Second Brain Triage

Usage Scenarios

Scenario 1: Triage a Single Task

User input: "Categorize this task: 'Complete quarterly report, due next Friday'"

Expected output: The tool classifies the item under "Projects" with high urgency score (7-8), recommends processing within 24 hours, and provides the full triage summary including type, category, and action.

Scenario 2: Batch Organize Bookmarks into PARA Categories

User input: "Organize my 20 bookmarked articles and resources using PARA"

Expected output: The tool processes each item through batch triage, categorizes them into Projects/Areas/Resources/Archive with individual urgency scores, and returns a consolidated report.

Scenario 3: Export Triage Report as Markdown

User input: "Export all my triaged items as a formatted Markdown report"

Expected output: The tool generates a readable Markdown report with all categorized items, their scores, and recommended actions, saved or printed to stdout.

Intelligent information triage system based on Tiago Forte's PARA method (Projects/Areas/Resources/Archive) for automatic categorization and priority scoring.

Scenario 4: 微信收藏夹快满了

User input: "我的微信收藏夹有324条内容，看了就想收藏但从来不回看，怎么清理和整理？"

Expected output: 提供'碎片信息分诊'三步法：1）批量导出微信收藏夹内容到Notion/飞书/印象笔记；2）用PARA分类法（Projects/Areas/Resources/Archives）快速标记；3）制定每日15分钟回顾规则：只保留当前需要的信息，其他直接归档。建议以后看到好内容先判断：对我当前的项目有用吗？如果5分钟内用不上就暂时不收藏。

Features

• Content Analyzer: Automatically identify content types (articles, videos, tasks, code, etc.) and extract metadata
• PARA Classifier: Smart categorization into Projects/Areas/Resources/Archive
• Urgency Scorer: Multi-dimensional algorithm to evaluate processing priority (1-10 scale)
• Relatedness Detector: Discover similarities and relationships between content items

Usage

Basic Usage

const { SecondBrainTriage } = require('./src');

const triage = new SecondBrainTriage();

// Triage single content item
const result = triage.triage('TODO: Complete project report, due this Friday');
console.log(result.summary);
// {
//   title: "Complete project report, due this Friday",
//   type: "task",
//   category: "Projects",
//   urgency: "High urgency",
//   urgencyScore: 8,
//   action: "Process today: recommend completing within 24 hours"
// }

// Batch triage
const results = triage.triageBatch([
  'https://github.com/user/repo',
  'Notes on learning React Hooks',
  'TODO: Fix login bug',
]);

// Export report
const report = triage.exportReport(results, 'markdown');

CLI Usage

# Analyze single content item
node scripts/triage.js "Text content to process"

# Analyze file
node scripts/triage.js --file ./notes.txt

# Batch analysis
node scripts/triage.js --batch ./items.json --output report.md

Classification Guide

PARA Categories

Urgency Levels

Technical Architecture

src/
├── content-analyzer.js    # Content type recognition and metadata extraction
├── para-classifier.js     # PARA classification algorithm
├── urgency-scorer.js      # Urgency scoring algorithm
├── relatedness-detector.js # Relatedness detection
└── index.js               # Main entry and API

Scoring Algorithm

Urgency Scoring Dimensions

1. Time Sensitivity (30%): Deadlines, time keywords
2. Action Requirement (25%): Action verbs like must/plan/maybe
3. Consequences (20%): Potential impact of not processing
4. Context Signals (15%): Blockers, external dependencies
5. User Preferences (10%): Configurable priorities

Relatedness Detection

• Tag similarity (Jaccard coefficient)
• Title/description text similarity (Cosine similarity)
• Semantic similarity (based on semantic groups)
• Type matching

Configuration Options

const triage = new SecondBrainTriage({
  enableRelatedness: true,    // Enable relatedness detection
  urgencyThreshold: 5,        // Urgency threshold
});

Output Formats

Supported export formats:

• JSON (complete data)
• Markdown (readable format)
• CSV (spreadsheet format)

Dependencies

• Node.js >= 14
• No external dependencies (pure JavaScript implementation)

License

MIT