- How will you measure if this product is working?
+- What user behavior indicates success?
+- What business metrics matter most?
+### Area 5: Constraints & Scope
+- What technical constraints exist?
+- What timeline are you working with?
+- What resources are available?
+- What are you explicitly NOT building in v1?
+## Phase 2: Verify Completeness
+2. **Before generating PRD**, confirm you have:
+- ✅ Core problem clearly defined with specific user experiencing it
+- ✅ 1-3 specific personas with roles, pain points, motivations
+- ✅ THE ONE critical feature + 2-4 supporting must-have features
+- ✅ Specific, measurable metrics (not vague goals)
+- ✅ Clear list of what is NOT being built in v1
+- ✅ Timeline and constraints understood
+**IF ANY MISSING**: Ask more questions first. Do NOT generate PRD with gaps.
+## Phase 3: Generate PRD
+3. **Create PRD.md** with this structure:
+```markdown
+# Product Requirements Document: [Product Name]
+## Product Vision
+**Problem Statement**
+[2-3 sentences describing the core problem]
+**Solution**
+[2-3 sentences describing how this product solves it]
+**Success Criteria**
+- [Metric 1: quantitative measure]
+- [Metric 2: user behavior indicator]
+- [Metric 3: business outcome]
+## Target Users
+### Primary Persona: [Name]
+- **Role**: [User role/context]
+- **Pain Points**:
+  - [Pain point 1]
+  - [Pain point 2]
+- **Motivations**: [What drives them]
+- **Goals**: [What they want to accomplish]
+### Secondary Persona: [Name] (if applicable)
+- **Role**: [User role/context]
+- **Pain Points**: [Key challenges]
+- **Motivations**: [What drives adoption]
+## Core Features (MVP)
+### Must-Have Features
+#### 1. [Feature Name]
+**Description**: [What this feature does - 2-3 sentences]
+**User Value**: [Why this matters to users]
+**Success Metric**: [How to measure if working]
+#### 2. [Feature Name]
+**Description**: [Feature explanation]
+**User Value**: [Benefit to users]
+**Success Metric**: [Measurement criteria]
+#### 3. [Feature Name]
+**Description**: [Feature explanation]
+**User Value**: [Benefit to users]
+**Success Metric**: [Measurement criteria]
+### Should-Have Features (Post-MVP)
+- [Feature 1]: [Brief description]
+- [Feature 2]: [Brief description]
+## User Flows
+### Primary User Journey
+1. [Step 1: User action]
+2. [Step 2: User action]
+3. [Step 3: User action]
+4. [Step 4: Outcome achieved]
+## Out of Scope (v1)
+Explicitly NOT building in MVP:
+- [Feature/capability 1]
+- [Feature/capability 2]
+- [Feature/capability 3]
+## Open Questions
+- [Question 1 needing resolution]
+- [Question 2 requiring validation]
+## Success Metrics
+**Primary Metrics**:
+- [Metric 1]: [Target/goal]
+- [Metric 2]: [Target/goal]
+**Secondary Metrics**:
+- [Metric 3]: [Target/goal]
+## Timeline & Milestones
+- **MVP Completion**: [Target timeframe]
+- **First User Testing**: [Target]
+- **Launch**: [Target]
+```
+## Phase 4: Save and Next Steps
+4. **Save PRD.md** in project directory
+5. **Suggest next steps**:
+   > "PRD created! Next steps in order:
+   > 1. `/prompts/saas-create-architecture` - Design technical architecture
+   > 2. `/prompts/saas-define-pricing` - Define pricing strategy
+   > 3. `/prompts/saas-create-landing-copy` - Create landing page copy"
+</process>
+<constraints>
+**DISCOVERY RULES**:
+- NEVER write PRD until ALL 5 information areas are covered
+- Ask 2-3 questions at a time, not all at once
+- Dig deeper based on responses - don't accept vague answers
+- Push for specific metrics, not goals like "increase engagement"
+**MVP FOCUS**:
+- Maximum 3-5 must-have features
+- If user lists 10 features, help them prioritize to 3-5
+- "Out of Scope" section is critical - define boundaries clearly
+- Post-MVP features go in "Should-Have" section
+**OUTPUT RULES**:
+- Target 2-3 pages max, not a 20-page spec
+- Every feature needs Description, User Value, Success Metric
+- Be specific and actionable, not vague
+**CONVERSATION STYLE**:
+- Keep it conversational, not interrogation
+- Adapt questions based on their industry/context
+- Help users think through trade-offs
+</constraints>
+<success_criteria>
+- All 5 information areas have complete answers
+- 1-3 specific personas defined with pain points
+- 3-5 must-have features clearly described
+- Each feature has measurable success metric
+- Out of Scope section explicitly defines boundaries
+- PRD saved as PRD.md in project directory
+- Next steps provided for workflow continuation
+</success_criteria>

package/agents-config/commands/prompts/saas-create-tasks.md ADDED Viewed

@@ -0,0 +1,240 @@
+---
+description: Generate implementation task files from PRD and Architecture documents
+---
+<objective>
+Generate a structured set of implementation task files from existing PRD and Architecture documents.
+Create well-scoped, coherent tasks that an AI agent can execute autonomously. Each task represents 1-3 hours of focused work with clear deliverables and success criteria. This comes AFTER PRD and ARCHI - both must exist first.
+</objective>
+<context>
+Current project: !`ls -la`
+Existing specs folder: !`ls specs/ 2>/dev/null || echo "No specs folder"`
+Package.json: !`cat package.json 2>/dev/null | head -20 || echo "No package.json"`
+</context>
+<process>
+## Phase 1: Understanding
+1. **Ask for project location**:
+   > "Where is your project located? Are you starting from scratch or using a boilerplate?"
+2. **Explore existing codebase**:
+   - Read `package.json` for dependencies
+   - Check folder structure (`app/`, `src/`, `components/`)
+   - Identify what's already implemented
+3. **Document current state**:
+   - ✅ Already implemented: [list]
+   - 🚧 Partially implemented: [list]
+   - ❌ Not yet built: [list]
+4. **Ask for PRD location**, then read completely:
+   - Extract all features
+   - Note user flows
+   - Identify MVP scope vs post-MVP
+5. **Ask for ARCHI location**, then read completely:
+   - Note all technology choices
+   - Understand patterns to follow
+   - Identify integration points
+## Phase 2: Deep Analysis
+6. **Map features to implementation**:
+   - [Feature 1] → Database models, UI, API, validation needed
+   - [Feature 2] → Real-time, file storage, background jobs needed
+   - [Feature 3] → Auth, permissions, notifications needed
+7. **Create dependency graph**:
+```
+Foundation Layer (Database, Auth, Base UI)
+    ↓
+Core Features Layer (Primary user functionality)
+    ↓
+Integration Layer (Third-party services)
+    ↓
+Polish Layer (Error handling, optimization)
+```
+8. **Identify what can be parallelized** vs what is sequential
+## Phase 3: Verify Before Creating
+9. **Confirm checklist complete**:
+   - ✅ Codebase understood
+   - ✅ All PRD features extracted
+   - ✅ All ARCHI decisions understood
+   - ✅ Dependencies mapped
+   - ✅ Foundation identified
+   - ✅ Integration points clear
+   - ✅ Success criteria defined
+**IF ANY MISSING**: Do more analysis first. Do NOT create tasks with gaps.
+## Phase 4: Generate Task Files
+10. **Create folder structure**:
+    ```bash
+    mkdir -p specs/01-mvp
+    ```
+11. **Create individual task files** in `specs/01-mvp/`:
+    - Naming: `01-task-name.md`, `02-task-name.md`
+    - Use 2-digit numbering with kebab-case names
+12. **Task file structure**:
+```markdown
+# Task [Number]: [Action-Oriented Title]
+## Context
+[Connection to PRD feature and overall vision]
+## Scope
+[Specific deliverables for this task]
+## Implementation Details
+### Files to Create/Modify
+- `path/to/file.ts` - [Purpose]
+### Key Functionality
+- [What the code should do]
+### Technologies Used
+- [From ARCHI with how it's used]
+### Architectural Patterns
+[Patterns from ARCHI to follow]
+## Success Criteria
+- [ ] [Testable outcome 1]
+- [ ] [Testable outcome 2]
+- [ ] [Testable outcome 3]
+## Testing & Validation
+### Manual Testing Steps
+1. [Step]
+2. [Step]
+### Edge Cases
+- [Edge case]
+## Dependencies
+**Must complete first**:
+- Task [N]: [Name]
+**Blocks**:
+- Task [X]: [Name]
+## Related Documentation
+- **PRD**: [Reference]
+- **ARCHI**: [Reference]
+---
+**Estimated Time**: [1-3 hours]
+**Phase**: [Foundation / Core / Integration / Polish]
+```
+13. **Create README overview** at `specs/README.md`:
+```markdown
+# Implementation Tasks Overview
+## Project Summary
+**From PRD**: [Summary]
+**Tech Stack**: [From ARCHI]
+**Current State**: [What exists]
+## Task Execution Guidelines
+- Read complete task before starting
+- Check dependencies are met
+- Follow ARCHI patterns
+- Validate against success criteria
+## MVP Tasks (specs/01-mvp/)
+### Phase 1: Foundation
+- [ ] `01-name.md` - [Description]
+- [ ] `02-name.md` - [Description]
+### Phase 2: Core Features
+- [ ] `03-name.md` - [Description]
+- [ ] `04-name.md` - [Description]
+### Phase 3: Integration
+- [ ] `05-name.md` - [Description]
+### Phase 4: Polish
+- [ ] `06-name.md` - [Description]
+## Dependency Map
+[Visual dependency graph]
+## PRD Coverage
+- ✅ [Feature]: Task [N]
+- ✅ [Feature]: Task [N]
+## Total Estimated Time: [X-Y hours]
+```
+</process>
+<constraints>
+**TASK SIZING**:
+❌ **Too Small**:
+- "Add TypeScript type"
+- "Create button component"
+- "Add validation to field"
+❌ **Too Large**:
+- "Implement complete auth system"
+- "Build entire dashboard"
+✅ **Just Right** (1-3 hours):
+- "Create user authentication flow with email/password and session management"
+- "Build dashboard layout with navigation and responsive design"
+- "Implement checkout with Stripe integration and webhooks"
+**OUTPUT FORMAT**:
+- NEVER create single document with all tasks
+- ALWAYS create individual files in `specs/01-mvp/`
+- ALWAYS create `specs/README.md` overview
+- Use 2-digit numbering (01-, 02-, 03-)
+**QUALITY**:
+- Every task links to PRD feature
+- Every task follows ARCHI patterns
+- Every task has testable success criteria
+- Every task has clear dependencies
+</constraints>
+<output>
+**Created files**:
+- `specs/README.md` - Overview with dependency map
+- `specs/01-mvp/01-name.md` through `specs/01-mvp/[N]-name.md` - Individual task files
+**Each task file contains**:
+- Context connecting to PRD
+- Specific scope and deliverables
+- Files to create/modify
+- Technologies from ARCHI
+- Testable success criteria
+- Dependencies (before and blocks)
+- Time estimate
+</output>
+<success_criteria>
+- All PRD features mapped to specific tasks
+- All ARCHI decisions reflected in task patterns
+- Tasks ordered correctly by dependencies
+- Each task is 1-3 hours of work
+- Each task has clear success criteria
+- Individual files created in `specs/01-mvp/`
+- README created with complete overview
+- No orphan tasks (everything connected)
+</success_criteria>