生产力提升
Lord1Egypt2026-10-25查看 641复制 1078
| 个人生产力追踪与分析技能。通过语音/文本输入记录工作和生活活动, 追踪时间、分类任务,并生成带有洞察和改进建议的每日日志。 当用户想要追踪时间、记录活动、分析生产力模式或获取每日工作/生活总结时使用。
📘如何使用这个技能
- 1复制技能内容·点击上方「Copy SKILL.md」复制全部内容,或 Download 下载到本地
- 2打开你的 AI 客户端·支持 Claude Code、Trae、Cursor、ChatGPT、DeepSeek、Kimi 等任意 AI 助手
- 3粘贴为系统提示/知识文件·Claude Code:放入 .claude/skills/ 目录;普通对话:作为首条消息发送并说「请严格按这份技能文档执行」
- 4直接提出你的需求·AI 会按技能定义的工作流、格式和约束来处理,无需再解释背景
💡 提示:技能生效后,每次对话只需一句需求,AI 会按 SKILL.md 的流程执行。
SKILL.md5067 chars
---
name: productivity-improving
description: |
Personal productivity tracking and analysis skill. Records work and life activities via voice/text input,
tracks time, categorizes tasks, and generates daily logs with insights and suggestions for improvement.
Use when the user wants to track time, log activities, analyze productivity patterns, or get daily work/life summaries.
triggers:
- start
- complete
- log
- what did I do today
- productivity
- time tracking
- work log
- efficiency analysis
---
# Productivity Tracker
Track, categorize, and analyze your work and life activities to improve efficiency and maintain balance.
## Input Methods
### Voice/Text Input
- **Start Activity**: "start [activity name]"
- **Complete Activity**: "complete"
- **Quick Log**: "log [activity] took [duration]"
### Natural Language Examples
```text
"start coding"
"start meeting"
"complete"
"log workout 45 minutes"
"what did I do today"
"analyze my productivity this week"
```
## Core Features
### 1. Activity Recording
- Real-time activity tracking with start/end timestamps
- Automatic duration calculation
- Support for interruptions and resumption
- Voice and text input support
### 2. Smart Categorization
Auto-categorize activities into:
- **Work**: coding, meetings, emails, planning
- **Learning**: reading, courses, research
- **Health**: exercise, meditation, sleep
- **Life**: cooking, cleaning, family time
- **Rest**: entertainment, social media, breaks
### 3. Time Analysis
- Daily/weekly/monthly time distribution
- Focus time vs. fragmented time analysis
- Peak productivity hours identification
- Work-life balance metrics
### 4. Daily Report Generation
```markdown
# 2026-03-19 Work Log
## Overview
- Total activities: 12
- Focus time: 6.5 hours
- Rest time: 2 hours
- Work/Life ratio: 65%/35%
## Time Distribution
| Category | Duration | Percentage |
|----------|----------|------------|
| Deep Work | 4h | 40% |
| Meetings | 1.5h | 15% |
| Learning | 1h | 10% |
| Life Tasks | 2h | 20% |
| Rest | 1.5h | 15% |
## Key Activities
- Completed project docs (2h, Deep Work)
- Team weekly meeting (1h, Meetings)
- Read tech article (45min, Learning)
## Insights
Highlight: Peak focus 9-11 AM, core tasks completed
Improvement: Frequent interruptions 3-4 PM, reserve block time
Trend: Deep work time increased 15% vs last week
## Tomorrow's Suggestions
1. Maintain morning deep work routine
2. Batch email processing after 4 PM
3. Reserve 30 minutes for tomorrow's planning
```
## Data Storage
### Local Storage
- **Activities**: `data/activities.json`
- **Daily Logs**: `data/logs/YYYY-MM-DD.md`
- **Analytics**: `data/analytics/`
### Export Options
- Daily/weekly Markdown reports
- CSV for spreadsheet analysis
- JSON for API integration
## Workflow
### Phase 1: Capture
1. User says "start [activity]"
2. Record timestamp and activity name
3. Auto-categorize based on keywords
4. Confirm category with user if uncertain
### Phase 2: Track
1. Monitor active activity
2. Allow "pause" and "resume"
3. Handle interruptions gracefully
4. Record end timestamp on completion
### Phase 3: Analyze
1. Calculate duration and metrics
2. Update category totals
3. Compare with historical patterns
4. Generate insights
### Phase 4: Report
1. Generate daily summary at user-defined time
2. Weekly review with trends
3. Monthly analysis with recommendations
4. Export to Obsidian or other tools
## Commands
| Command | Description |
|---------|-------------|
| `/track start [activity]` | Start tracking an activity |
| `/track stop` | Stop current activity |
| `/track status` | Show current activity and today's summary |
| `/track log [activity] [duration]` | Quick log a completed activity |
| `/track report daily` | Generate today's report |
| `/track report weekly` | Generate weekly analysis |
| `/track category [name]` | Show time spent in category |
| `/track insights` | Get productivity suggestions |
## Configuration
```json
{
"dailyReportTime": "21:00",
"categories": {
"work": { "color": "#4CAF50", "keywords": ["code", "meeting", "email"] },
"learning": { "color": "#2196F3", "keywords": ["read", "study", "course"] },
"health": { "color": "#FF9800", "keywords": ["exercise", "meditation", "sleep"] },
"life": { "color": "#9C27B0", "keywords": ["cook", "clean", "family"] },
"rest": { "color": "#607D8B", "keywords": ["rest", "entertainment", "break"] }
},
"focusThresholdMinutes": 25,
"breakReminderIntervalMinutes": 90
}
```
## Privacy & Security
- All data stored locally
- No cloud sync by default
- Optional encryption for sensitive logs
- User owns all data
## Integration
- Export to Obsidian daily notes
- Sync with calendar events
- Connect with health apps (optional)
- API for custom workflows
## Technical Info
| Property | Value |
|----------|-------|
| **Name** | Productivity Tracker |
| **Slug** | `productivity-improving` |
| **Version** | 1.0.5 |
| **Category** | Productivity / Lifestyle |
| **Tags** | time-tracking, productivity, analytics, daily-log |