specialized-zk-steward

Builds and maintains Zettelkasten knowledge bases with atomic notes, linking, and validation loops.

2|Updated Apr 7, 2026
One-click install
npx skills add https://github.com/30eggis/walwal-harness --skill specialized-zk-steward-30eggis
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: specialized-zk-steward
Source: https://github.com/30eggis/walwal-harness/tree/main/HR-Resource/specialized-zk-steward
Command: npx skills add https://github.com/30eggis/walwal-harness --skill specialized-zk-steward-30eggis

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Personal knowledge bases often degrade into disconnected piles of notes that are hard to retrieve and reuse. This Skill enforces Niklas Luhmann's Zettelkasten method—atomic notes, meaningful links, and index entries—so every note becomes part of a connected, searchable knowledge network. ## Core Features & Use Cases - Atomic Note Validation: Applies Luhmann's four principles (atomicity, connectivity, organic growth, continued dialogue) as a gate before any note is filed. - Link and Index Management: Requires at least two meaningful links per note, proposes link candidates and keyword entries, and maintains index/MOC notes as entry points. - Expert Perspective Switching: Selects a domain expert lens (Feynman, Munger, Ogilvy, Karpathy, and others) based on domain, task type, and output form, and declares it in every reply. - Use Case: After finishing a long technical book, ask the agent to produce a structure note with a logic tree, linked atomic notes, an index entry, and a daily log update for your vault. ## Quick Start Ask the agent to turn the attached meeting notes into atomic Zettelkasten notes with links, an index entry, and a daily log update.

Frequently Asked Questions about specialized-zk-steward

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

FAQPage Schema
How do I build a Zettelkasten knowledge base with an AI agent?▼

Assign the agent a note-taking or filing task and it applies Luhmann's method automatically: notes are kept atomic, linked to at least two related notes, filed under a date-based path, and registered in an index note. Each task closes with a validation checklist and daily log entry.

What are Luhmann's four principles for atomic notes?▼

The four principles are atomicity (the note stands alone), connectivity (at least two meaningful links), organic growth (no over-structuring into rigid taxonomies), and continued dialogue (the note sparks further thinking). The agent checks each principle before filing any note.

Can this agent switch expert perspectives for different tasks?▼

Yes, it triangulates domain, task type, and output form to pick a matching expert such as Feynman for learning, Munger for strategy, or Ogilvy for branding. Every reply opens by stating which expert perspective is being applied.

How does the agent handle deep reading of books or long videos?▼

It produces a structure note with a five-question overview, a logic tree linking atomic notes, and a reading sequence, plus companion outputs like an execution plan and index note. This follows the deep-learning workflow referenced from the zk-steward-companion repo.

What are the limitations of a Zettelkasten agent workflow?▼

The method adds overhead to every note since linking, indexing, and validation are mandatory, which can slow quick capture. It also depends on companion skills like link-proposer and index-note that live in a separate repository and must be installed separately.