collect

Save URLs, notes, files, or excerpts into a PARA-structured knowledge base.

37|7|Updated Mar 19, 2026
One-click install
npx skills add https://github.com/unclejobs-ai/second-claude-code --skill collect-unclejobs-ai
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: collect
Source: https://github.com/unclejobs-ai/second-claude-code/tree/main/skills/collect
Command: npx skills add https://github.com/unclejobs-ai/second-claude-code --skill collect-unclejobs-ai

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

It prevents knowledge-base rot by ensuring every saved item is reduced, classified into PARA, and connected using a specific concept rather than vague similarity.

Core Features & Use Cases

  • PARA classification at capture time: Stores knowledge as project, area, resource, or archive based on explicit criteria.
  • Source reduction with quality gates: Extracts and reduces content into exactly three key points plus a short summary, avoiding verbatim storage.
  • Bias-resistant concept linking: Runs a separate connector that finds shared principles/patterns/concepts using only the saved knowledge base.

Quick Start

Use the collect skill to save a URL into your PARA knowledge base by calling /second-claude-code:collect with the URL as input.

Frequently Asked Questions about collect

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I capture URLs and notes into a PARA knowledge base without creating duplicates?

The collect skill prevents duplicates by running duplicate detection against stored knowledge search before accepting any URL, note, or file into your PARA knowledge base.

What is the PARA method for structured knowledge storage?

The PARA method structures knowledge storage into Projects, Areas, Resources, and Archives, and the collect skill classifies every captured item into one of these categories at capture time.

How does concept linking work when archiving research?

Concept linking works by dispatching a dedicated connector subagent that finds shared principles and patterns exclusively within your saved knowledge base, avoiding vague similarity matches.

Can I save full verbatim text excerpts into the knowledge base?

No, the collect skill enforces source reduction by extracting exactly three key points plus a short summary, with strict length validation to prevent verbatim text dumps from entering the knowledge base.

What file formats does the knowledge base write to?

The collect skill writes both JSON index and YAML markdown files under the configured data directory, providing structured storage that supports search and retrieval of your reduced knowledge representations.

How do I avoid knowledge-base rot when capturing information?

You avoid knowledge-base rot by using the collect skill to reduce content into validated summaries, classify items into PARA, and create concept-level connections rather than storing raw dumps.