vault

odysseus0's avatarfrom odysseus0

Format conventions for markdown knowledge vaults. Use when writing notes or when format/syntax questions arise. Covers the dual-reader principle (human + AI), YAML frontmatter, wikilinks, and sources.

⭐0stars🔀0forks📁View on GitHub🕐Updated Jan 7, 2026⬇️1downloads

When & Why to Use This Skill

This Claude skill provides comprehensive format conventions for markdown knowledge vaults, specifically optimized for the 'dual-reader principle.' It ensures that digital notes are perfectly structured for both human readability in tools like Obsidian and seamless parsing by AI agents. By standardizing YAML frontmatter, wikilinks, and source attribution, it creates a high-quality, interoperable environment for long-term information management.

Use Cases

  • Personal Knowledge Management (PKM): Organizing and standardizing notes within Obsidian or Logseq to ensure long-term clarity and searchability.
  • AI-Enhanced Research: Structuring markdown files so that AI agents can accurately extract metadata and navigate internal links within a local knowledge base.
  • Technical Documentation: Establishing a consistent style guide for markdown-based documentation that requires both human review and automated data processing.
  • Digital Gardening: Implementing the dual-reader principle to make public or private knowledge bases accessible to both users and web crawlers/LLMs.
namevault
descriptionFormat conventions for markdown knowledge vaults. Use when writing notes or when format/syntax questions arise. Covers the dual-reader principle (human + AI), YAML frontmatter, wikilinks, and sources.

Vault Format

Dual-Reader Principle

Notes are read by both humans (via Obsidian/editors) and AI (raw markdown). Write so both have a good experience.

Avoid:

  • Dataview blocks (AI sees dead code)
  • Content that depends on embeds ![[]] resolving
  • Heavy callout syntax

Use freely:

  • Wikilinks [[Note Title]] (both can follow)
  • YAML frontmatter (both can parse)
  • Standard markdown

Note Structure

---
tags:
  - tag1
  - tag2
---

# Title

Content here.

## Sources
- (Author, Year) for claims

Conventions

Frontmatter: YAML with 3-7 specific tags

Links: Inline wikilinks [[Note Title]] or [[Note Title|display text]]

  • No separate "Related Notes" section - links belong in context

Sources: Inline attribution (Author, Year), full references in ## Sources section