Obsidian is a note-taking app that works on local markdown files. Claude Code is Anthropic’s AI running in your terminal - it can read and write those same files. Together they’re a different thing from chatting with an AI in your browser.

Want to set up the system? Skip the philosophy - use the CCO Setup Wizard or clone the template on GitHub.

The Core Insight

You control the context. The AI doesn’t guess it.

Instead of dumping everything into one endless chat history and hoping the model “remembers” the right things, you:

  • Explicitly choose which files the AI reads for each conversation
  • Save outputs back into your own file system
  • Reuse those outputs as context for future conversations

This creates a compounding loop - each conversation produces refined thinking that becomes input for the next. And you decide what goes back into your canon. Not everything the AI produces is worth keeping.

Why This Combination Is Powerful

1. Precise Context Control

Web chat: “Here’s everything I’ve ever said to you, try to figure out what’s relevant.”

Claude Code + Obsidian: “Read these 3 specific files. Ignore everything else.”

You choose the lens width for each conversation - narrow (“only my health notes”), medium (“my travel planning folder”), or broad (“my life philosophy + current projects”).

2. Outputs Persist and Compound

Web chat: insights get buried in chat history you’ll never revisit.

Claude Code + Obsidian: the AI writes directly to your notes. The output is the documentation. Next conversation builds on it. You can pause, come back tomorrow, and pick up exactly where you left off - the state is in your files, not trapped in a chat window.

3. The Frictionless Flywheel

  1. “Read these files to understand my context”
  2. AI synthesises, asks clarifying questions
  3. Back-and-forth refines the thinking
  4. “Write the output to this location”
  5. Output becomes input for future sessions

No copy-paste. No reformatting. No “where did I save that?”

A Concrete Example

I started with a messy document of pasted ChatGPT/Claude memories, scattered travel notes, and half-formed ideas. Over a weekend:

Travel planning — a 4-month trip itinerary with routing logic, conference schedules, friend visits, anxieties and mitigations. Multiple iterations refined through rapid back-and-forth. Draft emails ready to send.

Personal context system — a core CLAUDE.md (~80 lines) with essentials any session should know, plus 7 modular context files (Technical Infrastructure, Photography, Health, Relationships, Spiritual Practice, Philosophy, Key People) loaded selectively per topic.

Quick captures — the friction is so low that even small things get documented instantly:

❯ summarise Caplan's "The Case Against Education" in 1000 words
  and put it in 05 Resources/Personal philosophy

● Write(05 Resources/Personal philosophy/The Case Against
  Education - Bryan Caplan - Summary.md)
  ⎿  Wrote 87 lines

One sentence. Instant documentation. That summary is now part of the knowledge base, ready to be referenced in future conversations.

Worked Example: Actual Back-and-Forth

Here’s a real excerpt. We were building a personal context file:

Me: “Create a context file about my philosophy and worldview”

Claude Code: reads existing notes, writes a draft

Me: “The ‘Background’ section is great, but I can’t justify elevating it over the other aspects of my life philosophy”

Claude Code: “Fair point. The signaling framing got elevated but Internal Locus of Value, My Machine philosophy are all in the modular file. Inconsistent. Two options: 1) Remove it entirely, 2) Brief mention + pointer…”

Me: “option 2”

Claude Code: revises to a one-line pointer

Me: “do a final proofread”

Claude Code: “Found one issue: Line 66 ‘Health & optimization’ uses US spelling, but the filename also uses US spelling. Options: 1) Leave as-is, 2) Rename to Australian spelling…”

Me: “2”

Claude Code: renames file, updates reference

Dozens of rounds like this over a weekend. The output got sharper each iteration. The key insight: you don’t need to get your prompt perfect. Start rough, iterate, refine. The conversation is the refinement process.

Who This Is For

Good fit:

  • You value owning your data
  • You have multiple domains of life you think about seriously
  • You want AI as a long-term thinking partner, not just a chat toy
  • You’re willing to spend an hour getting set up for long-term payoff

Less good fit:

  • You just want quick answers to quick questions
  • You prefer everything to “just work” without any configuration

Series: Claude Code + Obsidian

Further Reading