{"repo":"arturseo-geo/llm-knowledge-base","free":true,"listed":false,"github":"https://github.com/arturseo-geo/llm-knowledge-base","clone":"git clone https://github.com/arturseo-geo/llm-knowledge-base.git","description":"A schema standard for LLM-compiled personal knowledge bases. AGENTS.md spec, templates, worked example, spaced repetition learning layer.","language":null,"stars":39,"topics":["agents","ai","knowledge-base","llm","markdown","obsidian","pkm","rag","schema","spaced-repetition"],"license":"MIT","category":"ai-agents","readme_excerpt":"LLM Knowledge Base Built by Artur Ferreira @ The GEO Lab · 𝕏 @TheGEO Lab · LinkedIn · Reddit A schema and workflow for LLM-compiled personal knowledge bases. Inspired by Andrej Karpathy's April 2026 post on using LLMs to build and maintain markdown wikis, this repo formalises the pattern into a reusable, versioned schema — including a structured learning layer that the original workflow leaves out. The core idea: you collect raw source material. The LLM compiles it into a structured wiki. You never edit the wiki manually. You just learn from it. --- What this repo contains File Purpose --- --- AGENTS.md The schema spec — every LLM agent operating on your wiki reads this first templates/ Starter templates for each file type in the schema examples/ A worked example wiki (AI alignment topic, fully populated) docs/why-not-rag.md When to use a compiled wiki vs a vector RAG stack docs/learning-layer.md How the spaced repetition and gap tracking system works docs/contamination-mitigation.md Preventing quality degradation in agent-generated wikis docs/two-vault-setup.md Obsidian two-vault setup: agent vault vs personal vault docs/finetune-path.md Turning your wiki into a fine-tuning corpus --- The workflow in one diagram --- Quickstart 1. Clone and initialise your vault Or just copy AGENTS.md into the root of an existing Obsidian vault. 2. Drop source material into raw/ Anything works: PDFs, markdown files clipped with Obsidian Web Clipper, .py scripts, .csv datasets, images. The LL","default_branch":null,"files":null,"tree":[],"storefront":"/r/arturseo-geo","claimed":false,"request_supported":{"post":"https://gitbuyer.com/r/arturseo-geo/llm-knowledge-base/request-supported","requests":0},"note":"indexed from public GitHub; nothing is for sale on this page. Clone it from GitHub. Paid listings live at /search."}