Second Brain Triage

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

Security Scan

The skill's code, documentation, and runtime instructions are consistent with a local PARA-based triage tool and do not request unrelated credentials, installs, or network access.

Purpose & Capability

Name/description (PARA triage, urgency scoring, relatedness) match the included code (ContentAnalyzer, ParaClassifier, UrgencyScorer, RelatednessDetector). No unrelated credentials, binaries, or config paths are requested.

Instruction Scope

SKILL.md and CLI instructions only describe local analysis, batch processing, and exporting (JSON/Markdown/CSV). The runtime code reads provided files or CLI args and performs in-memory analysis. Instructions do not direct the agent to read unrelated system state or send data externally.

Install Mechanism

There is no installer spec in the registry (instruction-only), but the package contains source files and package.json. The code uses only Node.js standard libraries (fs/path) and declares no external download/install steps — low install risk.

Credentials

requires.env and primary credential are empty, and the code does not access environment secrets or external service tokens. All environment/credential demands are proportional (none).

Persistence & Privilege

The skill is not force-included (always: false) and does not modify other skills or system-wide agent settings. It runs as a normal, local library/CLI without elevated persistence.

Assessment

This skill appears to be a self-contained Node.js library/CLI for classifying and scoring personal content using PARA and local heuristics. It does not request credentials or contact external servers. Things to consider before installing: (1) origin unknown — review the code yourself or run in a sandbox before granting it access to sensitive files; (2) it reads any file path you pass to the CLI, so avoid pointing it at directories with secrets; (3) there are no external dependencies, but test the package (npm test) to verify behavior in your environment; (4) if you plan to integrate it into an automated agent, remember the agent could run it on arbitrary content you supply — ensure inputs are trusted.

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

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

Category	Description	Examples
Projects	Items with clear goals and deadlines	"Develop new feature", "Complete report"
Areas	Long-term responsibilities and standards	"Health management", "Skill development"
Resources	Topics of interest and reference materials	"Technical articles", "Learning notes"
Archive	Completed or inactive items	"Finished projects", "Historical records"
Inbox	Temporary storage for uncategorized items	Content that cannot be determined

Urgency Levels

Score	Level	Description	Recommendation
9-10	Critical	Process immediately	Take action now
7-8	High	Process today	Complete within 24 hours
5-6	Medium	Process this week	Schedule within the week
3-4	Low	Low priority	Can be deferred
1-2	Minimal	Archive for reference	No immediate action needed

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