The article introduces a personal knowledge management system based on Claude Code and Obsidian, emphasizing AI's continuous maintenance of a structured knowledge base (rather than a one-time RAG query), achieving knowledge accumulation and compounding through three types of operations: ingestion, querying, and verification, enhancing cognitive efficiency and output quality, and providing a 5-minute setup tutorial and advanced usage tips.
Article author, source: Rhythm BlockBeats
Editor's note: This article introduces a personal knowledge system built on Claude Code and Obsidian, where the core is no longer the traditional RAG model of 'temporary retrieval for each query,' but instead attempts to let AI continuously build and maintain an evolving knowledge base (Wiki).
Structurally, the system can be divided into three layers:
· The first layer is the raw data layer, including notes, articles, transcripts, and other unmodifiable input sources;
· The second layer is a structured knowledge base maintained by AI, which completes cross-referencing and relationship building through continuous updates;
· The third layer is the Schema rule layer, used to standardize the organization of knowledge and the logic of system operation.
Around this structure, the system operates through three core operations: Ingest, which continuously incorporates external information into the system; Query, which enables instant access to knowledge; and Lint, which checks structural consistency and fixes potential issues.
In this mechanism, knowledge no longer remains as one-time conversation results but is gradually precipitated into reusable long-term assets through the cycle of "writing—organizing—reusing." The author proposes that this model allows knowledge to have a compounding effect: on one hand, it reduces individual cognitive burdens, and on the other hand, it enhances the accuracy and contextual consistency of model outputs.
However, the effective operation of this system is also based on one premise—continuous input and maintenance. Without stable data injection and structural updates, this "Second Brain" will struggle to form a genuine accumulation effect, and its advantages will diminish accordingly.
The original text is as follows:
Claude Code + Obsidian is the most powerful AI combination I have ever used.
I've almost built an "AI Second Brain," incorporating all my thoughts, readings, writing, online research, and more into it. This includes my business plans, all the YouTube videos I've published, articles I've written, and everything that is important to me.
Claude Code + Obsidian has quickly gained popularity across platforms, and this is not a coincidence.
For me personally, this AI system has greatly reduced my cognitive burden, enabling me to focus on what truly matters—whether it’s business or personal life.
This system may seem a bit complicated, but in reality, it only takes 5 minutes to set up. More importantly, it comes with a memory mechanism that will continuously optimize itself with use.
Next, I will guide you step by step to replicate this "AI Second Brain" system; it can truly enhance your efficiency.
I suggest you read to the end of the article—I will attach a complete Claude Code + Obsidian operation quick reference table and all the resources mentioned in the text (all free).
Before we begin
This system is not solely my original creation; its inspiration comes from a viral tweet by Andrej Karpathy a few days ago about "LLM Knowledge Base."
The reason this tweet went viral is that it provided ideas for solving a key pain point in the current AI development.
The problem is: every time you start a new conversation or switch to a new AI tool, you have to repeatedly input prompts and provide context, which is almost like starting from scratch.
By combining this system's prompts with Obsidian and Claude Code, this issue can be thoroughly resolved, while significantly enhancing the quality of AI output.
How does this system operate?
The entire system consists of four core modules:
1. Your data: including articles, notes, transcripts, inspiration ideas, etc.
2. Organization Method: Automatically organize through Claude Code in Obsidian
3. Instant Access: You can ask this "database" questions at any time to get answers.
4. Evolving Memory: The system will become smarter with continued use.
What is the real power of this system?
As humans, our cognitive bandwidth is limited. We forget, and sometimes it is difficult to connect different ideas; the information that can be tracked and processed at the same time ultimately has its limits.
By using this system composed of four modules, you are actually relieving your cognitive burden, delegating the work of "connecting, organizing, and understanding information" to Obsidian and Claude Code.
Your ideas start to be systematically connected; one note can automatically relate to another note, and you can always re-extract, combine, and invoke this content through Claude.
In this structure, your knowledge is no longer fragmented but rather a network that can be constantly invoked and restructured—almost without limits.
How to build your AI brain in 5 minutes
1. Download Obsidian
Official website: https://obsidian.md/
2. Create your Vault (Knowledge Base)
After the download is complete, Obsidian will prompt you to create a "Vault."
You can think of it as a folder on your computer, where we will store all content and allow Claude Code to access and manage this data.
The name of this Vault can be set freely—for example, I call it "Obsidian Vault."
This Vault is where Obsidian stores all your data and notes, and all content will be saved as MD (Markdown) files.
3. Set up Claude Code
Next, you need to configure a way to access Claude Code. For me (and most likely for most people), the simplest method is to use the desktop client directly.
In the main chat interface, click "Select Folder," then find and select the Obsidian Vault you just created.
4. Set up system prompts (System Prompt)
Once you select the folder, the next step is to paste Andrej Karpathy's system prompts into the main chat box.
You can copy this prompt here: https://gist.github.com/karpathy/442a6bf555914893e9891c11519de94f
Your input should look like this:
Tip: If you don't want to, you can completely avoid manually opening Obsidian. Just give the MD folder (which is your Vault) and related data to Claude Code, and it can directly read, write, and modify these files—while this content will automatically sync to your Obsidian "Second Brain."
5. Build your database
After you finish entering the above system prompts, Claude Code will start asking you for some data sources to initialize and gradually fill your "Second Brain."
You can think of Obsidian as a "blank notebook"—initially, you need to actively input content for the database to gradually build up. Content that can be imported includes notes, CSV files, Markdown/text files, etc.
Some practical suggestions:
· Export data from your existing note-taking tools
· If you use Notion, you can export it as a CSV file
· Let Claude (or other large models) organize a summary of your information to initialize your "Second Brain"
· Import all your existing articles, collections, inspiration ideas, etc., at once—this is the best time to establish initial data, and you can always supplement it later.
It is important to note that a large database like mine is not built overnight; it is formed by continuously inputting and gradually accumulating over time.
That's it, your "AI Second Brain" is now built and can start operating. Next, I will share some advanced techniques to help you use it more efficiently.
Advanced Techniques (Pro Tips)
1. Obsidian Chrome extension
If you want to add data to the system more easily, just install the Chrome extension for Obsidian. It allows you to click "Add to Obsidian" while browsing the web, saving content directly to your knowledge base with one click. This will make the process of building your "Second Brain" very seamless.
I often use this feature to collect articles, web data, research materials, and so on.
It is important to note that data added through extensions is initially just an "isolated data source."
Next, you can tell Claude Code: "I just added [x] in Obsidian, please help me integrate this content into my Wiki."
Claude Code will automatically associate this new data with existing content, generating links that truly integrate into your "Second Brain." This is also the reason why this tool combination is powerful.
2. Create separate folders (Vaults)
Andrej Karpathy recommends using two separate folders (Vaults):
· A tool for work / business content
· A tool for personal life / goal management
My own experience is that this structure is the clearest and most effective.
3. Practicality
I found that the most valuable use of this system is actually very simple: to make your LLM prompts more precise.
When the model can access your complete personal information, business plans, writing background, and other contexts, it can generate higher-quality prompts that are more customized and closer to real situations (even "super prompts").
Of course, the use of this system goes far beyond these, but if you want to start with the most practical scenario, I strongly recommend beginning with "Improving Prompt Quality."
4. Orphans (Isolated Nodes)
In Obsidian, "Orphans" refers to those data points that have not established connections with other notes.
This feature is very useful because it can help you:
· Find ideas that have not been integrated yet
· Discovering the "weak areas" in the database
· Determine which content is worth expanding or deepening
In other words, it is not only an organizing tool but also a mechanism that helps you discover cognitive blind spots.
You can click the "three dots" in the upper right corner to find and turn on the Orphans switch to see which content has not yet established connections.
Potential drawbacks of this system
We have already discussed many advantages, use cases, and optimization methods, so what are its shortcomings? In what situations might you not be well-suited to use this system?
1. People who are not accustomed to visualization
One core advantage of this system is that it can visualize data. If you do not rely on or are not accustomed to this way, it may be of limited help to you.
2. Requires some maintenance cost
If you are unwilling to continuously maintain a database, this system may not be suitable for you. Although the maintenance cost is not high, if you do not continuously input data into the "Second Brain," it will be difficult to realize its value.
3. Storage usage
All content will be stored locally in the form of Markdown files, which will occupy a certain amount of device space. This also needs to be considered in advance.