We are seeing this work incredible well at work, where agents will build out a detailed picture of technical architecture, as well as any decisions/technical progress. But also, I have gotten an agent to keep me an Obsidian vault of all of the resources that I have read/consumed.

It’s taken a little while to get here though!

On the first iteration of using Claude with my Obsidian vault, I had it writing directly into the vault. This very quickly felt like something I didn’t want to happen.

Firstly, it became hard to tell which notes were written by myself, and which were created or edited by an agent. For me, this is an important distinction to make - my goal with using a LLM in this way is to help my thinking, not take over it.

Secondly, the amount of notes I was able to add made traversing my vault pretty hard. Adding every Granola transcript I have ever collected (500+) or bringing in all of the markdown transcripts of Lenny’s podcast, was an interesting proposition to be able to use as reference material. Not very fun to navigate manually!!

Having a separation of concerns, and keeping a second vault which was to be used only by agents seemed like a sensible choice to explore.

On reading Karparthy’s notes on “LLM Wiki’s”, it seemed to tick all of boxes I wanted to achieve, and supported the idea of having a separation between human and LLM-generated content.

“the LLM incrementally builds and maintains a persistent wiki— a structured, interlinked collection of markdown files that sits between you and the raw sources. When you add a new source, the LLM doesn’t just index it for later retrieval. It reads it, extracts the key information, and integrates it into the existing wiki — updating entity pages, revising topic summaries, noting where new data contradicts old claims, strengthening or challenging the evolving synthesis. The knowledge is compiled once and then kept current, not re-derived on every query.”

The main benefit I have found, is the volume of content which you can feed it, and how well it does to find themes/patterns/structures and to present them.

If I read five articles in a day - generally I will pick one or two to write some notes on that have resonated with me the most, or I have found the most interesting. The rest might be useful context at some point, but are not useful to me now.

With an LLM-only vault, I would drop all of the articles in, and it would extract all of the useful context, and structure it in a way that it is easy for me to come back to later.

It doesn’t replace the flow I have currently of reading and making my own notes, it compliments it and works along side it.

The Structure of the Vault

Kaparthy’s structure is pretty simple:

/raw : this is where you add your original sources - documents, articles, X threads, videos etc etc.

/wiki : a directory of LLM-generated markdown files. Summaries, entity pages, concept pages, comparisons, an overview, a synthesis. The LLM owns this layer entirely.

/schema : the inner workings of the vault. CLAUDE.md or AGENTS.md that focuses on structure, what the vault conventions are and any workflows to follow.

I took this structure and ran with it. I named my agent-only vault “Foundry” - the idea being that stuff gets poured in raw and comes out the other side as something more useful.

Here is the actual folder structure:

foundry/
├── CLAUDE.md       — the schema
├── inbox/          — raw drops: PDFs, URLs, pasted text
├── sources/        — atomic source notes, one per item
└── wiki/           — concepts, people, queries, index, log

The flow is simple. I drop things into inbox. Claude reads them and writes atomic source notes into /sources. When there are enough sources on a single theme, Claude compiles a concept page into /wiki.

One decision I am glad I made is that I kept the front-matter in Foundry consistent with my personal vault. Same tag taxonomy, same date format, same #type/, #area/, #keyword/ conventions. It meant I could cross-link between the two without the stitching feeling ugly.

The thing that makes this all hold together is the CLAUDE.md or AGENT.md file. It is the file that tells your agent what this vault is for and what it can and can’t do. Things like:

• my personal vault is read-only - never write to it

• every concept must cite its sources

• don’t spin a concept page out of a single source, wait for two

• no emojis, no TODO comments, no speculative helper files

Writing this file is probably the most important thing you do when setting up a vault like this, and working with any agent, because this is how they follow along and know what to do. It is the thing that stops the agent from doing agent things - sprawling, inventing, helpfully rearranging your folders. It can be updated on the fly depending on your preferences as you go along.

The Skills

As per Kaparthys original document, he suggests four skills to help with the processing. Each one has one job to do.

/foundry-ingest — processes whatever I have dropped into the inbox. Reads each source, writes a clean atomic note with a summary, key points, and a short “Claude’s notes” paragraph (what is interesting, where it connects, what it contradicts). Updates the index, appends to the log, clears the inbox.

/foundry-compile — scans the sources folder for themes that have at least two sources behind them. Either extends an existing concept page, or spins out a new one. Themes with only a single source get logged as Candidates and have to wait for more evidence. If you are old-school “tools for thought” like myself, this is essentially something that looks over your “seedlings” and seeks to create “evergreens”.

/foundry-ask — answers a question across both vaults. Writes the answer into a dated query file in /wiki. Every claim cites the source. Gaps get captured as Open Questions. Anything essay-shaped ends up as a “Prompt for 2026.”

/foundry-lint — a health check. Counts orphans, finds dangling links, flags candidates that now have enough sources to compile, spots keyword drift.

The pattern that makes this work is that each skill does one thing, has a clear output, and has a list of things it will not do. If you give an agent a vague brief across a folder full of your notes, you end up with a mess. If you give it a narrow job with explicit boundaries, it does the job well.

How it works in Practice

Most of the time, Foundry is a background process. I read something, I drop it in the inbox, I move on. You can find a video at the bottom of this page of my running through the process from start to finish.

A typical day looks something like this:

• I read a couple of articles in the morning. I share it to the inbox.

• I listen to a podcast. I drop the transcript in.

• I come across something that sticks in a book. I paste it into a markdown file and drop it in.

By the end of the day there might be between five or ten items in the inbox. I run /foundry-ingest. Claude processes every item into a clean source note, cross-links where appropriate, and clears the inbox.

Once or twice a week, I run /foundry-compile. If there are enough sources behind a theme, Claude writes a concept page. Each concept page has the same shape - what it is, why it matters, key points, evidence across sources, open questions, and “Prompts for 2026.”

That last section is the payoff. “Prompts for 2026” is a list of essay-shaped questions where the concept intersects with things I have already written in my personal vault. They are things I could sit down and write about right now.

When I actually want to use the vault, I run /foundry-ask. I ask a question. Claude researches across sources and concepts, answers with citations, and the good answers get filed back into the vault. This is particularly useful when writing an article, or if I am thinking in particular about

An example of using this in anger recently was when we were discussing a change at work, in particular merging two smaller teams into one larger team. I asked the vault for evidence regarding merging teams in the context of AI-first organisations, and it gave me an answer based on the content I had given it. Sure, it can be biased, but it will flag up when it feels like it hasn’t given a balanced argument - and the value for me in this instance was the resources it directed me back to so I could re-read them with a new question top of mind.

Referencing My Personal Vault

This doesn’t replace my personal vault, and nor would I ever want it to! They both serve completely different purposes (I will touch in this next week). The intentional act of writing and curating an Obsidian vault is something I have taken great pride in, and it has bought me hours and hours of enjoyment over the years. Foundry is there to help, support and feed it.

My personal vault is read-only to Claude. Claude can read it, quote it sparingly, and link into it with [[2026/...]] style references when a Foundry note is informed by something I have written. But Claude cannot write, edit, move, or delete anything there.

My writing needs to stay mine. The whole point of setting it up this way is to keep my thinking mine - Claude can do the heavy lifting on everything else. Writing in my Obsidian vault, making notes, processing thoughts is still very much something I make time and space for.

I love writing in my Obsidian vault. It is one of the highlights of my day to sit with a coffee, process my thoughts and do some writing. For the years I have been using it, it has helped me distill my thinking and I genuinely see this as being a practice that will continue to enable me.

Being able to utilise AI to help me to write clearer, think deeper, understand my own nuances more deeply and build connections that I missed myself is my goal.

Having this agent-only vault that can digest and articulate knowledge faster than me, and help support what I was getting from my initial note-taking practice is something that I am genuinely reaping the benefits from.

Other Use Cases

I believe this paradigm is going to become more and more common, not just in personal knowledge bases, but for company wide knowledge bases also.

For example, we have started using an agent kept wiki at work to keep on top of architectural/technical changes across the stack. Whilst Foundry and my own content wiki is most useful for myself only, this is useful for a team or a collective. Below is an example of this vault being referenced in Slack.

Another example is a personal use case I have for helping keep track of various parts of my day to day work an an engineering manager. I have designed the LLM wiki to keep track of the below topics as they happen through Linear, Granola, GitHub, Slack & Incident.io.

1. If you want to try the content vault for yourself, then I have open sourced it. This is set up to work with Claude only (for now). All you need to do is download it, and open it in Claude and it should work.

2. I have filmed and recorded a video of me using the vault and running through how I use it on a daily basis - how I capture content with Obsidian web clipper, and how this content progress’ through the vault. This is something I am going to be doing more of from now on, and with every article there will be a video demonstration and deep dive. This will be kept as a benefit for my growing list of paid subs <3. I am offering a discount for a short term for annual plans so there is not a better time to join!

This practice has been utilised by thinkers & creatives alike since the middle ages - and in my opinion it is one of the highest value things you can be doing. Not only does it come in value to have a reference point when producing your own work - it is also inherently useful for thinking deeper, being able to look for and build connections between different topics or different resources to create new ideas.

The commonplace is a curation of your taste and thought.

My commonplace (both in physical form as it used to be, and digital form) has helped me out tremendously both in my career in tech, but also with my writing.

It’s ultimately a repository of your own context. With the ability to give AI the context of your Obsidian vault, and therefore access to your commonplace/second brain etc - it is becoming even more of a valuable practice to start collecting and building connections between ideas.

This article will be a deep dive into how I structure my Commonplace folder in Obsidian, and how I am using it with AI to help me think more deeply and more clearly.

The Structure and Contents

My commonplace folder is by design unstructured. There are no sub-folders, and I would argue for the most part having a rigid structure defeats the point of it (more on this in the navigation section).

my-vault/
├── 3. commonplace/
│   ├── advocate for simplicity
│   └── ai learning resources
│   ├── creativity vs productivity

I put anything in here that I have found online that I have particularly enjoyed or thought was interesting or useful in some way - quotes, book notes, video notes, images, sections of courses, designs I have liked, poems - and then I generally build on top of them with my own thinking or writing. It doesn’t need to be made up entirely of things you agree with either, there are a few notes I have written that follow the format of “I disagree with X, this is why”.

It also houses ideas/thoughts that I have had throughout the day. This could start as something as simple as creating a note for a single line/idea, but over time it is something I keep coming back to and building upon.

As a general rule, I try and keep my notes pretty specific to one topic or idea. So for example - if I am reading a book, I will have a parent note which captures some high level thoughts about the book, and then separate ideas I have taken from said book will have their own note.

An example below could be some of the notes I have taken from reading The Creative Act by Rick Rubin:

my-vault/
├── 3. commonplace/
│   ├── the creative act
│   └── looking for clues
│   ├── how to be creative
│   ├── create space for thought

You can see above I have a note “the creative act”, but I also have notes for ideas that I have taken from that book. They all link up via tags and backlinks - but are able to stand alone as their own independent notes.

Having this separation is important in my opinion, as it allows you to make connections with the specific ideas themselves, and this is easier when the notes are smaller and more atomic. For the engineers, it’s like building in components or functions - each one is able to be referred to independently. For instance “looking for clues” might connect up with an idea I have read about in another book, but not be related to “how to be creative”.

Key Takeaways

• Start collecting anything you find useful when reading, browsing and thinking.

• Don’t be tempted to create structure through folders. Building connections through tagging, backlinks or AI is much more efficient.

• Keep notes fairly small in context. When a note gets too big it is ok to split it down into multiple notes.

Navigation and Connection

Looking at the example note from my Commonplace above, I have multiple forms of navigation going on within, but the foundation of them is simplicity, hence this is a pretty short section.

I have the tags at the top of every note - the purpose of which is to help me “find” this note again. As an example, I could be writing about creativity and I can simply look at the “creativity” tag and it will show me every note that I have put under that tag.

I also have a backlink to “Chasing craft”, which is another note in my commonplace that I believe is relevant. The purpose of backlinks is to “link” and “connect” ideas.

Over time this compounds and you build up a resource of connected themes and ideas.

Key Takeaways

• Tags should be used for “finding” the note again.

• Backlinks should be used for “connecting” ideas together.

• Keep this navigation as simple as possible. Don’t overwhelm yourself by trying to be too clever. Simple tagging and back-linking is enough to build meaningful connections.

Using this with Claude

As I have discussed in previous articles - hooking Claude up to your Obsidian vault and allowing it to use your notes as context to help with your thinking is a really interesting and powerful proposition for using it as a ‘tool for thought’.

The commonplace is a great resource to allow Claude to have access to, because what you collect in your commonplace speaks a great deal about you as a person, what you value, what you thinking about and what interests you. Using AI effectively requires you to provide it with the sufficient context. Whether that is a codebase and specific details of what task you want it to achieve, or a commonplace that houses all of your personal ideas and interests.

For example - here I asked it to reveal what it understands about me as a person from my commonplace folder alone.

Prompt:
Can you build a short list of some of what you know or think about me from the commonplace folder?

Output:
Based purely on what the notes reveal — not flattery, just observation:

You care deeply about quality, and it bothers you when things aren’t done well. The craft notes aren’t aspirational in a vague sense — they’re specific about the kind of people you admire (Cillian Murphy, Daniel Day-Lewis, Redzepi). These are people who made their discipline a life’s work, not a job. That’s a high bar, and you hold it for yourself too.

You’re somewhere between consuming ideas and acting on them — and you know it. The project note is the tell. You can describe exactly what you want to make, why it matters, what it would feel like. The gap isn’t vision; it’s the first move. The collection itself is, in part, a way of staying close to something without having to commit to it yet.

Simplicity is a value, but complexity keeps finding you. The task management note is a quiet self-portrait: you can build elaborate systems, you know they don’t work for you, and you still find them appealing. The reminder you left yourself — “if this works well, stick with it. Don’t be sold” — suggests you’ve been here before.

You’re professionally embedded in AI but personally uncertain about what it means. You’re learning it, working with it, thinking about how organisations adopt it. But the project note imagines making something about what AI does to human memory and meaning. That’s a different relationship — more ambivalent, more philosophical. Both are real, and they sit uneasily together.

The cancer note matters to you in a way that’s hard to articulate. You saved a stranger’s grief about love, loss, and what remains when everything else falls away. In a collection mostly about becoming and achieving, that note is doing something different — it’s a reminder of what anchors all of it. You know the stakes aren’t really about craft or manifestation.

You want to be great at something, and you want that ambition to feel legitimate. The manifestation notes, the grit notes, the “I want to be one of the greats” note — there’s a recurring thread of permitting yourself to want big things. That suggests the ambition is real but not yet fully owned.

The short version: someone who thinks carefully, feels things seriously, and is in the middle of deciding what to actually do with both.

Whilst there is a lot there, it is able to articulate me and what I hold dear incredibly well. It spots common themes, and builds a picture of who you are. I have been consistently surprised at how well it picks up on small nuances and can flesh out patterns that even I (as someone who considers themself generally pretty self-aware) have not been able to uncover.

More practically, you can then use this context to cross-reference and ask it questions about your thinking.

An example below, when asking about suggesting topics for me to write about - rather than doing what people typically do and saying something like “what articles should I write about X topic” - it can suggest things that it notices I take a lot of notes on but that are missing from my outputs.

Prompt:
Using this context, which topics do you think it would be interesting for me to write about, that I haven’t yet covered in the writing in my outputs folder?

Why do you think that would be good for me to cover, knowing my goals with my writing?

Output:
Here are six topics that are conspicuously absent from your writing — and why each one makes sense for you specifically:

1. “All the Answers, None of the Action”

Theme: The gap between knowing and doing

You’ve written more about productivity systems than almost anything else. This is the essay that earns that body of work — making the uncomfortable case that understanding a problem can become its own form of avoidance. That fluency in the language of change isn’t the same as change.

Why you: Your task management writing is honest and non-preachy. You’ve already built the credibility to say “I know all of this and it’s still hard.” That’s rarer and more useful than any framework. This is the essay your existing writing has been circling without landing.

Key Takeaways

• Use your contents of the commonplace folder to allow Claude to build a picture of the topics you are interested in and are thinking/writing about.

• Use this as a reference point for when you work with it.

• Ask Claude to disagree with your ideas, or provide new perspectives. This is a challenge to our thinking that we don’t often have the luxury of having unless we share it with others.

Limitations

It’s worth pointing out, I think there are some limitations with using just Obsidian for curation of taste and commonplace.

Obsidian is a text first tool, and it does this incredibly well. For me, imagery and aesthetic plays a massive role in things I find inspiring - and whilst I do collect some images in Obsidian, it is not efficient for keeping and traversing and building connections between lots of visual elements. I do use it

For this, I use a product called MyMind - I have been using this for over three years now, and have built up a collection of nearly 4500 cards.

The benefit of MyMind is that it is much more visual - I use it generally for saving images/videos/websites that I find inspiring. There are no folders or structure generally, it is a consistent feed of things that you have saved from around the internet.

Connections here are done by “vibe”, so if I were to select and image I can then look for other things that have a similar vibe, and it uses AI to surface these other relevant pieces of content. Below is an example where it is showing me images I have saved with the “same vibe” as the first image:

At the start of any visual project, this will be the place I turn to.

Obsidian is the main bulk of my written thinking and written organisation - MyMind is the majority of my visual thinking. I use the two of them in collaboration.

Every now and then there are some cases where I will move an image from MyMind to Obsidian, if I want to write about it in more detail. There are a few examples of where I have dived deeper specifically into why a particular visual has resonated with me, and the various things I notice about it.

If I were somehow able to hook MyMind up to Claude and use it in tandem with Obsidian, it would be fucking incredible - but this is not the case as of yet! From my experimenting with image generation and using Claude to build websites/applications, using good and cohesive reference material for visualisation is all the more important than the efficiency/effectiveness of the prompt - so I am predicting that as time goes on it’s going to be even more valuable to have a collection of visual references.

I don’t think there is a decent product that exists that allows you to collate a really good collection of both written content and visual content. However for me, I quite like having the separation of concerns between a “visual” app and an app for “writing”.

Leave a comment

Share

This is going to be a walkthrough of all my set up. By the end of this article, you will have a vault that looks identical to my own.

How Obsidian looks is a big reason I use it and have been using it for six plus years now. It’s simplicity is beautiful. We are very fortunate we have a piece of software and a tool that is flexible enough that we can customise it to be exactly how we want.

My vault is the fruit of six years of tweaking and getting it just right. I hope you enjoy.

Theming

The theme I use, and always have used, is “Minimal” by Kepano.

There are some plugins I will discuss later down the line that change the look and feel of the theme even more. However I love how clean and, well, “minimal”, it is by default.

Fonts

I use two main fonts in my vault.

iA Writer Quattro is the main font and this is one I get asked about a lot. Back in the day i used to use the writing app iA Writer - and the font and clean UI was one of my favourite things about the app. This feeling was something I wanted to replicate in Obsidian.

It is available as a free download here.

For the monospace font (which isn’t something that is used an awful lot across my vault) I use Geist mono by Vercel.

This can be downloaded for free here.

Do check out the pixel variety too! Really nice :)

Plugins

My ethos with plugins has always been to keep it quite simple. Hence I don’t use many plugins.

In all honesty, I found it a bit overwhelming the amount of different things that can be installed, and I was very hesitant to build workflows around plugins that were ultimately created by the community and there was no guarantee that they would be continued to be developed. I wanted to build a note-taking system that would last a long long time, and this was a brittleness I didn’t want to introduce.

I keep my usage to seven plugins, and I have been using these now for years. They are all installable via the community plugins finder, so I won’t include links to them specifically.

Calendar

This works really well with the daily note core plugin - and is something I use pretty heavily. It shows a calendar in the sidebar, and you can easily transverse to a particularly daily note. It also visually shows you the days where you created a daily note with dots.

I use this often when going back to the previous week to look for what I wrote on a specific day.

Hider

Hider lets you hide/change parts of the Obsidian UI, and this is a critical plugin for getting the clean/minimalist aesthetic.

The settings I use are as follows:

Hide tab bar: Off
Hide status bar: On
Hide vault name: On
Hide scroll bars: On
Hide sidebar toggle buttons: Off
Hide tooltips: On
Hide file explorer buttons: On
Hide search suggestions: Off
Hide count of search term matches: Off
Hide instructions: On
Hide properties in reading view: Off

Kanban

I use the kanban plugin specifically to keep on top of writing tasks in my vault. This isn’t one I use extensively, but its definitely a nice quality of life improvement to be able to visualise the state of the different pieces of writing that sit within my vault.

It lives within my “Outputs” folder.

Minimal Theme Settings

Another plugin focused on styling! This handily lets you control some of the different elements of the Minimal theme.

I get asked a lot how you change the look of the navigation in Obsidian, and this is where this is controlled via the “Text labels for primary navigation”.

My settings are as follows:

Colour scheme: Flexoki for light and dark
Background contrast: Default
Text labels for primary navigation: On
Colourful window frame: Off
Colourful active states: Off
Colourful headings: Off
Minimal status bar: On
Trim file names in sidebars: On
Workspace borders: On
Focus mode: Off
Underline internal links: On
Underline external links: On
Maximize media: On

Then for the typography I use the following settings:

Outliner

Coming to Obsidian from Roam Research, I got very used to working with nested bullets. This replicates some of the same behaviour that you found in Roam, and being able to traverse lots of nested bullets.

My writing has moved more away from nested bullets now, so I am not sure whether this is still completely necessary for my vault and the way that I work - however it does no harm so I keep it! :)

Style Settings

There are a tonne of options here to control every aspect of visual settings of you vault. However, most of what I keep here is default. The one thing that it does change however is the styling for my tags.

I use tags a lot, and every note I have in my vault uses tags at the top of the page that will act as metadata. Being able to move away from the standard “pill” shape of the tags and have plain text was important.

The only setting I have changed in this plugin is:

Plain tags: On

Terminal

The most recent addition, and I have installed this to use as an integrated terminal when speaking with Claude.

This is still something that I am working out though - I find myself using it sometimes, and other times using my default terminal (Warp) and having both Warp and Obsidian open at the same time. Only time will tell which I prefer I guess!

Custom CSS

I have only written custom CSS for one small aspect of my vault, which is the callouts.

IMO callouts are a pretty underutilised feature in Obsidian generally, in that I rarely see others using them. If you have used Notion, you are likely familiar with the callout box, and that is something I use often there also.

Having a custom colour to match the monochrome nature of my vault was something I wanted to do - and it’s pretty easy in the sense you just create a .css file and then drop it in your ./obsidian/snippets folder. This can then be loaded under the “Appearance” tab in settings.

The CSS for my callout snippet is:

.callout[data-callout=”north-star”] {
--callout-color: 158, 158, 158;
--callout-icon: sparkle;
}

This controls the colour, but it also controls the icon that appears at the top of the callout. I use the callout in my daily note for my daily goals, and have used a star icon.

Conclusion

The flexibility of Obsidian can be overwhelming for folks - and this is one of the biggest sticking points I have found as to why people don’t continue with it. Taking a small amount of time to get your vault feeling as you want it to I think is a worthwhile investment of time.

Of course, its all very tempting to continuously tweak things (been there done that) but I am pleased to say that these configurations have been pretty consistent in my vault for the last four/five years now and my tweaking is kept to a minimal!

If there is any other aspect of my vault that you would like to find out more about, or for me to dive deeper into then please let me know by leaving a comment below or reaching out to me directly.

There has been a lot of information shared recently about Obsidian and Claude - some of it I believe is great, some I believe is not so great. I wanted to make sure that I had a robust and thoughtful way of working that stayed true to my values, rather than contributing something brittle when there was a lot of hype happening and noise.

The last couple of months, I have been entirely focused on finding the best way to integrate Claude into my daily work to help me get more efficient at both my job but also writing and thinking.

I believe now though I am ready to share my workflow. I am pleased with how it is, the value it is bringing to me, and I don’t see this changing drastically in the near future.

Feel free to take inspiration, or use it in your own setup. This has been working very well for me - and I hope it will work well for you also!

Also - please ask any questions as a comment. I am more than happy to help, offer guidance or just chat about my workflow!

1. File Structure

Initially, I started putting everything in to my Obsidian vault (meeting transcripts, generated files) however quickly realised this was not a good fit for me.

My knowledge graph was getting diluted by a lot of files/transcriptions - these are useful to speak with Claude about, but less useful to have visually in Obsidian. I try and keep a separation of concerns between notes that I have personally written, and things that are generated by AI.

I now have a dedicated Claude folder, which contains everything that I work with Claude on. There are three sub-folders:

Github
This contains both my personal codebases, work codebases and our company repository of skills/docs (affectionately named “Zkills”). Everything goes in there from PRDs, to customer research transcriptions to sales calls to Claude configuration files for our repositories.

I use Claude to parse a lot of this information, and reference company documents when making engineering decisions. This is becoming our default knowledge base over something like Notion - and through MCPs the flows are made much more efficient.

Example prompt: “Look at the latest PRD and create me a Linear project from the PRD. Create specific tasks for each stage”.

Meeting notes
I use a Raycast extension to export all of my Granola notes into their own dedicated folder. This has been really helpful in being able to use Claude to recall recent actions.

Example prompt: “I have a 121 upcoming with X person, what were the actions we took in our last meeting, and is there anything I need to do prior to our call?”.

Obsidian Vault
My Obsidian vault is where all of my writing happens. Daily notes, commonplace notes, written thinking, articles and social media posts. This is a practice I have been keeping now for the last six years with Obsidian specifically, and then before that other tools (Roam, Notion etc).

Claude has been ridiculously valuable when given access to all of this context, and has been transformational in how I think and write.

This article is about this section specifically. Let me dive deeper...

Key Takeaways

• Setup a dedicated “claude” folder on your machine. All context lives within here.

• Separate AI generated notes away from your Obsidian vault. This helps keep the vault/knowledge graph clean. Mass amounts of generated transcripts/notes in your vault is not actually that useful on the knowledge graph - however Claude having access to them is useful.

2. Obsidian Vault Structure

The main structure of my Obsidian vault has not changed with the introduction of Claude.

My ethos from starting to use Obsidian years ago has always to keep things as simple as possible - and I have found that this works well, especially when working with AI. My setup has not overly changed in the last six years, because it has been created around how I like to think and how I work.

Across my Obsidian vault I utilise three main forms of navigation: folders, tags and backlinks.

Folders

I keep my folder structure very simple. I have experimented with no-folders, but like this minimal folder setup.

As you can see from the lefthand sidebar, I have 5 folders:

Polaris
Polaris is my North Star section of the vault - this is where I keep my main goals, aspirations, “top of mind”, amongst other things.

This has been one of the biggest wins in terms of Claude efficiency, as it uses this folder as a guiding reference point for when it does my thinking. (More on this later).

Logs
This is where the daily notes live. I treat a daily note as a scratchpad, writing things down on the fly. An example can be seen in the picture above.

Commonplace
This is the main part of my vault. Lots of separate notes live in there, and each represent a different thought - staying true to the original concept of a “commonplace book”.

Outputs
This is where I do all my writing for things that are going to be shared - currently writing this in the Outputs section of my vault :)

Utilities
Bits that help the vault work - images, templates, canvas’, people etc

Tags

Tags is one of the most criminally underutilised ways of navigating through an Obsidian vault, in my humble opinion. I don’t see many people working in a similar way, and I think it is a shame.

Firstly, you can nest tags. Looking at the left sidebar above, in the tag pane I have “parent” tags, these can be opened and sub-tags kept within them.

Secondly, tags are very portable and keep with the “file over app” mentality. I have tried migrating my vault to other applications (Bear notes, as an example) and it works seamlessly.

Thirdly, tags works well with Claude. I can get more specific by prompting “look for notes within the articles tag” and it will filter them.

Each note I create has a set of tags that act as metadata at the top of them. Above is a screenshot of the tag structure as it currently stands.

Backlinks

Backlinks are used to make connections manually between notes. It is a good way to leave trails for yourself to help with your thinking.

The example above is a note that I made when reading an article on manifestation. There were several ideas I took from this article, and I decided they warranted their own notes. Building a connection between the two means that original context is not lost.

Also - Claude is very good at suggesting backlinks and connections between notes that you might have missed. I have a command specifically for it.

Key Takeaways

• Keep vault organisation simple and predictable. Ideally this is something that you pick and should work with for a long time without changing.

• Tags are an under-utilised source of Obsidian vault navigation (in my humble opinion!). They work well natively, and with Claude!

• Understanding how your vault is organised, can help you make Claude more efficient.

• Claude is great at building connections and surfacing connections that you may have missed.

3. Using Claude with Obsidian

At first I was skeptical about how valuable this would be - but as time goes on I become more and more impressed just by how helpful Claude is for my own thinking and workflows. It continues to surprise me at how well it is able to build connections in my thinking, challenge my assumptions and prod at areas that I need to dig deeper into

One of my Favourite Examples

Recently, I found out I was going to be becoming a father for the first time. It’s been a long road - and this is a whole new set of emotions/thoughts for me. Everything I think of goes into my vault.. I write every little thought, idea, concern, anxiety down onto these pages - I have done this for years.

Claude picked up that whilst I wrote about everything, I didn’t write much about the thought of being a father - despite it being something that is top of my mind. It asked why that might be.

This was something I hadn’t picked up on - but it prodded a vulnerability that hadn’t even crossed my mind. Why was I avoiding those thoughts?

This is something it continues to help me with and it features heavily in my responses especially when speaking about.

In the screenshot above it flagged up a connection between a fiction writing project I have been working on. “Do the math”. It is getting snarky at my own emotional incompetence and naivety. 😆

Utilising “Top Of Mind” & “Polaris”

In my “Polaris” section of my vault, I keep everything that is important to me. One of those notes is “Top Of Mind”.

I can use this as a reference point for Claude to refer back to when I am thinking of anything new.. It holds me accountable for losing focus on things and chasing shiny objects.

This is a document that I update every couple of weeks generally - it has been really valuable for making Claude more focused and efficient. The descriptions don’t need to be long here at all, as long as the point is clear.

The Life Razor is a concept I picked up from Sahil Bloom - where your life’s mission is defined by one sentence. Coming up with this has been a valuable exercise it itself I have used Claude for - however, now I have this statement it is again really useful as a reference point.

Example prompt: “How are my current actions aligned with whats top of my mind?”

Example prompt: “I am thinking about taking on X opportunity.. How does this help or detract from my life razor?”

Idea Reports

I will likely do another post/article somewhen on some of the commands/prompts I have written for Claude specifically to use in Obsidian.. but one of my favourite is the “Idea report”.

The prompt here is: “I have a few hours of spare focus time - what should I work on”?

This then breaks makes suggestions for different areas - things to build, people to reach out to, ideas to investigate or build on top of as well as actions that I may have missed.

It has the context of everything that I have been thinking about, and the output above you can clearly see the cross-referencing between my thoughts as well as other things I have been writing about.

Making Use of Vault Structure

Having a simple vault structure can make navigation much easier. An example is here below..

I am starting to write about “craft” as a topic, and I want to pick up the references from within my vault. I know I have a craft tag, so I can ask Claude to pick me up the files that sit within that tag.

Then, we can dive down on specific notes or look for specific themes across each of them.

Example prompt: “Show me all of my Obsidian notes that have the tag craft”

Example prompt: “Is there any connected ideas amongst these resources that would make an interesting topic to write about specifically?”

Key Takeaways

• Claude performs incredibly well at looking for threads that haven’t been explored, and asking questions for deeper thought.

• Introduce a “polaris” type document/folder - this should include what is top of your mind, goals, aspirations, personal values. Claude will reference this often and these principles help guide it’s thinking.

• Create commands/prompts that are specific to your ways of working.

• Keeping a simple vault structure will help the efficiency of Claude.

Conclusion

What I have found with Obsidian, pretty continuously over the last few years, is there is a lack of information that focuses very specifically on how to use it day to day. As a tool that provides little guidance and a lot of flexibility, it is very intimidating to get started.

There are great tutorials on how to build a specific thing, but never very many on how to build a workflow that covers everything. Or there are great tutorials for people doing a research job or that are studying full time.

Claude is another such tool, and it is becoming more and more challenging to find real world use cases amongst a barrage of hype.

Hopefully the above has given people a decent foundation for which to get started with Obsidian and Claude - two incredibly powerful tools.

As an idea it is deceptively simple - define what needs to be worked on, compile a list and then see it through. We have a tendency to over complicate it however.

More and more I find myself thinking about how simple productivity is as a whole, and how we have become trapped into trying to overcomplicate the art of “being productive”. I have fallen foul of this trap too.

We have a glut of tools that allow us to manage our tasks and stay on top of things. There is a whole ecosystem of productivity guru’s who each claim to have the golden nugget of information that will help us become our best productive selves.

The landscape is quite overwhelming, so I understand deeply the confusion people have that lead to reaching out and asking me about how to manage tasks.

I will share the process I follow and some of the traps I have fallen into in this article. But the overarching theme is the simplest form you can get away with for being productive the better. If you take one thing away from these words, let it be that whatever you are doing right now is probably overkill. Stop overthinking and reduce the form.

History

I first stumbled across the Bullet Journal method when I spotted the book in a little homeware shop me and my wife liked to visit in Exeter. It called out to me for reasons unknown - but I knew I had to purchase it. Up until that point task management was not really something I had ever given too much consideration to. What I needed to get done, I kept in my head and then just did it when I found some time.

I bought a journal and kept this practice up for several years. I loved it and it worked. Limitations started forming however, when I noticed myself getting busier and busier. I wanted to capture tasks when I didn’t have my notebook to hand, and I wanted to store tasks for a longer period of time - things that I knew I needed to do but didn’t want to do now.

In hindsight this was my first critical mistake and it lead to a few years of pain. I flitted around all the usual suspects for YEARS.. Todoist, Things, Tweek, Notion kanbans.. You name the tool I have probably tried it.

Nothing stuck. I built systems and flows in these tools each time hopeful that this was going to be the thing that let my stay on top of all of the mountain of work that was piling up on me, but none of them lasted more than a few days.

Strangely enough, last year me and my wife were looking around a homeware shop in the Cotswolds, when I spotted a weekly spread on paper. Eerily similar to buying the bullet journal method all those years ago, I knew that I had to have it and give it a try.

Capturing tasks on that weekly spread was the first time in a long time that I was able to successfully keep up and on top of things. I kept it up for a while, but have since moved to travellers style notebook where I have a dedicated calendar notebook as well as a writing pad. I have thought a lot about what changed and why this method proved so well for me.

I don’t believe it is an analog vs digital debate specifically. What I have put it down to is you are forced to be much more intentional when you only have a limited space to capture things.

The error in moving away from the Bullet Journal method was in the fact that a digital tool made it easier to collect tasks whenever they came to my head. Whilst I thought that was great at the time, it is actually damaging.

The friction of writing something down by hand means that you are more likely to make a decision to decide that I won’t do a certain task. You limit the things that come into play and that are in your line of sight, and I think this is CRUCIAL.

My Current Flow

Each morning, I wake up make a coffee and dedicate ten minutes to writing down what needs to be done today.

This is written on paper. It’s usually four or five things that are on top of my mind. I find this is incredibly useful just to get anything out of my brain that is being held on to. The act of getting it down on to a page is both useful and meditative. If its top of your mind and you can recall it in this way, then I feel pretty confident it is important to do.

I keep a weekly list in a Raycast note, with each day of the week. After I have written the tasks down manually, I will open my Raycast note, and see what hasn’t been done from the day before and how it fits with what I wrote down.

I then set a 30 minute focus session (on Raycast) and try and get as much as them done as possible without distraction. This is a super critical step for me actually - usually tasks are much quicker to do that you imagine them to be, and getting a few quick wins in first thing in the morning is a great feeling.

Anything else that is bigger, I will be sure to schedule time in the calendar for that day to getting it done. Tasks should be limited to the maximum of an hour - if it is longer than that it should be broken down.

I use Raycast notes over Obsidian (which would also work) purely for the fact that Raycast notes persists at the side of my screen when I open another window. Being consistently reminded that these things need doing and being able to see them consistently is really helpful. Sure there is a little bit of friction there, and it would probably be a bit tighter/neater in Obsidian, but I am ok with the small level of friction. In this case it is quite helpful.

One of the things I was worried about was not having the ability to write down “someday” tasks that I wanted to do at some point in the future - but I think ridding your mind of the burden of these things is healthy. I have found no downside to not having a longer time list of things that need doing. I simply decide to do it at some point this week, and if it isn’t going to be finished this week then I don’t think about it at all.

Lessons Learned

1. Simple is best - You don’t want to be spending longer than necessary or any brain space on creating a workflow to manage your tasks. The simplest form you can get away with is probably the best.

2. Removing tasks is as important as doing them - Making the decision to not do something is as important as getting things done. It stops you from focusing on the fluff, and keeps your working on the critical.

3. Get shit done - I found myself stressing and worrying about how tasks were being captured and organised more so than doing the actual work. Productivity is as simple as just doing the thing that needs to be done. If you are putting more thought into it, then you are overthinking.

4. Be honest with yourself - I knew for a long time trying these different tools wasn’t working for me, but I wasn’t honest with myself. Probably because I wanted to try and be smart and find a clever way of managing tasks. I would have saved a shit load of time had I just cut my losses and been honest with myself.

5. Make time - Having the time to look over, write tasks down and be intentional is critical. This is a step many people forget, and then wonder why things aren’t happening. At the start of each day, carve out some time to schedule your day.

6. Schedule creativity - Whilst tasks generally are something you can tick off and complete, I have found that scheduling time to “think about X” has been really beneficial. This is a very recent thing for me, but it has already proved to be incredibly valuable. It could be as simple as sitting and thinking about a topic, making some notes or mind mapping.

7. Reflect on each week - Looking back over the week and seeing how you have done is a nice practice. I often found how I have deviated from my intentions at the start of the week, and there are few things I intended to do that haven’t been done. Being aware of this, and taking it into account will only make you stronger and more efficient.

Self-help is a type of book I have veered away from the last couple of years however for one simple reason. It’s literally the same few messages repackaged in a different way a thousand times over.

I have come to refer to that message as “get your ducks in a row”. A message I find I have to give myself and remind myself of often.

My favourite few non-fiction books as an example - Dan Charmas “Work Clean”, Oliver Burkemans “4000 weeks”, Greg McKeowns “Essentialism” and “Effortless”, Cal Newports “Digital Minimalism”, James Clears “Atomic Habits”. As much as I love these books, the message I take away from all of them is about stripping things back and doing the basic and (often) most obvious things well.

Unfortunately myself and doing the basics well have had a turbulent relationship. It turns out I am not a natural duck herder, and my ducks are all over the fucking place.

Not revising at school and then wondering whey my exam results were not as good as they could have been. Not putting the time in to consistently write, and then wondering why essays are not appearing on my Substack. Not tidying the house and then cursing when its a mess. Not exercising and then wondering why I am putting on weight.

See I am happy to fall into the trap of thinking about the more glamorous and exciting things that come later. I will spend days perfecting the exact topics and long term strategy of what I want to write, and how it’s all going to piece together in some master plan that will see me on my way to success, yet I will neglect the one thing that will actually move the needle.. Writing some words.

I will fall down rabbit holes of curating the perfect lifting routine based on research and science that will have the maximum affect but I struggle to go outside and lift the weights. Even something as simple as dropping to the floor and doing fifty push ups each day would get me exponentially further to the end goal than my inaction and overthinking.

At work, I can spend time chasing and researching the new technologies and novel ways to save age old problems rather than getting shit done.

But this is all a facade. It’s the same as all of these self-help books that regurgitate the same message a million times over. It’s not that I or you or we don’t know that these things are helpful - it is obvious - but actually having the discipline to sit down and be real with yourself about doing the unglamorous work is everything.

Maybe this plain straight talking advice doesn’t really sell well. It’s not sexy, it’s not exciting. People look for the magical routine that will solve all their problems and bestow them with unlimited riches and success and they will forever be the envy of others. This is why I believe we have so many different flavours of the same piece of advice.

For me, I am in my best place mentally when I am focusing on making sure all bases are covered and that the simple things are being done to a high standard - probably because I can see that things are getting done and that I am moving in the right direction. It might seem overly simplistic, but I am surprised everytime I catch my ducks getting out of line and I need to resort to falling back to these things, just how powerful they actually are.

This means:

• Eating healthy.

• Exercising regularly.

• Keeping my physical and digital spaces tidy of clutter.

• Setting goals each morning, and getting them done.

• Spending time to write each morning.

• Spending time to read each morning.

If you do those things, these few simple things in and throughout 2026, I guarantee you you will be ahead of the majority of people.

Get the basics sorted. Do them well.

Get your ducks in a row.

The core concepts are rooted around becoming less reliant on technology to live our every day lives. We can't escape it in it's entirety, but being mindful about social media, conscious of screen time and having a simpler approach to using digital tools that so many of us rely on day to day.

It's grown in popularity, especially over the last few years and rightly so. We are becoming more attuned to the negative impact living on a screen brings - and we are questioning if there is a better way of doing things. People are intentionally cutting back on social media, finding simpler ways of building and creating - and I am all for it.

Yet somewhere along the way, digital minimalism has seemingly turned into an aesthetic. We have started focusing on the surface level - apps that look simple, or quitting a particular social media rather than making actual change.

Don't get me wrong. I love the look of minimalist interface, and it's something that I take great influence over in my own work, but there's a risk here. When we focus on the appearance of minimalism or the surface level, we miss the deeper work that Cal wrote about so well.

Deleting apps or simplifying a home screen is great - but only if it’s part of a larger process that affects your behaviour and how you are living your digital life. Without that foundation, aesthetic minimalism becomes little more than digital feng shui. Simple and clean interfaces are easy to fetishise as it gives us the feeling of clarity and being on top of everything without necessarily helping us achieve it. And in doing so, they can distract from the harder, deeper task of rethinking our relationship with technology.

Imagine it in terms of physical clutter. If you are tidying your house and you move every little bit of mess or rubbish into one room and then shut the door, the illusion is that you might be on top of everything, but the bad habits that got you into such a state in the first place still exist - they are just hiding behind the door. Nothing has been sorted, it's just been moved somewhere out of sight.

This is where most people get stuck in that they change the surface, but not the core. They use tools that look simplistic, but spend days building complex systems and workflows that bring very little benefit. They organise their apps nicely on the home screen, then spend hours mindlessly on them. They delete Tiktok, then binge YouTube. They announce their departure from Instagram and then doom scroll Substack. They spend hours making their Notion "aesthetic AF ✨", but then never actually use it for what they have designed it for.

I speak from experience by the way, I have fallen into every trap imaginable, believing that I was making positive change in my life but swapping one vice for another. I cringe when I think back to how overly complex I made things for myself in the guise of "minimalism". So much time wasted.

For me I was always chasing the one way of working that would help me solve all of my problems - but being true to myself, the only issue was I was avoiding doing the things that I was supposed to. I wasn’t getting shit done. All I was doing was plastering over my lack of agency with a facade of things that were making me feel “productive”. That is a bad place to be, readers.

Yet I have learned from my mistakes - and I want to help you embrace digital minimalism and digital clarity also.

Design your digital life around your values, from your goals. Start from the inside, then work outward. Most importantly, be an honest critic for yourself and hold yourself truly accountable for building healthier habits.

Here’s how I think about it:

1. Define your outcome.

What do you actually want from your digital life? More focus? Deeper work? Less comparison? Just to be behind a screen less? Clarity starts with direction.

For me it was about having less comparison, less busy work and more creativity. I let go of the places that made me feel like shit (looking at you, Instagram) and dived deeper into actions like writing in my physical journal, or capturing inspiration via Obsidian and MyMind.

These things are good for me, and again, digital minimalism is not about abstaining from all digital tools, that is not realistic, it is about being mindful and intentional.

2. Audit your tools.

Every app should earn its place. If it doesn’t help you create, connect, or grow - it’s probably noise. Delete the apps that are sitting on your phone unused. Get rid of the social media accounts where you don't actually post but just doom scroll.

This took me a lot of time to do, and was hard, but it is an important process. I said goodbye to social media accounts that had 10,000+ followers that had been the accumulation of years of work as your own mental health is always more valuable than social clout. I deleted all of the productivity apps from both my phone and laptop that I liked, but only used occasionally.

I speak to so many people who are constantly chasing the new tool to solve all of their problems... Pick tools for simplicity but also longevity. Systems take months to develop and grow. Constantly migrating is no good.

(I am going to be writing more about this particularly in future, if folks are interested.)

3. Add friction with intention.

Make distractions harder to access. Log out. Remove shortcuts. Put your phone in another room. Write more things by hand. Avoid needless automation.

One of the things I try and teach is friction is good. Whether its adding enough friction to break you out of the muscle memory habit of opening social media, or manually writing things down so your brain processes them more efficiently, it is helpful.

Adding the screen time widget to the phone home screen is another small hack that adds just enough friction. A pang of guilt when you open your phone and see the wasted time initially is scary, but effective in changing bad habits.

I hope this resonated with some of you. It's certainly been resonating with me more and more over the last year, where I have been working on myself more than I ever have done in the past.

To truly embrace and get the benefits of digital minimalism, look past the aesthetics. Look at the core. If your apps look minimalist, but your digital habits are a complete mess and you are not working/building with the clarity and purpose that you want - then it is never too late to make some impactful change.

Until the next time friends!