adhd-productivity
H2 Tank Designer (proagetic-dfx)的全面用户验收测试(UAT)自动化。执行30个关键冒烟测试,涵盖需求输入、3D可视化、分析面板、合规性、验证和导出。使用Chrome DevTools浏览器自动化,并强制进行屏幕截图分析。在流程完成后捕获屏幕截图,用详细描述分析每个截图,增量更新报告。在使用基于证据的测试验证H2 Tank Designer功能时使用。
📖 中文导读:H2 Tank Designer (proagetic-dfx)的全面用户验收测试(UAT)自动化。执行30个关键冒烟测试,涵盖需求输入、3D可视化、分析面板、合规性、验证和导出。使用Chrome DevTools浏览器自动化,并强制进行屏幕截图分析。在流程完成后捕获屏幕截图,用详细描述分析每个截图,增量更新报告。在使用基于证据的测试验证H2 Tank Designer功能时使用。 以下正文为该技能的原始 SKILL.md 内容(英文),上方安装命令可直接使用。
name: adhd-productivity description: "ADHD-optimized productivity techniques and interventions. Invoke when user shows signs of task abandonment, context switching, or needs focus assistance." allowed-tools: - Read - Write - Bash
ADHD Productivity Skill
When to Invoke
- User starts new topic without closing previous task
- User mentions feeling stuck or overwhelmed
- Session duration exceeds 30 minutes without completion
- User asks for help focusing
Intervention Techniques
Micro-Commitment
Break tasks into smallest possible step:
- "Just open the file"
- "Just write the first line"
- "Just run one command"
Body Doubling
Virtual presence accountability:
- "I'm here. You: [action]. Me: ⏱️ Waiting..."
- Set 5-minute timer
- Check in after timer
Chunking
Divide overwhelming tasks:
- Maximum 3 steps visible at once
- Complete step 1 before revealing step 2
- Celebrate each chunk completion
Pattern Recognition
Track and call out patterns:
- "This is the 3rd time this week you've switched away from [task]"
- "You typically abandon tasks at the [specific point]"
- "Your completion rate improves when you [specific behavior]"
Task State Management
INITIATED → SOLUTION_PROVIDED → IN_PROGRESS → COMPLETED
↘ ABANDONED (log pattern)
↘ BLOCKED (external wait)
↘ DEFERRED (intentional)
Response Templates
Level 1 Intervention (0-30 min)
"📌 Quick check: [task] - still on it?"
Level 2 Intervention (30-60 min)
"🔄 I notice [task] from earlier. Pattern: [observation]. Continue or defer?"
Level 3 Intervention (>60 min)
"⚠️ ACCOUNTABILITY: [task] started [time] ago. Status? Be honest."
Success Metrics
Track in task_completion_streaks table:
- Daily completion count
- Streak length
- Common blockers
- Peak productivity times