How I Use Obsidian + Claude Code to Run My Life

Summary

Greg sits down with Internet Vin to walk through pairing Obsidian (plain-text markdown notes) with Claude Code as a thinking partner. The pitch: stop re-explaining context to agents. Point Claude Code at your vault and let it read everything you’ve already written about yourself, your projects, and your beliefs.

The setup is simple. Obsidian becomes the long-term memory — daily notes, project files, references, journal entries — all in plain markdown. Claude Code, running locally, has filesystem access to the vault. From there you can ask questions that span months of notes, generate summaries, refactor your own thinking, and notice patterns you wouldn’t catch by scrolling. Vin demonstrates custom slash commands and prompt files that turn the workflow into reusable rituals.

The strongest argument is around self-observation. Most people underuse AI for introspection because the model has no context on who they are. With a real archive of your own writing piped in, the model can reflect things back you didn’t know you’d said — patterns in decisions, recurring blockers, contradictions between what you claim and what you do. Less productivity hack, more cognitive prosthetic.

For anyone running on AI-assisted workflows: the question stops being “which model is best” and starts being “what context am I giving it, and how persistent is it.”