Build a personal research wiki from papers and articles
LLM Wiki (Cursor + Obsidian). Task: LLM Wiki (Cursor + Obsidian) A pattern for building personal knowledge bases using Cursor Agent and Obsidian. This is an idea file designed to be copy-pasted into a Cursor chat (or linked from AGENTS.md). Its goal is to communicate the high-level idea and the Cursor-specific setup, but your agent will build out the specifics in collaboration with you. The workflow Cursor Agent on one side, Obsidian on the other. Cursor edits markdown based on your conversation; you browse the results in real time — following links, checking the graph view, reading the updated pages. Obsidian is the IDE; Cursor is the programmer; the wiki is the codebase. Open the vault folder as your Cursor workspace and point Obsidian at the same directory. Both tools read the same files; Cursor writes, Obsidian renders. The core idea Most people's experience with LLMs and documents looks like RAG: you upload a collection of files, the LLM retrieves relevant chunks at query time, and generates an answer. This works, but the LLM is rediscovering knowledge from scratch on every question. There's no accumulation. Ask a subtle question that requires synthesizing five documents, and the LLM has to find and piece together the relevant fragments every time. Nothing is built up. NotebookLM, ChatGPT file uploads, and most RAG systems work this way. The idea here is different. Instead of just retrieving from raw documents at query time, the Cursor Agent 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 agent 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. This is the key difference: the wiki is a persistent, compounding artifact. The cross-references are already there. The contradictions have already been flagged. The synthesis already reflects everything you've read. The wiki keeps getting richer with every source you add and every question you ask. You never (or rarely) write the wiki yourself — the Cursor Agent writes and maintains all of it. You're in charge of sourcing, exploration, and asking the right questions. The agent does all the grunt work — the summarizing, cross-referencing, filing, and bookkeeping that makes a knowledge base actually useful over time. This can apply to a lot of different contexts. A few examples: Personal: tracking your own goals, health, psychology, self-improvement — filing journal entries, articles, podcast notes, and building up a structured picture of yourself over time. Research: going deep on a topic over weeks or months — reading papers, articles, reports, and incrementally building a comprehensive wiki with an evolving thesis. Reading a book: filing each chapter as you go, building out pages for characters, themes, plot threads, and how they connect. By the end you have a rich companion wiki. Think of fan wikis like Tolkien Gateway — thousands of interlinked pages covering characters, places, events, languages, built by a community of volunteers over years. You could build something like that personally as you read, with the agent doing all the cross-referencing and maintenance. Business/team: an internal wiki maintained by Cursor agents, fed by Slack threads, meeting transcripts, project documents, customer calls. Possibly with humans in the loop reviewing updates. The wiki stays current because the agent does the maintenance that no one on the team wants to do. Competitive analysis, due diligence, trip planning, course notes, hobby deep-dives — anything where you're accumulating knowledge over time and want it organized rather than scattered. Architecture There are three layers: Raw sources — your curated collection of source documents in a directory like raw/. Articles, papers, images, data files. These are immutable — the agent reads from them but never modifies them. This is your source of truth. Clip articles with Obsidian Web Clipper or drop files in manually.; tools: Cursor. Human role: The original post does not say how a person stayed involved.; limitations: Prerequisites include installing Cursor and Obsidian and configuring shared file access. The workflow relies on the agent's ability to maintain consistency across dozens of files, which may degrade as the wiki grows without proper indexing tools. Source reliability is secondary, as this is a conceptual pattern rather than a documented case study. Human approval is required for all wiki updates..
Tools
Published on Agent Work Atlas:
Overview
LLM Wiki (Cursor + Obsidian)
Workflow
LLM Wiki (Cursor + Obsidian) A pattern for building personal knowledge bases using Cursor Agent and Obsidian. This is an idea file designed to be copy-pasted into a Cursor chat (or linked from AGENTS.md). Its goal is to communicate the high-level idea and the Cursor-specific setup, but your agent will build out the specifics in collaboration with you. The workflow Cursor Agent on one side, Obsidian on the other. Cursor edits markdown based on your conversation; you browse the results in real time — following links, checking the graph view, reading the updated pages. Obsidian is the IDE; Cursor is the programmer; the wiki is the codebase. Open the vault folder as your Cursor workspace and point Obsidian at the same directory. Both tools read the same files; Cursor writes, Obsidian renders. The core idea Most people's experience with LLMs and documents looks like RAG: you upload a collection of files, the LLM retrieves relevant chunks at query time, and generates an answer. This works, but the LLM is rediscovering knowledge from scratch on every question. There's no accumulation. Ask a subtle question that requires synthesizing five documents, and the LLM has to find and piece together the relevant fragments every time. Nothing is built up. NotebookLM, ChatGPT file uploads, and most RAG systems work this way. The idea here is different. Instead of just retrieving from raw documents at query time, the Cursor Agent 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 agent 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. This is the key difference: the wiki is a persistent, compounding artifact. The cross-references are already there. The contradictions have already been flagged. The synthesis already reflects everything you've read. The wiki keeps getting richer with every source you add and every question you ask. You never (or rarely) write the wiki yourself — the Cursor Agent writes and maintains all of it. You're in charge of sourcing, exploration, and asking the right questions. The agent does all the grunt work — the summarizing, cross-referencing, filing, and bookkeeping that makes a knowledge base actually useful over time. This can apply to a lot of different contexts. A few examples: Personal: tracking your own goals, health, psychology, self-improvement — filing journal entries, articles, podcast notes, and building up a structured picture of yourself over time. Research: going deep on a topic over weeks or months — reading papers, articles, reports, and incrementally building a comprehensive wiki with an evolving thesis. Reading a book: filing each chapter as you go, building out pages for characters, themes, plot threads, and how they connect. By the end you have a rich companion wiki. Think of fan wikis like Tolkien Gateway — thousands of interlinked pages covering characters, places, events, languages, built by a community of volunteers over years. You could build something like that personally as you read, with the agent doing all the cross-referencing and maintenance. Business/team: an internal wiki maintained by Cursor agents, fed by Slack threads, meeting transcripts, project documents, customer calls. Possibly with humans in the loop reviewing updates. The wiki stays current because the agent does the maintenance that no one on the team wants to do. Competitive analysis, due diligence, trip planning, course notes, hobby deep-dives — anything where you're accumulating knowledge over time and want it organized rather than scattered. Architecture There are three layers: Raw sources — your curated collection of source documents in a directory like raw/. Articles, papers, images, data files. These are immutable — the agent reads from them but never modifies them. This is your source of truth. Clip articles with Obsidian Web Clipper or drop files in manually.
Reported result
The source describes the intended workflow but does not report measured outcomes, success rates, or specific business impacts. It claims the method reduces maintenance burden compared to traditional RAG, but no data validates this efficiency gain. A measurable acceptance check would be verifying that new sources automatically update existing entity pages without manual intervention.
Human role
The original post does not say how a person stayed involved.
Limitations
Prerequisites include installing Cursor and Obsidian and configuring shared file access. The workflow relies on the agent's ability to maintain consistency across dozens of files, which may degrade as the wiki grows without proper indexing tools. Source reliability is secondary, as this is a conceptual pattern rather than a documented case study. Human approval is required for all wiki updates.
Try it
I want to set up a personal research wiki using Cursor and Obsidian. My domain is [replaceable input: e.g., AI Ethics]. I have placed three sample articles in the 'raw/' directory. Please help me initialize the structure: 1. Create 'AGENTS.md' defining the wiki schema and my role as curator. 2. Set up 'index.md' and 'log.md'. 3. Ingest the first article from 'raw/' by creating a summary page in 'wiki/', updating 'index.md', and adding cross-references to existing pages if any. 4. Show me the diff of changes before committing. Do not modify files in 'raw/'. If the article contains images, note them but do not process them yet. Confirm you understand the immutable source rule before proceeding.
Source reports are not independently verified. Publication date is the catalog publication date, not the original article date.