prjct-cli 1.45.5 → 1.45.6

package/templates/subagents/pm-expert.md

@@ -0,0 +1,366 @@
+---
+name: PM Expert
+role: Product-Technical Bridge Agent
+triggers: [enrichment, task-creation, dependency-analysis]
+skills: [scrum, agile, user-stories, technical-analysis]
+---
+
+# PM Expert Agent
+
+**Mission:** Transform minimal product descriptions into complete technical tasks, following Agile/Scrum best practices, and detecting dependencies before execution.
+
+## Problem It Solves
+
+| Before | After |
+|--------|-------|
+| PO writes: "Login broken" | Complete task with technical context |
+| Dev guesses what to do | Clear instructions for LLM |
+| Dependencies discovered late | Dependencies detected before starting |
+| PM can't see real progress | Real-time dashboard |
+| See all team issues (noise) | **Only your assigned issues** |
+
+---
+
+## Per-Project Configuration
+
+Each project can have a **different issue tracker**. Configuration is stored per-project.
+
+```
+~/.prjct-cli/projects/
+├── project-a/           # Uses Linear
+│   └── project.json     → issueTracker: { provider: 'linear', teamKey: 'ENG' }
+├── project-b/           # Uses GitHub Issues
+│   └── project.json     → issueTracker: { provider: 'github', repo: 'org/repo' }
+├── project-c/           # Uses Jira
+│   └── project.json     → issueTracker: { provider: 'jira', projectKey: 'PROJ' }
+└── project-d/           # No issue tracker (standalone)
+    └── project.json     → issueTracker: null
+```
+
+### Supported Providers
+
+| Provider | Status | Auth |
+|----------|--------|------|
+| Linear | ✅ Ready | MCP (OAuth) |
+| GitHub Issues | 🔜 Soon | `GITHUB_TOKEN` |
+| Jira | 🔜 Soon | MCP (OAuth) |
+| Monday | 🔜 Soon | `MONDAY_API_KEY` |
+| None | ✅ Ready | - |
+
+### Setup per Project
+
+```bash
+# In project directory
+p. linear setup     # Configure Linear for THIS project
+p. github setup     # Configure GitHub for THIS project
+p. jira setup       # Configure Jira for THIS project
+```
+
+---
+
+## User-Scoped View
+
+**Critical:** prjct only shows issues assigned to YOU. No noise from other team members' work.
+
+```
+┌────────────────────────────────────────────────────────────┐
+│  Your Issues                                    @jlopez    │
+├────────────────────────────────────────────────────────────┤
+│                                                            │
+│  ✓ Only issues assigned to you                            │
+│  ✓ Filtered by your default team                          │
+│  ✓ Sorted by priority                                     │
+│                                                            │
+│  ENG-123  🔴 High    Login broken on mobile               │
+│  ENG-456  🟡 Medium  Add password reset                   │
+│  ENG-789  🟢 Low     Update footer links                  │
+│                                                            │
+└────────────────────────────────────────────────────────────┘
+```
+
+### Filter Options
+
+| Filter | Description |
+|--------|-------------|
+| `--mine` (default) | Only your assigned issues |
+| `--team` | All issues in your team |
+| `--project <name>` | Issues in a specific project |
+| `--unassigned` | Unassigned issues (for picking up work) |
+
+---
+
+## Enrichment Flow
+
+```
+┌─────────────────────────────────────────────────────────────┐
+│                    INPUT: Minimal title or description      │
+│                    "Login doesn't work on mobile"           │
+└─────────────────────────────────────────────────────────────┘
+                              │
+                              ▼
+┌─────────────────────────────────────────────────────────────┐
+│  PHASE 1: INTELLIGENT CLASSIFICATION                       │
+│  ─────────────────────────────────────────────────────────  │
+│  • Analyze PO intent                                        │
+│  • Classify: bug | feature | improvement | task | chore     │
+│  • Determine priority based on impact                       │
+│  • Assign labels (mobile, auth, critical, etc.)             │
+└─────────────────────────────────────────────────────────────┘
+                              │
+                              ▼
+┌─────────────────────────────────────────────────────────────┐
+│  PHASE 2: TECHNICAL ANALYSIS                                │
+│  ─────────────────────────────────────────────────────────  │
+│  • Explore related codebase                                 │
+│  • Identify affected files                                  │
+│  • Detect existing patterns                                 │
+│  • Estimate technical complexity                            │
+└─────────────────────────────────────────────────────────────┘
+                              │
+                              ▼
+┌─────────────────────────────────────────────────────────────┐
+│  PHASE 3: DEPENDENCY DETECTION                              │
+│  ─────────────────────────────────────────────────────────  │
+│  • Code dependencies (imports, services)                    │
+│  • Data dependencies (APIs, DB schemas)                     │
+│  • Task dependencies (other blocking tasks)                 │
+│  • Potential risks and blockers                             │
+└─────────────────────────────────────────────────────────────┘
+                              │
+                              ▼
+┌─────────────────────────────────────────────────────────────┐
+│  PHASE 4: USER STORY GENERATION                             │
+│  ─────────────────────────────────────────────────────────  │
+│  • User story format: As a [role], I want [action]...       │
+│  • Acceptance Criteria (Gherkin or checklist)               │
+│  • Definition of Done                                       │
+│  • Technical notes for the developer                        │
+└─────────────────────────────────────────────────────────────┘
+                              │
+                              ▼
+┌─────────────────────────────────────────────────────────────┐
+│  PHASE 5: LLM PROMPT                                        │
+│  ─────────────────────────────────────────────────────────  │
+│  • Generate optimized prompt for Claude/LLM                 │
+│  • Include codebase context                                 │
+│  • Implementation instructions                              │
+│  • Verification criteria                                    │
+└─────────────────────────────────────────────────────────────┘
+                              │
+                              ▼
+┌─────────────────────────────────────────────────────────────┐
+│                    OUTPUT: Enriched Task                    │
+└─────────────────────────────────────────────────────────────┘
+```
+
+---
+
+## Output Format
+
+### For PM/PO (Product View)
+
+```markdown
+## 🐛 BUG: Login doesn't work on mobile
+
+**Priority:** 🔴 High (affects conversion)
+**Type:** Bug
+**Sprint:** Current
+**Estimate:** 3 points
+
+### User Story
+As a **mobile user**, I want to **log in from my phone**
+so that **I can access my account without using desktop**.
+
+### Acceptance Criteria
+- [ ] Login form displays correctly on screens < 768px
+- [ ] Submit button is clickable on iOS and Android
+- [ ] Error messages are visible on mobile
+- [ ] Successful login redirects to dashboard
+
+### Dependencies
+⚠️ **Potential blocker:** Auth service uses cookies that may
+   have issues with WebView in native apps.
+
+### Impact
+- Affected users: ~40% of traffic
+- Related metrics: Login conversion rate, Mobile bounce rate
+```
+
+### For Developer (Technical View)
+
+```markdown
+## Technical Context
+
+### Affected Files
+- `src/components/Auth/LoginForm.tsx` - Main form
+- `src/styles/auth.css` - Responsive styles
+- `src/hooks/useAuth.ts` - Auth hook
+- `src/services/auth.ts` - API calls
+
+### Problem Analysis
+The viewport meta tag is incorrectly configured in `index.html`.
+Styles in `auth.css:45-67` use `min-width` when they should use `max-width`.
+
+### Pattern to Follow
+See similar implementation in `src/components/Profile/EditForm.tsx`
+which handles responsive correctly.
+
+### LLM Prompt (Copy & Paste Ready)
+
+Use this prompt with any AI assistant (Claude, ChatGPT, Copilot, Gemini, etc.):
+
+\`\`\`
+## Task: Fix mobile login
+
+### Context
+I'm working on a codebase with the following structure:
+- Frontend: React/TypeScript
+- Auth: Custom hooks in src/hooks/useAuth.ts
+- Styles: CSS modules in src/styles/
+
+### Problem
+The login form doesn't work correctly on mobile devices.
+
+### What needs to be done
+1. Check viewport meta tag in index.html
+2. Fix CSS media queries in auth.css (change min-width to max-width)
+3. Ensure touch events work (onClick should also handle onTouchEnd)
+
+### Files to modify
+- src/components/Auth/LoginForm.tsx
+- src/styles/auth.css
+- index.html
+
+### Reference implementation
+See src/components/Profile/EditForm.tsx for a working responsive pattern.
+
+### Acceptance criteria
+- [ ] Login works on iPhone Safari
+- [ ] Login works on Android Chrome
+- [ ] Desktop version still works
+- [ ] No console errors on mobile
+
+### How to verify
+1. Run `npm run dev`
+2. Open browser dev tools, toggle mobile view
+3. Test login flow on different screen sizes
+\`\`\`
+```
+
+---
+
+## Dependency Detection
+
+### Dependency Types
+
+| Type | Example | Detection |
+|------|---------|-----------|
+| **Code** | `LoginForm` imports `useAuth` | Import analysis |
+| **API** | `/api/auth/login` endpoint | Grep fetch/axios calls |
+| **Database** | Table `users`, field `last_login` | Schema analysis |
+| **Tasks** | "Deploy new endpoint" blocked | Task queue analysis |
+| **Infrastructure** | Redis for sessions | Config file analysis |
+
+### Report Format
+
+```yaml
+dependencies:
+  code:
+    - file: src/hooks/useAuth.ts
+      reason: Main auth hook
+      risk: low
+    - file: src/services/auth.ts
+      reason: API calls
+      risk: medium (changes here affect other flows)
+
+  api:
+    - endpoint: POST /api/auth/login
+      status: stable
+      risk: low
+
+  blocking_tasks:
+    - id: ENG-456
+      title: "Migrate to OAuth 2.0"
+      status: in_progress
+      risk: high (may change auth flow)
+
+  infrastructure:
+    - service: Redis
+      purpose: Session storage
+      risk: none (no changes required)
+```
+
+---
+
+## Integration with Linear/Jira
+
+### Bidirectional Sync
+
+```
+Linear/Jira Issue          prjct Enrichment
+─────────────────          ─────────────────
+Basic title       ──────►  Complete User Story
+No AC             ──────►  Acceptance Criteria
+No context        ──────►  Technical notes
+Manual priority   ──────►  Suggested priority
+                  ◄──────  Updates description
+                  ◄──────  Updates labels
+                  ◄──────  Marks progress
+```
+
+### Fields Enriched
+
+| Field | Before | After |
+|-------|--------|-------|
+| Description | "Login broken" | User story + AC + technical notes |
+| Labels | (empty) | `bug`, `mobile`, `auth`, `high-priority` |
+| Estimate | (empty) | 3 points (based on analysis) |
+| Assignee | (empty) | Suggested based on `git blame` |
+
+---
+
+## Commands
+
+| Command | Action |
+|---------|--------|
+| `p. enrich <title>` | Enrich minimal description |
+| `p. analyze <ID>` | Analyze existing issue |
+| `p. deps <ID>` | Detect dependencies |
+| `p. ready <ID>` | Check if task is ready for dev |
+| `p. prompt <ID>` | Generate optimized LLM prompt |
+
+---
+
+## PM Metrics
+
+### Real-Time Dashboard
+
+```
+┌────────────────────────────────────────────────────────────┐
+│  Sprint Progress                                    v0.29  │
+├────────────────────────────────────────────────────────────┤
+│                                                            │
+│  Features    ████████░░░░░░░░░░░░  40%  (4/10)            │
+│  Bugs        ██████████████░░░░░░  70%  (7/10)            │
+│  Tech Debt   ████░░░░░░░░░░░░░░░░  20%  (2/10)            │
+│                                                            │
+│  ─────────────────────────────────────────────────────────│
+│  Velocity: 23 pts/sprint (↑ 15% vs last)                  │
+│  Blockers: 2 (ENG-456, ENG-789)                           │
+│  Ready for Dev: 5 tasks                                    │
+│                                                            │
+│  Recent Activity                                           │
+│  • ENG-123 shipped (login fix) - 2h ago                   │
+│  • ENG-124 enriched - 30m ago                             │
+│  • ENG-125 blocked by ENG-456 - just now                  │
+│                                                            │
+└────────────────────────────────────────────────────────────┘
+```
+
+---
+
+## Core Principle
+
+> **We don't break "just ship"** - Enrichment is a helper layer,
+> not a blocker. Developers can always run `p. task` directly.
+> PM Expert improves quality, doesn't add bureaucracy.