Why Your Second Brain Needs an AI Companion

Updated 19 April 2026

Part 1 in my AI-Augmented Second Brain series

For the past 6 years, I’ve been extensively building a “Second Brain”, a collection of notes, meetings, weekly and monthly reflections, technical deep-dives, projects, and people. I structure this using Tiago Forte’s CODE/PARA methodology inside an Obsidian vault.

Working in technology and being exposed to so many projects, people, and information, my vault has grown over these years, with much of it still relevant and updated regularly. However, running this manually, I’d often forget to cross-link, disconnecting newer knowledge from the existing. This has led in many situations to me rediscovering areas or techniques that I’d already previously explored, distilled, and expressed in some form.

If you’ve built any kind of knowledge management system, whether it’s a Notion workspace, Confluence, or Obsidian, you’ve likely hit this same wall. Your second brain is only as good as the maintenance you put in, and it can be hard to scale.

This series is designed to solve that problem by exploring how AI agents, like GitHub Copilot or Claude Code, can not only build on your existing knowledge, but actively contribute to expanding it.

A Quick Primer on CODE and PARA

Here’s a 60-second version to get you up-to-speed:

CODE is the workflow engine, it explains how information flows through your second brain:

• Capture: Save insights, ideas, and information that resonates with you. Meeting notes, articles, quotes, observations. The goal is to eliminate the worry of forgetting.
• Organize: Sort captured information based on actionability, not topic. This is where PARA comes in.
• Distill: Summarize the key points, focusing on the most important insights, making information more accessible for your future use.
• Express: Turn distilled insights into something valuable. A blog post (like this), a presentation, a decision, or a new project. This is where captured knowledge becomes impact.

PARA is the structure for your second brain, where information lives:

• Projects: Active tasks or initiatives with clear goals and deadlines
• Areas: On-going responsibilities and knowledge domains
• Resources: Non-actionable content that supports learning
• Archive: Completed or inactive items from the above

I also incorporate a Techniques bridge as part of the PARA structure, that works to incorporate projects, areas, and resources into how-tos, frameworks, and content to Distill and Express.

What makes this system so powerful is its simplicity. Everything you capture has a clear place to live, and the workflow gives you a path from raw inputs to refined outputs.

If you want to dive deeper into Forte’s world of building a second brain with the CODE and PARA methodology, I highly recommend reading the book, Building a Second Brain: A Proven Method to Organize Your Digital Life and Unlock Your Creative Potential.

Where manual Second Brain management breaks down

Capturing is easy. Most tools make it trivial to jot down meeting notes, clip an article, or record your thoughts. If anything, we over-capture. This isn’t where the system fails.

Organizing is manageable, to a degree. But as your second brain grows into hundreds of notes, organization becomes more nuanced. Especially when you’re working out if you’ve already written something similar six months ago that you’ve since forgotten about or previously archived.

Distilling is where things get hard. Forte calls this Progressive Summarization, the practice of revisiting captured content and highlighting the most valuable parts, layer-by-layer, until you’ve extracted the essential insight. In practice, this is a huge effort on your part to revisit notes regularly, synthesize patterns, and produce those refined outputs to distill and express.

Expressing is where things can stall. The whole point of a Second Brain is to use what you’ve captured. But expression requires you to recall by finding the notes, connecting them, and producing something coherent.

And here’s the compounding problem, the stages reinforce each other’s failures. Poor organization makes distillation and expression hard, and when they fail, you lose motivation to capture in the first place.

What if your Second Brain could think with you?

This is the question that changed my approach in the last year. And given my extensive experience building and working with AI, now is the perfect opportunity to make the system participate in its own maintenance.

This isn’t about chatting over your data, it’s deeper. Using AI agents, like GitHub Copilot and Claude Code, to understand your Second Brain. Enabling it to know where every piece of information belongs, traverse the connections of the knowledge graph, and perform the cognitive heavy lifting of organizing and connecting your knowledge on your behalf.

This is what I’ve been building, and the approach rests on three core pillars that reinforce each other.

Pillar 1 – A Structured Knowledge Store

The foundation of my Second Brain is an Obsidian vault organized using PARA with one critical design constraint. It has been designed specifically to be machine readable. This means applying:

• Consistent templates for each note type containing section headings that follow predictable patterns,
• Frontmatter properties that encode metadata (dates, tags, etc.) as structured data,
• A README at the vault’s root that documents the entire system.

This approach not only serves as great documentation and standardization for yourself as a human, but these are the first things that an AI reads to understand how your Second Brain works. It’s your defined contract between you and AI.

Structure inherently enables AI, and without it, would need to infer the chaos and ultimately, will get it wrong more often than not.

Pillar 2 – Knowledge Graphs in Plain Markdown

Obsidian’s notes are essentially Markdown files, something that AI can already interpret well. And its knowledge graph is built up through [[wikilink]] links to other notes. So every time you write [[Azure OpenAI]] in a note for example, you create a directed edge in the knowledge graph to that Area. This then becomes a backlink without you touching the Area note at all.

Adding in Obsidian’s base query blocks, which are essentially dynamic, auto-populated tables, and you have a system where:

• An Area can automatically list every Resource associated with it,
• A Technique can automatically list every Area and Resource that influenced it,
• And the graph itself encodes where the gaps are in distilling and expressing what you’ve captured.

Essentially, the knowledge graph turns your Obsidian vault from a flat collection of files into a connected system that AI agents can easily traverse, query, and extend.

Pillar 3 – Specialized AI Agent Skills

A skill is a detailed, reusable set of instructions, stored as a Markdown file in your vault, that tells an AI agent exactly how to perform a specific knowledge workflow.

It’s important to understand that a skill is not just a prompt. A skill is a specific tool that an AI can leverage, with the instruction you provide, to perform that task on your behalf. For example:

Before creating a new Technique note, search all active folders and Archive for existing notes on this topic. If a match exists in Archive, reactivate it instead of creating a duplicate. Read the Technique template to understand the required sections. Populate frontmatter with tags and area links. For each area linked, verify the Area note exists. Search for related Resources that should be cross-linked. Set the last updated field to today’s date. Save to Techniques/<subfolder>/<title>.md.

The level of specificity, grounded in your vault’s structure, templates, and conventions, is what separates a reliable Second Brain agent from a Q&A AI chat that hallucinates.

For my own structure, I built five skills that cover:

• Note creation: creating new notes with a knowledge base scan, deeper MCP/tool extended knowledge retrieval, and automatic cross-linking.
• Note update: enriching existing notes, adding links, and decomposing overloaded notes into connected structures under the PARA system.
• Note retrieval: providing you the ability to use AI to trace the graph across all your notes to answer your queries, for recall, distillation, and expression.
• Weekly journaling: based on my notes and meetings captured, synthesizes activity into a structured, reflective weekly journal of what went well, what didn’t, and what can I learn from the week.
• Monthly journaling: distills weekly journals into monthly insights and patterns, useful for further reflection and great for personal and career check-ins.

Each skill follows the same instruction pattern: understand the structure, scan the knowledge base, gather context, produce outputs that strengthens the graph.

How the pillars reinforce each other

As I’ve established working with AI, structure enables AI. The PARA system and template consistency give AI reliable patterns to follow. Every time a skill is invoked and a note is recalled, created, or updated, AI knows exactly what to expect and do.

AI fundamentally strengths your Second Brain. Every skill-driven note change ensures cross-links are established, fills in gaps, and surfaces archived notes for reactivation. This is maintenance work that I know I’ve missed in the past. Over time, AI is actively improving the knowledge graph’s density, but more importantly, the accuracy.

Knowledge graphs guide AI. Backlinks and relationships between notes show AI what’s connected, what’s missing, and where to focus. The graph is ultimately both the input and output of AI’s work on your Second Brain.

This process creates a self-reinforcing loop. The more you use the system, the richer the graph becomes, and the better AI performs with it. The system compounds rather than decays over time.

What this series will build

Over the following articles in this series, I’ll walk you through building this system. Together, we’ll focus on:

• Why Your Second Brain Needs an AI Companion (you’re here!)
• Designing a Machine-Readable Knowledge Base with Obsidian
• Building out your Knowledge Graph in Markdown
• Your First AI Skill for Knowledge Management
• Leveraging the AI-Augmented CODE Workflow
• Making the AI-Augmented Second Brain Your Own

By the end, you’ll have a working Obsidian vault with your own AI skills that actively participate in your knowledge lifecycle.

And I’m sharing a companion base repository, obsidian-ai-second-brain on GitHub, for you to get started with your own Obsidian vault.

The Shift in Mindset

If there’s only one thing you take away from this article, it’s this:

Your Second Brain shouldn’t be a passive filing cabinet. It should be orchestrating itself to strengthen your knowledge over time.

The CODE/PARA methodology gives you the systems. Obsidian gives you the capabilities to deliver the methodology. And AI gives you the capacity to maintain, distill, and express at a scale that manual effort can’t match.

The methodology doesn’t change, the capability does.

Found this article useful?

Thank you for taking the time to read this article. I’m committing to sharing more of my knowledge through my blog and open-source projects! You’ll also catch me in casual conversation on Bluesky and LinkedIn too.

If you enjoy what you see, please consider subscribing to get a notification when new articles go live!

[mailerlite_form form_id=2]

---

This article is part of my AI-Augmented Second Brain series.

Next up: Designing a Machine-Readable Knowledge System with Obsidian.

You can find the companion template repository for an Obsidian vault at obsidian-ai-second-brain on GitHub.