# AI-Ready Second Brain
How to architect a [[Personal Knowledge Management (PKM)]] system that AI can consume as context. Not just "use Obsidian." The specific design principles that make a PKM system AI-accessible.
## Design Principles
1. **[[Atomic notes]].** One idea per note equals one loadable unit of context. AI can grab exactly what it needs without parsing a 5,000-word omnibus note.
2. **Consistent metadata.** Tags, properties, dates for filtering. AI can query "all notes tagged ai/governance created this month" only if the metadata is reliable.
3. **Plain text Markdown.** AI-readable, not proprietary formats. No vendor lock-in, no binary blobs. Every note is a text file an LLM can ingest directly.
4. **Rich linking.** Wikilinks AI can follow to traverse your knowledge graph. Links are the paths AI uses to find related context. See [[Knowledge Graph (KG)]].
5. **Clear folder structure.** Predictable locations. AI can find things by convention without searching the entire vault.
6. **Naming conventions.** AI can find things by pattern. "Person notes are in 36 People" is a rule an agent can follow.
7. **Identity notes.** Who you are, for the [[AI Master Prompt]]. Values, beliefs, writing style, goals. The foundation AI needs to act on your behalf.
8. **Skill files.** Codified procedures. Not "how I usually do X" in your head but written steps AI can execute. See [[Context-as-Code]].
9. **Memory systems.** Accumulated context that grows over time. What AI learned in past sessions carries forward.
## The Progression
Each step adds a capability AI can leverage:
**Basic vault.** Notes exist. They're in Markdown. AI can read them, but finding the right one is a needle-in-a-haystack problem.
**Organized vault.** Folder structure exists. AI knows where to look for specific note types. Finding relevant context goes from O(n) to O(1).
**Tagged vault.** Consistent tags on every note. AI can filter and query. "Show me all permanent notes about context engineering" actually works.
**Linked vault.** Wikilinks between related notes. AI can follow connections, discover adjacent context, build richer understanding. The [[Knowledge-Context Pipeline]] activates.
**AI-ready vault.** Identity notes, skill files, memory systems, context hierarchy. AI doesn't just read your notes; it understands your thinking, follows your processes, and improves over time. See [[Personal Context Management (PCM)]] and [[PKM-to-AI Readiness]].
## What Makes It Work
The gap between "I use Obsidian" and "my PKM is AI-ready" is not tooling. It's discipline. Consistent metadata. Atomic structure. Named conventions. The same practices that make a PKM useful for you make it useful for AI.
## References
## Related
- [[PKM-to-AI Readiness]]
- [[Personal Knowledge Management (PKM)]]
- [[Atomic notes]]
- [[Knowledge-Context Pipeline]]
- [[Personal Context Management (PCM)]]
- [[AI Master Prompt]]
- [[Context-as-Code]]
- [[Knowledge Graph (KG)]]
- [[Tolaria]] — markdown PKM app with native MCP server, designed AI-native