Files
gh-comzine-claude-code-mark…/skills/meta-automation-architect/examples/EXAMPLE_PROJECT_MANAGEMENT.md
2025-11-29 18:16:25 +08:00

9.6 KiB
Raw Blame History

Example: Project Management Automation

This example shows what the meta-automation-architect generates for a software delivery project management workspace.

Project Context

  • Type: Project Management (primary) + Programming (secondary)
  • Project: Q1 2025 Mobile App Development
  • Team: 8 people (2 devs, 2 designers, 1 PM, 1 QA, 2 stakeholders)
  • Duration: 3 months (12 sprints)
  • Pain Points: Manual status reporting, resource over-allocation, missing risk tracking
  • Priority: Automated reporting and resource management

Project Structure

mobile-app-project/
├── project-plan.md
├── roadmap.md
├── sprints/
│   ├── sprint-01/
│   │   ├── planning.md
│   │   ├── tasks.md
│   │   ├── retrospective.md
│   │   └── burndown-chart.png
│   ├── sprint-02/
│   └── ... (12 sprints)
├── resources/
│   ├── team-capacity.md
│   ├── allocation.xlsx
│   └── skills-matrix.md
├── risks/
│   ├── risk-register.md
│   └── mitigation-plans.md
├── milestones/
│   ├── m1-design-complete.md
│   ├── m2-mvp-ready.md
│   └── m3-launch-ready.md
└── reports/
    ├── weekly-status/
    └── monthly-summary/

User Conversation

User: "Set up automation for my mobile app project management"

Meta-Skill: Analyzing project...
- Detected: Project Management (78% confidence)
  - Found sprints/ directory with 12 sprint folders
  - Found project-plan.md and roadmap.md
  - Found resources/allocation.xlsx
  - Found risks/risk-register.md
- Secondary: Programming (22% confidence)
  - References to code deliverables

Meta-Skill: What are your main pain points?
Options:
- Manual status reporting (RECOMMENDED: 12 weekly reports expected, only 3 exist)
- Resource over-allocation (RECOMMENDED: Excel shows conflicts)
- Missing risk tracking
- Timeline validation

User: [Selects: Manual status reporting, Resource over-allocation]

Meta-Skill: Generating automation system...
Session ID: pm-mobile-app-xyz789

Generated Automation System

1. Custom Subagents (6)

Universal Agents

  • structure-analyzer - Project organization and folder hierarchy
  • workflow-analyzer - Sprint and delivery processes

Project Management Domain Agents

  • timeline-analyzer - Sprint schedules, dependencies, critical paths
  • resource-analyzer - Team allocation, capacity, conflicts
  • risk-analyzer - Risk identification and mitigation coverage

Implementation Agent

  • command-generator - Created 3 PM-specific commands

2. Custom Skills (3)

status-reporter - Auto-generates weekly status reports from sprint data

  • Reads sprint tasks, completion status, blockers
  • Generates formatted report with metrics
  • Saves time: 45 min/week (9 hours over 12 sprints)

resource-optimizer - Identifies and resolves allocation conflicts

  • Parses resource allocation data
  • Detects over/under allocation
  • Suggests rebalancing
  • Saves time: 30 min/sprint (6 hours total)

risk-tracker - Maintains risk register and tracks mitigation

  • Monitors risks from register
  • Tracks mitigation progress
  • Alerts on new risks
  • Saves time: 20 min/week (4 hours total)

3. Custom Commands (3)

/sprint-report

/sprint-report                # Current sprint
/sprint-report sprint-05      # Specific sprint
/sprint-report --all          # All sprints summary

Generates comprehensive sprint report:

  • Completed tasks vs. planned
  • Velocity and burndown
  • Blockers and risks
  • Team capacity utilization
  • Next sprint forecast

/resource-check

/resource-check               # Check current allocation
/resource-check --week 5      # Specific week
/resource-check --conflicts   # Show only conflicts

Analyzes resource allocation:

  • Capacity vs. assigned work
  • Over-allocated team members
  • Under-utilized resources
  • Skill match for tasks
  • Rebalancing suggestions

/timeline-validate

/timeline-validate            # Validate full timeline
/timeline-validate --critical # Show critical path
/timeline-validate --risks    # Timeline risks

Validates project timeline:

  • Dependency validation
  • Critical path analysis
  • Buffer analysis
  • Risk to deadlines
  • Suggested adjustments

4. Hooks (2)

update_progress.py (PostToolUse)

  • Triggers when task markdown files are updated
  • Extracts completion status
  • Updates sprint progress automatically
  • Regenerates burndown chart

resource_validation.py (PreToolUse)

  • Triggers when allocation files are modified
  • Validates no over-allocation
  • Blocks if conflicts detected
  • Suggests fixes before allowing change

5. Documentation

Impact Assessment:

Time Savings:

  • Weekly status reports: 45 min/week × 12 weeks = 9 hours
  • Resource planning: 30 min/sprint × 12 sprints = 6 hours
  • Risk tracking: 20 min/week × 12 weeks = 4 hours
  • Timeline validation: 1 hour/month × 3 months = 3 hours
  • Total: 22 hours saved

Quality Improvements:

  • 100% sprint coverage for status reports (up from 25%)
  • Zero resource conflicts (automated detection)
  • Real-time risk visibility (vs. monthly reviews)
  • Validated timeline with critical path analysis

Example Usage

Generating Sprint Report

/sprint-report sprint-05

Output:

# Sprint 05 Report
**Period:** Jan 15-19, 2025
**Team:** Mobile App Team

## Summary
✅ Sprint Goal: Complete user authentication flow - ACHIEVED

## Metrics
- **Planned Story Points:** 34
- **Completed Story Points:** 32 (94%)
- **Velocity:** 32 pts (trending up from 28)
- **Team Capacity:** 160 hours
- **Actual Hours:** 152 hours (95% utilization)

## Completed Tasks (8)
✅ User login UI (8 pts) - @designer1
✅ OAuth integration (13 pts) - @dev1
✅ Session management (5 pts) - @dev2
✅ Login testing (3 pts) - @qa
[... 4 more ...]

## Carried Over (2)
➡️  Password reset flow (5 pts) - blocked by API
➡️  Biometric auth (8 pts) - iOS issue

## Blockers
⚠️  API authentication endpoint delayed → unblocks next sprint
⚠️  iOS biometric API documentation unclear → dev investigating

## Risks
🔴 HIGH: OAuth provider latency affecting UX → mitigation: add loading states
🟡 MEDIUM: Test coverage only 65% → action: dedicated testing time in sprint 06

## Team Notes
- Great collaboration between design and dev
- Need more QA capacity in next sprint
- Consider adding API buffer time

## Next Sprint Preview
Sprint 06 will focus on:
- Complete password reset (carried over)
- User profile management
- Settings screen
- Target: 30-35 story points

---
Auto-generated by meta-automation-architect

Checking Resource Allocation

/resource-check

Output:

📊 Resource Allocation Analysis

Current Week: Week 6 (Jan 22-26, 2025)

✅ Properly Allocated (5):
- @dev1: 40/40 hours (100%)
- @designer2: 32/40 hours (80%)
- @qa: 40/40 hours (100%)
- @pm: 35/40 hours (88%)
- @stakeholder1: 10/10 hours (100%)

⚠️  Over-Allocated (2):
- @dev2: 52/40 hours (130% - CRITICAL)
  Tasks: Feature A (20h), Feature B (18h), Bug fixes (14h)
  Recommendation: Move Bug fixes to @dev1 (8h available)

- @designer1: 45/40 hours (113%)
  Tasks: UI mockups (25h), Asset creation (20h)
  Recommendation: Defer Asset creation to Week 7

🔵 Under-Utilized (1):
- @stakeholder2: 3/10 hours (30%)
  Opportunity: Review sessions, feedback collection

💡 Optimization Suggestions:
1. Redistribute 14h from @dev2 to @dev1
2. Move Asset creation from @designer1 to Week 7
3. Add review tasks for @stakeholder2

Estimated Rebalancing Time: 10 minutes
After optimization: 100% feasible allocation

Agent Communication

reports/timeline-analyzer.json (excerpt):

{
  "agent_name": "timeline-analyzer",
  "summary": "Timeline feasible but tight. Critical path includes 4 sprints with zero buffer.",
  "findings": [
    {
      "type": "risk",
      "severity": "high",
      "title": "Zero Buffer on Critical Path",
      "description": "Sprints 4, 7, 9, 11 are on critical path with no schedule buffer",
      "recommendation": "Add 10% buffer to each critical sprint or reduce scope",
      "time_impact": "Any delay in these sprints directly impacts launch date"
    },
    {
      "type": "opportunity",
      "severity": "medium",
      "title": "Parallel Workstreams Possible",
      "description": "Design and backend development can run in parallel in sprints 2-5",
      "recommendation": "Optimize resource allocation to leverage parallelism",
      "time_saved_if_optimized": "2 weeks off critical path"
    }
  ],
  "automation_impact": {
    "time_saved": "3 hours per month in timeline reviews",
    "quality_improvement": "Real-time critical path visibility vs. monthly checks"
  }
}

Result

PM now has powerful automation:

  • Weekly status reports generated in 30 seconds (vs. 45 minutes)
  • Resource conflicts detected instantly (vs. discovered in standup)
  • Risk register automatically maintained
  • Timeline validated continuously
  • 22 hours saved over project duration
  • Better decision making with real-time data

Before vs After:

Before (Manual):

  • Weekly status report: 45 min
  • Resource planning: 30 min/sprint
  • Risk review: 20 min/week
  • Timeline validation: 1 hour/month
  • Total: ~3 hours/week

After (Automated):

  • Status report: /sprint-report → 30 seconds
  • Resource check: /resource-check → 30 seconds
  • Risk tracking: Auto-updated from tasks
  • Timeline: Auto-validated on every change
  • Total: ~5 minutes/week

Impact: 97% time reduction + higher data quality!