Web Applications and Collaborative Workspaces
Browser-based graphical interfaces, self-hosted web applications, local server dashboards, and full-stack interactive graph platforms.
Astucia LLM Wikiastucia.wikiAn open-source, flat-file, Git-enabled self-hosted wiki designed for human-agent collaboration. Stores Markdown pages, diagrams, and logs as plain files without a database, featuring automated Git commit tracking, background AI agent cron jobs for page maintenance, interactive knowledge graph visualization, and a built-in Model Context Protocol (MCP) server.
BYO-LLM-WIKI (Lyra-stellAI)github.comA local-first Flask web application, JSON API, and CLI utility that implements the Karpathy LLM Wiki pattern using deepagents. Features an 8-layer on-disk knowledge graph (entities, topics, synthesis), contextual HNSW retrieval, cross-session memory, bi-directional MCP server capabilities, and human-gated LangGraph skill synthesis.
Cabinetruncabinet.comA free, open-source, file-based AI knowledge workspace that implements Karpathy's compilation loop, featuring git-backed auto-commits, scheduled agent automation cron-jobs, an integrated browser terminal, and embedded HTML application injection.
Eigmaeigma.appA local-first personal wiki platform and Model Context Protocol (MCP) connector for AI assistants (ChatGPT, Claude, Claude Code, Codex). Compiles conversations into chaptered, interlinked Markdown pages paired with spaced-repetition quizzes, featuring a review inbox for proposed changesets, interactive 3D galaxy visualization, and rich rendering (LaTeX, Mermaid).
Knemoraknemora.spaceAn open-source, local-first personal knowledge base and Second Brain engine implementing Karpathy's LLM Wiki pattern on Apple Silicon. Ingests multi-format documents (PDF, DOCX, spreadsheets, markdown) into immutable source stores, compiling them into a connected Markdown wiki with claim-level provenance, native graph views, and on-device inference via llama.cpp.
Kompl (tuirk)github.comA self-hosted compounding LLM Wiki application and personal second brain. Uses a local NLP pipeline (spaCy, TF-IDF) paired with Gemini or DeepSeek to ingest web pages, PDFs, and YouTube transcripts, compiling them into an interlinked wiki with entity graph views, page history, disagreement tracking, and an 8-tool MCP server.
LLM Wiki (ddsyasas)llmwiki.ccAn open-source, local-first web application and CLI implementing Karpathy's LLM Wiki pattern. Refactors multi-format sources into typed markdown pages (Concepts, Entities, Comparisons, Overviews) with an interactive 3D knowledge graph, two-pass linting, and support for both OpenRouter frontier models and local Ollama instances.
LLM Wiki System (ai-biz-app)github.comA local-first web application and FastAPI platform implementing Karpathy's LLM Wiki pattern with a vanilla JS SPA. Synthesizes URLs and multi-format documents into structured concept, entity, and analysis pages in an Obsidian-compatible vault, featuring full-text search, an asynchronous file queue, and a D3.js knowledge graph powered by Graphify with 3-tier confidence scoring.
LLM Wiki Toolsllmwiki.toolsA managed web platform and Model Context Protocol (MCP) service implementing Karpathy's LLM Wiki pattern. Converts uploaded multi-format documents (PDFs with OCR, Office docs, Markdown) into an interlinked wiki layer authored by Claude via remote MCP tools, featuring click-through source citations, compounding query synthesis pages, and automated inconsistency and orphan-node linting.
LLM Wiki v2 (crow-hq)github.comAn open-source, local-first web application, Python library, and CLI implementing the LLM Wiki pattern with CROW (Classifier Routing for Open Wikis) architecture. Replaces vector embeddings and monolithic index prompts by routing multi-format document ingestion (PDF, Office, Markdown) and hierarchical retrieval through a lightweight classifier, reserving the LLM for note synthesis to achieve 89% retrieval precision, 2.1× faster ingest, and 68% token savings.
localmdlocalmd.appAn open-source, local-first web application and agentic workbench operating directly on local folders via the File System Access API. Features an integrated agent that compiles and edits Markdown wiki notes with
[[wikilinks]], force-directed graph visualization, human-in-the-loop diff reviews, block-level PDF/EPUB citation chips, browser-native Git, and MCP server connectivity.
Mementogithub.comA self-hosted, local-first engine that compiles decades of local email data into an organized knowledge base across People, Projects, and Concepts. It runs deterministic graph and parsing algorithms to map connection structures, allows text narrative enrichment via local LLMs, and exposes its underlying FTS data directly to terminal agents over structural search tools.
Mempipemempipe.aiA collaborative web platform and shared workspace implementing Karpathy's LLM Wiki pattern across multi-agent setups. Provides a three-part directory layout (
raw/,wiki/,AGENTS.md) with click-through passage citations on original PDFs, optimistic version locking to prevent conflicting agent overwrites, full-text search, and remote MCP connectivity.
Myco (cmblir)github.comA personal knowledge base tool that compiles unstructured files (PDFs, docs, images, media) into an organized Obsidian-compatible wiki vault. Features a local MCP server integration for Claude Desktop or Claude Code, an offline semantic search index (via BGE-M3 and llama.cpp), and an autonomous researcher agent mode that crawls notes to answer questions with citations.
scrap (gilesknap)github.comAn open-source personal knowledge store and PWA implementing the Karpathy LLM Wiki pattern on SQLite. Keeps human notes and model-generated content strictly separated—labeling Claude's additions (photo OCR, whiteboard write-ups, tags, topic pages) as regenerable derivations—while exposing an MCP server, Claude Code skills (/sc-find), automated vault health checks, and Kubernetes Helm deployment with Litestream.
Silica (kiycoh)silica-agent.vercel.appA local-first transactional agent and web GUI workbench for Karpathy-style LLM Wikis and Google's Open Knowledge Format (OKF v0.2). Implements an FSM write path with post-write verification gates and rollback mechanisms, atomic graph-safe link redirects, zero-trust ingress staging, and published abstention benchmarks on LoCoMo.
Sofie Wikisofie.wikiA chat-driven AI knowledge base application that automatically writes, cross-links, and updates wiki pages from conversations, featuring local history and Model Context Protocol (MCP) integration.
talirezun/the-curatorgithub.comA full-stack, locally-hosted knowledge curation application that automates the Karpathy LLM Wiki pattern via an accessible browser GUI. Features an automated Atomic Decomposition pipeline separating files into interconnected Entity, Concept, and Summary nodes, built-in Obsidian graph visualization setups, and an integrated private Model Context Protocol (MCP) server bridge to expose local vaults directly to external AI clients.
WikiMind (Jayashan Manodya)wikimind-lac.vercel.appA full-stack, AI-powered knowledge management system inspired by the LLM Wiki pattern. Converts unstructured documents into an interlinked, Wikipedia-style knowledge base featuring a 2D physics-based interactive knowledge graph, a stateful LangGraph ingestion pipeline, LangSmith validation, a zero-hallucination grounded QA assistant with click-through citations, and cross-device Turso Cloud database connectivity.
XTrace MemHubmem.xtrace.aiA web application, context control panel, and browser extension that turns chat histories from ChatGPT, Claude, and Gemini into an interactive LLM Wiki mindmap. Extracts memories into an encrypted database and exports structured, interlinked Markdown vaults ready for Obsidian navigation with zero local CLI setup.
yologdev/yopediagithub.comA full-stack Next.js and TypeScript wiki application built entirely by autonomous agents. Implements a dual-surface architecture combining a human-readable Markdown layer (featuring inline citations, validation rules, and decay parameters) with an agent-consumable Model Context Protocol (MCP) server containing 28+ specialized data tools. Runs an ongoing multi-agent loop via GitHub Issues with a strict, automated shell-testing harness that handles continuous ingestion, claim linting, and automatic code reverts.
