atomization

Create schema-compliant atomic notes in NODES/ with deduplication, tagging, and graph linkage.

3|4|Updated Jul 20, 2026
One-click install
npx skills add https://github.com/xrcodexcode/xr_node --skill atomization-xrcodexcode
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: atomization
Source: https://github.com/xrcodexcode/xr_node/tree/main/claude/skills/atomization
Command: npx skills add https://github.com/xrcodexcode/xr_node --skill atomization-xrcodexcode

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Turning study notes, research, or raw ideas into reusable atomic knowledge notes is error-prone: duplicates creep in, tags fragment, and notes end up orphaned from the knowledge graph. This Skill enforces a governed pipeline so every note created in NODES/ is deduplicated, schema-valid, tagged from a controlled vocabulary, and linked into the MOC navigation graph. ## Core Features & Use Cases - 8-Step Creation Pipeline: Classifies the note type, runs a 3-tier deduplication protocol, discovers the owner MOC, maps relationships, selects controlled tags, drafts from templates, validates against a 12-point checklist, and writes the note. - Four Input Modes: Single concept creation, batch extraction from study notes, topic deep-dives, and merge/update of overlapping existing notes. - Strict Governance: Enforces Frontmatter Schema v4, kebab-case naming, 12 approved discovery tags, source provenance, and zero-orphan graph connectivity. - Use Case: After finishing a study note on a machine learning video, ask to atomize its downstream concept candidates — the Skill deduplicates each candidate against existing nodes, presents a batch plan, and writes approved atomic notes linked to the right MOC. ## Quick Start Ask the agent to create an atomic note for a specific concept, for example by saying: create an atomic note for the concept of Spaced Repetition.

Frequently Asked Questions about atomization

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

FAQPage Schema
How do I create atomic notes from study notes?

Point the Skill to a study note in 02_NEW-KNOWLEDGE/ containing a Downstream Atomic Concept Candidates section. It extracts each candidate, deduplicates against existing nodes and within the batch, presents a plan for approval, then writes the approved notes to NODES/.

How does atomic note deduplication work before creation?

A 3-tier protocol runs before any note is written: exact filename match, alias and title substring overlap, then semantic keyword co-occurrence across note bodies. Any match halts the pipeline and offers merge, differentiate, or cancel options.

What frontmatter fields are required for an atomic note?

Schema v4 requires id (UUID v4), title, type, status, domain, source_type, created/updated dates, review date, confidence, version, aliases, tags, owner_moc, sources, related, and schema_version. All fields are validated by a 12-point checklist before writing.

Can I use custom tags on atomic notes?

No. Only 12 approved discovery tags are allowed, such as beginner, comparison, implementation, and reference, with 0-3 tags per note. Tags not in the controlled vocabulary in tagging.md are rejected to prevent taxonomy fragmentation.

What happens if a new concept overlaps an existing note?

The Skill presents the existing note alongside the new concept and offers three options: merge the new information into the existing note, differentiate by creating a note with a refined scope, or cancel creation entirely.

What are the limitations of the atomization skill?

It does not create or modify MOC files, edit raw sources in 01_RAW/, or produce literature and study notes, which are upstream skill outputs. Concepts with no related notes are linked only to a MOC and flagged as isolated.