Build AI-powered knowledge bases with retrieval-augmented generation (RAG) — ingest documents, search semantically, and answer questions.
Knowledge Base & RAG tools are AI-powered software designed to help developers and teams tackle knowledge base & rag-related tasks more efficiently. These tools are typically published as open-source projects on GitHub and can be integrated into existing workflows via MCP (Model Context Protocol), Claude Skills, or standalone agent frameworks. On Agent Skills Hub, we index 230 quality-scored knowledge base & rag tools across languages including Python, C, TypeScript.
In 2026, the AI agent ecosystem is maturing rapidly. Knowledge Base & RAG tools can significantly boost development efficiency by automating repetitive tasks, reducing human error, and providing intelligent suggestions. The top 3 tools — ragflow, codegraph, LightRAG — have earned an average of 2,178 GitHub stars, reflecting strong community validation. 191 of the listed tools come with clear open-source licenses, ensuring freedom to use and modify.
When choosing a knowledge base & rag tool, consider these factors: 1) Community activity — GitHub stars and recent commit frequency indicate reliability; 2) Integration method — check if it supports MCP, Claude, or your preferred agent framework; 3) Language compatibility — the most common language in this list is Python; 4) Quality score — Agent Skills Hub's composite score evaluates code quality, documentation completeness, and maintenance activity. Our recommendation: start with ragflow — it ranks highest in both star count and quality score.
RAGFlow is a leading open-source Retrieval-Augmented Generation (RAG) engine that fuses cutting-edge RAG with Agent capabilities to create a superior context layer for LLMs
Pre-indexed code knowledge graph, auto syncs on code changes, for Claude Code, Codex, Gemini, Cursor, OpenCode, AntiGravity, Kiro, CoPilot, and Hermes Agent — fewer tokens, fewer tool calls, 100% local
[EMNLP2025] LightRAG: Simple and Fast Retrieval-Augmented Generation
Self-evolving Context Database for AI Agents. Unify Agent Memory, Knowledge RAG and Skills.
Turn any technical book PDF into a Claude Code skill — ready to study, reference, and use while you work.
Cognee is the open-source AI memory platform for agents. Give your AI agents persistent long-term memory with small models for free
🔥 MaxKB is an open-source platform for building enterprise-grade agents. 强大易用的开源企业级智能体平台。
Open Source DeepWiki: AI-Powered Wiki Generator for GitHub/Gitlab/Bitbucket Repositories. Join the discord: https://discord.gg/gMwThUMeme
OpenWiki is a CLI that writes and maintains agent documentation for your codebase.
Self-organizing AI second brain for Obsidian + Claude Code. Drop any source and Claude reads, links, and files it into one connected knowledge graph of plain Markdown you own. AI note-taking, personal knowledge management (PKM), and an open-source Notion alternative. Based on Karpathy's LLM Wiki pattern.
可私有部署的多租户知识智能体平台：统一 RAG、知识图谱、多智能体、MCP/Skills、沙盒与权限管理。Yuxi = Cloud Agents + Knowledge RAG, Self-hosted knowledge agent platform for RAG, knowledge graphs and multi-agent workflows.
Open-source context retrieval layer for AI agents
The ultimate RAG for your monorepo. Query, understand, and edit multi-language codebases with the power of AI and knowledge graphs
Persistent memory for Claude Code and 6 other CLI agents, stored as plain markdown in your Obsidian vault. Stop re-explaining your projects, decisions and people every session. 45 commands: hybrid semantic search, self-rewriting notes, key-less web research, and scheduled agents that maintain the vault while you sleep.
A bio-inspired cognitive memory engine — a new paradigm for Graph RAG.
The open-source context layer for AI agents. PipesHub turns your company's knowledge (Slack, Drive, Jira, GitHub, Microsoft 365 and 40+ connectors) into a permission-aware workspace that agents can search, grep, navigate and cite. MCP, SDKs and built-in agents. Self-hosted.
Knowhere extracts, parses, and outputs structured chunks ready for AI Agents and RAG.
A personal knowledge base that builds and maintains itself. Drop in sources — Claude (or Codex/Gemini) reads them, extracts knowledge, and maintains a persistent interlinked wiki. Works with Claude Code, Codex, OpenCode, Gemini CLI. No API key needed.
ReMe: Memory Management Kit for Agents - Remember Me, Refine Me.
Framework for AI agents to build and maintain a digital brain through Obsidian wiki | Memory System for Agents
```bash
git clone https://github.com/Ar9av/obsidian-wiki.git
cd obsidian-wiki
bash setup.sh
```Claude Code plugin that generates individualized knowledge systems from conversation. You describe how you think and work, have a conversation and get a complete second brain as markdown files you own.
```
/plugin marketplace add agenticnotetaking/arscontexta
```A persistent, unified memory layer for all your AI agents (e.g. Claude Code, Codex, DSH), backed by Markdown and Milvus.
Agent Skills-compatible LLM wiki for Claude Code, Cursor, and Codex. Build a Karpathy-style knowledge base from raw sources, citations, and linting.
```bash
npx add-skill Astro-Han/karpathy-llm-wiki
```A bio-inspired cognitive memory engine — a new paradigm for Graph RAG.
Web Crawling and RAG Capabilities for AI Agents and AI Coding Assistants
The knowledge compiler. Raw sources in, interlinked wiki out. Inspired by Karpathy's LLM Wiki pattern.
Open-source, AI-native knowledge base & Evernote alternative with native MCP. Zero-cost on Cloudflare or Docker.
Official Microsoft Learn MCP Server and CLI tool – powering LLMs and AI agents with real-time, trusted Microsoft docs & code samples.
Team memory for engineers and their AI agents. Lives in your repo. Shared through Git.
Open Source Implementation of Karpathy's LLM Wiki. Upload documents, connect your Claude account via MCP, and have it write your wiki !
Synthadoc: An open-source LLM knowledge compilation engine that turns raw documents into structured, local-first wikis. A transparent, human-readable alternative to traditional RAG, which can be self-managed and self-improved without the use of any tools.
Your entire engineering context, deeply understood
LLM-compiled knowledge bases for any AI agent. Parallel multi-agent research, thesis-driven investigation, source ingestion, wiki compilation, querying, and artifact generation.
🧠 RepoBrain (formerly Antigravity) — Give your repo a brain. ChatGPT for your codebase: works in Claude Code, Cursor, Codex, Windsurf & more.
ApeRAG: Production-ready GraphRAG with multi-modal indexing, AI agents, MCP support, and scalable K8s deployment
Give Claude Code, Cursor, Codex CLI a ChatGPT for your codebase. Multi-agent knowledge engine, grounded Q&A with file paths and line numbers. Works in any AI IDE.
A personal context store for AI agents and assistants—reuse your existing coding agent CLI (Codex/Claude/OpenCode) with built‑in Skills/tools and a desktop GUI to capture, search, and reuse project knowledge across agents and repos.
Arkon: Enterprise AI Knowledge Hub & MCP Server. Self-hosted knowledge base for teams to manage RAG contexts, access policies, and AI skills. Connect Claude and other LLMs via Model Context Protocol (MCP) for automated, secure organizational knowledge integration.
把中文全渠道内容（抖音 / B站 / 小红书 / 公众号 / X / 播客）采集进个人知识库的 14 个 AI Skill：图文存图、视频转文字稿、字幕优先免 GPU，附带知识库 MCP server。 ｜ Ingest Chinese content into your personal knowledge base — image/video routing, subtitle-first transcription, and a KB MCP server.
A local-first AI knowledge base & NotebookLM alternative built with Electron. More convenient, more lightweight, and understands you better!
All-in-One Multimodal Parsing Engine + Ontology-Powered, LLM Wiki-Driven AI-Ready Knowledge Engine
A modular RAG (Retrieval-Augmented Generation) system with MCP Server architecture. Using Skill to make AI follow each step of the spec and complete the code 100% by AI.
On-premises conversational RAG with configurable containers
AI-powered StartUp Accelerator Engine built with Next.js, LangChain, PostgreSQL + pgvector. Upload, organize, and chat with documents. Includes predictive missing-document detection, role-based workflows, and page-level insight extraction.
100% Rust implementation of code graphRAG with blazing fast AST+FastML parsing, surrealDB backend and advanced agentic code analysis tools through MCP for efficient code agent context management
An AI second brain that maintains itself. Full guide, starter vault, agent skills and scripts for a self-organizing knowledge base in Claude Code and Obsidian.
Long-term memory for AI assistants. Graph + vector store that recalls decisions, relationships, and context across sessions.
One memory layer, every AI tool. Store anything once — recall it in Claude, ChatGPT, Cursor, or any MCP client. Self-hosted on Cloudflare's free tier.
A codebase wiki for AI coding agents. Captures what the code can't say: decisions, flows, invariants, gotchas.
Give your AI agents persistent, collective memory — with deduplicating absorb, supersession lineage, semantic search, and a graph UI. Speaks MCP.
MindOS is a Human-AI Collaborative Mind System, where human thinks and agents act. Globally sync your mind for all agents: transparent, controllable, and evolving symbiotically.
RAGLight is a modular framework for Retrieval-Augmented Generation (RAG). It makes it easy to plug in different LLMs, embeddings, and vector stores, and now includes seamless MCP integration to connect external tools and data sources.
The local-first LLM Wiki: open-source knowledge graph builder, RAG knowledge base, and agent memory store. Built on Andrej Karpathy's pattern. An Obsidian alternative for personal knowledge management, AI second brain, and durable Claude Code / Codex / OpenClaw memory.
Asisten crypto berbahasa Indonesia: RAG pengetahuan 267 topik + data pasar realtime (6 bursa, WebSocket, derivatif, on-chain, TVL, DeFi) + tool-calling agent + LLM synthesis
Karpathy-style LLM knowledge base Agent Skill for OpenClaw/Codex. Experimental — will iterate over time.
⚡ DocsAgent — give your AI agents instant, private access to your personal knowledge base (Zotero, Obsidian, Apple Notes supported now, local docs on the road). Native C++ search core: BM25 + passage ranking in ~15 ms. MCP server for Claude, Cursor, Cline & any MCP client.
Agentic RAG for local and self-hosted document search: hybrid retrieval, reranking and multimodal RAG on embedded LanceDB, with Docling parsing and an MCP server
Agent-native knowledge engine with MCP tools for document indexing, wiki organization, fast retrieval and deep reading across PDF/DOCX/PPTX/Markdown
Self-maintaining, Obsidian-compatible knowledge base for pi — turn raw sources into an interlinked wiki that compounds. Native Open Knowledge Format (OKF) v0.2.
📚 A zero-dependency, git-backed micro-lesson library for AI Agents to asynchronously share and search verified debugging experience. | https://misakanet.org
Karpathy's LLM Wiki implementation plugin for Obsidian - turns notes and PDFs into a linked, LLM-powered knowledge base with entity pages, concept pages, graph-powered Q&A, and local-first privacy.
Python toolkit for building graph-enhanced GenAI applications
Your First LLM-Wiki Conversation Knowledge Base
A local-first personal knowledge layer for AI agents. Build evolving knowledge spaces with LLM Wiki, knowledge graphs, and agent workflows.
Talk with your notes in Claude. RAG over your Apple Notes using Model Context Protocol.
```bash
git clone https://github.com/RafalWilinski/mcp-apple-notes
cd mcp-apple-notes
```Local-first RAG server for developers. Semantic + keyword search for code and technical docs. Works with MCP or CLI. Fully private, zero setup.
Ask questions across your Markdown notes using a fully local Graph RAG engine. Built for Obsidian vaults, works with any folder of Markdown files. Extracts entity-relation triples from wikilinks & YAML frontmatter, retrieves answers via hybrid search (vector + BM25 + temporal). Multilingual. No cloud. Runs on Ollama.
An always-on second brain you talk to. Voice notes in Telegram → typed, linked knowledge in your Obsidian vault. Runs 24/7 on the Claude subscription you already have.
LLM-powered knowledge base from your Claude Code, Codex CLI, Copilot, Cursor & Gemini sessions. Karpathy's LLM Wiki pattern — implemented and shipped.
MCP Documentation Server - Bridge the AI Knowledge Gap. ✨ Features: Document management • Gemini integration • AI-powered semantic search • File uploads • Smart chunking • Multilingual support • Zero-setup 🎯 Perfect for: New frameworks • API docs • Internal guides
MCP server to interact with LogSeq via its Local HTTP API - enabling AI assistants like Claude to seamlessly read, write, and manage your LogSeq graph.
Turn your Claude Code session history into a searchable local knowledge base — 13 MCP tools, pure stdlib, nothing leaves your machine
Local-first persistent agentic memory powered by Recursive Memory Harness (RMH). Open source must win.
Claude Code plugin that compiles markdown knowledge files into a topic-based wiki. Implements Karpathy's LLM Knowledge Base pattern.
A-RAG: Agentic Retrieval-Augmented Generation via Hierarchical Retrieval Interfaces. State-of-the-art RAG framework with keyword, semantic, and chunk read tools for multi-hop QA.
台灣法律 MCP 伺服器 + CLI（免費、免註冊、免 API key）：2,250 萬筆裁判書、行政函釋、憲法法庭裁判，附引用查核。Free Taiwan legal MCP server for Claude/ChatGPT/Codex — bring your own LLM, retrieval-only.
Persistent project knowledge graph for coding agents. MCP server with semantic search, in-process embeddings, and web explorer.
Self-improving, AI-native markdown vault you hand to an AI agent. GitHub-style file tree + Notion editing, exposed to Claude/Cursor via a built-in MCP server (24 tools): semantic & hybrid search, RAG, cited answers. Learns your voice from your edits; human-approved review queue. Local-first .md.
Revornix is an open-source, local-first AI information/markdown workspace. It helps you collect fragmented inputs, turn them into structured knowledge, generate reports with images and podcast audio, and deliver the output through automated notifications.
Pharos — local-first agentic RAG for your team's document library: multi-format ingest, hybrid retrieval, enterprise ACL, dual HTTP + MCP exits.
🦀 Prevents outdated Rust code suggestions from AI assistants. This MCP server fetches current crate docs, uses embeddings/LLMs, and provides accurate context via a tool call.
Turn Google Search, Papers, and Codebases into an Automatic Local Knowledge Graph for AI Agents.
Local RAG MCP server for Claude Code — hybrid search (semantic + BM25), cross-encoder reranking, 13 MCP tools, 20 format parsers. Zero external servers, zero API keys.
Build your own LLM-native WIKI (knowledge library). Search, extract, summarize, Q&A with contextual RAG, layered knowledge graph, and reinforced memory. Importantly use selected context to automatically generate skills, empowered by Claude subagents + CodeAct pipeline and gated by human review. **Try Live Demo**: https://byo-wiki-demo.vercel.app
A library-science-inspired personal knowledge management system with LLM agents
Pharos — local-first agentic RAG for your team's document library: multi-format ingest, hybrid retrieval, enterprise ACL, dual HTTP + MCP exits.
Build your own LLM-native WIKI (knowledge library). Search, extract, summarize, Q&A with contextual RAG, layered knowledge graph, and reinforced memory. Most importantly you can use selected context to automatically generate skills, empowered by Claude subagents + CodeAct pipeline and gated by human review.
Amazon Bedrock Foundation models with Amazon Opensearch Serverless as a Vector DB
301 open-source coding agent skills across 22 domains — methodology, judgment & gotchas packaged as Claude Agent Skills for the Asgard AI Platform.
an AI interaction tool with RAG hybrid search, conversation context, web content processing and structured data analysis with LLM / GPT
Hệ thống knowledge base cá nhân hoàn toàn tự động, vận hành bởi LLM. Dựa trên pattern LLM Wiki của Andrej Karpathy.
LLM Wiki - 用 LLM 构建持续积累的个人知识库，含 Claude Code Skill 和实战经验
Local-first code intelligence for AI assistants. Turns your codebase into a knowledge graph your AI can query, navigate, and remember. 25 MCP tools.
📚 Student LLM Wiki — AI-compiled knowledge base for university students. Drop course slides, get a persistent interlinked wiki. Feynman review, exam prep, confidence decay, cross-course connections. Based on Karpathy's LLM Wiki pattern. Works with Claude Cowork / Claude Code / Claudian + Obsidian.学生大语言模型维基百科——专为大学生打造的AI编译知识库。
OnCo: total information dominance on cancer. A public, cited knowledge graph of oncology with a website, JSON API, MCP server and CLI: one page per cancer, treatment, target, trial, institution, person and idea.
Python, LlamaIndex, LangChain, 15 Property Graph, 4 RDF , 10 Vector, OpenSearch, Elasticsearch, Alfresco, Nuxeo DBs. 14 data sources (10 auto-sync), KG auto-building, Ontologies, LLMs, Docling, LlamaParse, LiteParse, GraphRAG, RAG, Hybrid Search, AI Chat. TypeScript React, Vue, Angular frontends, REST, MCP Server. Options: Langflow, CocoIndex
The World's Most Comprehensive, Authoritative, and Structured Open Source Data Source Knowledge Base
LLM-maintained personal wiki skills for Claude Code — implements Karpathy's LLM Wiki pattern
```bash
/plugin marketplace add kfchou/wiki-skills
/plugin install wiki-skills@kfchou/wiki-skills
```A comprehensive knowledge base for Huawei Ascend NPU development, structured as distributed Agent Skills. https://ascend-ai-coding.github.io/awesome-ascend-skills/
Self-improving, AI-native markdown vault you hand to an AI agent. GitHub-style file tree + Notion editing, exposed to Claude/Cursor via a built-in MCP server (24 tools): semantic & hybrid search, RAG, cited answers. Learns your voice from your edits; human-approved review queue. Local-first .md.
Transform your codebase into an intelligent knowledge base for AI-powered development with Cursor IDE, Google AntiGravity, and MCP-enabled assistants
Harness engineering applied to knowledge production: a self-evolving multi-agent newsroom that turns your documents into a cross-linked markdown wiki. A "reground" loop pulls published pages back in before they go stale — writer ≠ reviewer, local-first, a structured alternative to RAG.
A Model Context Protocol (MCP) server that implements the Zettelkasten knowledge management methodology, allowing you to create, link, explore and synthesize atomic notes through Claude and other MCP-compatible clients.
AI-native codebase intelligence skills — generate persistent .nexus-map/ knowledge bases and instantly query file structure, dependency graphs, and change impact. Built for Copilot, Cursor, and any tool-calling LLM. 面向 AI 编程助手的代码库感知技能——生成持久化知识图谱，精准查询文件结构、依赖关系与改动影响半径。
AKB — Agent Knowledgebase. Organizational memory for AI agents: vault-scoped docs / tables / files unified by URI graph, served over MCP.
OKF (Open Knowledge Format): Durable, structured memory for AI agents. Author, validate, consume, and maintain portable knowledge bundles through an ecosystem of Skills, MCP, an interactive graph, TUI, CLI, Docker, and a Claude Code plugin. 100% local.
Model Context Protocol server to allow for reading and writing from Pinecone. Rudimentary RAG
Local-first scientific literature RAG with quote-anchored evidence annotations, human paper screening, frozen provenance, and model-free research worksheets.
Build Karpathy's LLM Wiki with Claude Code. L1/L2 cache architecture. Logseq + Obsidian support.
MCP server that lets Claude Code and other AI agents read and search large PDFs, one file or a whole folder: agentic RAG with hybrid semantic + keyword search, selective page reads, tables, images, OCR, chart data, and multi-column/CJK layouts.
DocMason is a repo-native agent that turns your complex office files into a local LLM knowledge base and your second brain. The repo is the app. Codex is the runtime.
Persistent memory plugin for Codex. Obsidian-style knowledge base that survives across sessions
A self-hosted knowledge platform for humans and AI agents — publish wikis, blogs, and portable Agent Skills.
Turn scattered notes, docs and transcripts into a queryable Markdown wiki — an LLM knowledge-base compiler with MCP access, no embeddings, self-hosted.
Open Source, Self-Hosted, AI Search and LLM.txt for your website
Agent Skill for building evidence-backed Markdown knowledge bases with zero-cost setup, image-aware capture, automatic wiki maintenance, and an interactive knowledge graph—without requiring a RAG stack or Obsidian.
Kappa Graph — κ(G). A semantic knowledge graph where knowledge has weight. Extracts concepts, measures grounding strength, preserves disagreement, traces everything to source.
Local, git-versioned memory for AI coding agents. No RAG, no Docker, no external service. Capture, compile, recall over a local LLM wiki with on-device embeddings and an MCP server.
Open Knowledge Format for coding agents. Author, validate, lint, search, and visualize portable Markdown knowledge bundles. The okf gem carries the agent skill, CLI, Ruby library, and an interactive graph; okf-mcp serves the same bundles to any MCP host. Docker and Claude Code plugin included, 100% local.
An agentic memory database that cuts session tokens by 82–99%. One portable SQLite file — your agent's memory, anywhere.
Local-first AI knowledge app and Agent Skill with evidence-backed Wiki, an interactive knowledge universe, Viki Q&A, and shareable knowledge galaxies.
OKF-powered knowledge context for Claude Code — injects your project's knowledge base at every session
```
/plugin marketplace add guhcostan/claude-mega-brain
/plugin install mega-brain@mega-brain
```A lightweight CPU only memory approach with ranked retrieval. Simple, yet effective.
MCP server providing AI agents with instant access to Apple developer documentation via RAG
A context harness for AI agents: all your scattered context — code, memory, docs, databases, SaaS — in one searchable, browsable, file-like interface.
Local-first RAG memory for AI agents: hybrid + GraphRAG search, MCP server with OAuth 2.1, plugin for Claude Code / OpenCode / Codex.
Local AI-powered document search and editing with first-in-class hybrid retrieval, LLM answers, WebUI, REST API and MCP support for AI clients.
A git-native, review-gated knowledge base for AI agents: they propose writes, you approve them. Every claim cites a source, every change is a diff in your repo. MCP + CLI.
基于 [Andrej Karpathy]提出的 [LLM Wiki 模式]构建的 Agent Skill，通过四阶段流水线将碎片化信息转化为结构化、可检索、持续增长的个人知识库。
An "LLM wiki" upgraded to a real database — typed entities, graph relations, HTTP API, and a built-in natural-language agent.
Open-source semantic document search (RAG) engine with FastAPI and instant self-hosted deployment
Self-hosted AI agent memory — MCP memory server with durable recall for agents
An AI-powered Quran knowledge graph — place verses on a canvas and discover thematic, linguistic, and theological connections.
Karpathy's LLM Wiki idea as a product — an AI that builds and maintains a markdown wiki from your notes and sources. MCP server + web UI, runs on free local models (Ollama), no API key needed. MIT.
Self-hosted RAG platform for AI document search across GitHub, Notion, Google Drive, local files, and web sources with citations.
Karpathy Wiki - Claude Code skills for building persistent, compounding knowledge bases. Based on Andrej Karpathy's LLM Wiki pattern.
AI-powered universal search for all your personal data, tailored just for you. Goal:The world's first product with "edge-side LLMs + consumer data localization" as its core development direction.
Andrej Karpathy's LLM Wiki pattern as a skill & Claude Code plugin — turn accumulated sources into a self-maintaining, scalable markdown knowledge base.
CLI tool for LLM agents to build and maintain personal knowledge bases
Indexed knowledge bases with command-line tools for agents.
OKF (Open Knowledge Format) — curated catalog of tools, plugins, skills, proposals, and docs for agent-friendly knowledge. YAML-driven, agent-searchable, MCP-ready.
Open-source knowledge base for people and AI agents. Rich-text editing, hybrid search, GraphRAG, and MCP. Self-host with Docker or use DocsMint Cloud.
An agent based LLM assistant that extends RAG with batch entity extraction and SQL querying to improve performance on multi-step and analytical questions.
Local-first AI daemon for Logseq OG: background semantic indexing, link hygiene, and agent-ready CLI/MCP — edits Markdown on disk (no cloud, no Logseq API). Karpathy LLM-Wiki inspired.
Turn your team's AI coding sessions, GitHub code, and docs into one shared, searchable memory — a self-hosted git truth store you query right from your editor over MCP.
A queryable second brain over your scattered notes and docs - hybrid retrieval (vector + BM25), section-level citations, and an MCP server so AI agents can use it. ~300 lines, no LangChain.
Modular Context | Karpathy LLM Knowledge Base + Gmail & G-Cal — multi-account MCP server for Claude Code, encrypted local-first
本地优先的 GitHub Stars AI 桌面应用，用于同步、摘要、标签、搜索和发现项目。Local-first AI desktop app for GitHub Stars.
Your brain, then your team's brain, then your agents' brain. Drop in documents and it compounds them into an interlinked markdown wiki you own — readable in Obsidian, synced through your own private GitHub repo. Share it with a cohort. Coding agents resume from it across sessions, models and machines.
Claude Code Plugin for Self-maintaining research knowledge graph for Claude Code + Obsidian
An agent skill to evolve the quality of LLM-Wiki (Graphify) at test time.
Second Brain automation for Obsidian vaults — entity management, ingestion, compression, and sync via Claude Code skills
The theory of LLM wikis, running as one. A framework for agent-operated knowledge: typed, linked, review-gated markdown your agents execute.
Two Claude Code skills for building a Karpathy-style LLM wiki — a compounding AI second brain. Install: git clone https://github.com/NulightJens/ai-second-brain-skills.git && ln -s $(pwd)/ai-second-brain-skills/llm-wiki-setup ~/.claude/skills/llm-wiki-setup && ln -s $(pwd)/ai-second-brain-skills/wiki-self-heal ~/.claude/skills/wiki-self-heal
Turn Google Drive PDFs into Obsidian wiki notes via NotebookLM MCP without loading full PDFs into Claude context
AI agent skill that remembers every technical decision & bug fix across sessions — and learns from them. v4.0, MIT. | 跨会话记忆的AI编程助手知识大脑
Semiont supports human+ai collaborative knowledge work. Use it as: a Wiki, Semantic Layer, Context Graph, Knowledge Base, Annotator, Research Tool, or Agentic Memory...
Local-first compile step for knowledge bases. Save 350x cost, 1000x speed vs. frontier models
Official Findings of EMNLP 2026 implementation of Corpus2Skill: compile a document corpus into a navigable skill hierarchy that LLM agents explore at query time, with document lookup instead of a serving-time vector-search service.
Socratic is a framework for building reliable vertical AI agents by letting human experts teach agents interactively, turning tacit domain knowledge into a continuously improving knowledge base.
Wenlan is a knowledge base for the AI-native age. Your AI agents capture what they learn, Wenlan keeps it current and distills it into source-cited wiki pages you can trust
Open source memory infrastructure for AI and humans. Turn scattered knowledge into AI-ready context across Claude, Cursor, ChatGPT, and any MCP client.
Cross-border e-commerce AI knowledge base, read by people and installed by agents: 69 trilingual guides, 878 prompts, a 100-entity / 322-constraint ontology, and 9 skills as a Claude Code plugin or over MCP. Factual claims are dated and CI-verified. CC0.
A beginner-friendly and extensible Agentic RAG project that demonstrates the full pipeline of document parsing, retrieval, reranking, workflow orchestration, tool calling, and answer generation, designed for both learning and secondary development.
经管实证研究的 AI 知识库 — 从文献阅读到 Stata 执行，一条流水线串到底。基于 Karpathy 的 LLM-Wiki 理念，按实证研究 10 类实体（变量 / 数据集 / 模型 / 机制 / 假设 / 识别策略 / 稳健性 / 异质性 / 表格 / 论文）定制
Skill for Claude, Cursor & Copilot that automates the Karpathy LLM Wiki workflow: ingest web, GitHub, and YouTube URLs into a well-structured, citable, cross-referenced knowledge base with automatic linting.
A compounding LLM-wiki, best fit as a personal second brain that organizes and updates itself as it grows, zero upkeep.
MyKG Knowledge Graph Engine: Turn raw files into knowledge graphs with induced ontology
Nia is a context-augmentation layer for agents, primarily designed for coding agents. It provides them with an up-to-date knowledge base and improves their performance by 27%.
A fast codebase indexer and knowledge wiki for AI agents.
Agent Fusion is a local RAG semantic search engine that gives AI agents instant access to your code, documentation (Markdown, Word, PDF). Query your codebase from code agents without hallucinations. Runs 100% locally, includes a lightweight embedding model, and optional multi-agent task orchestration. Deploy with a single JAR
Skill for a persistent LLM-managed wiki — the LLM writes and cross-references while you curate sources.
```bash
git clone https://github.com/sametbrr/llm-wiki-manager ~/.claude/skills/llm-wiki-manager
```Open-source, self-hostable markdown team wiki with a built-in MCP server — so AI agents read, write, and search your docs as first-class teammates. Live multiplayer, semantic + full-text search. Go + PostgreSQL + React/Milkdown.
LLM Wiki template — Karpathy 3-layer pattern + Gold In Gold Out purpose gate + dual Claude Code·Codex harness (11 commands, 2 hooks, 18 web clipper templates)
Turn your markdown vault into a compounding knowledge wiki (Karpathy inspired). Six agent skills - knowledge grows with every conversation. Works with Obsidian, Logseq, etc. or just folders on your local drive. Compiled memory for your LLM sessions. Crossplatform. GUI install on Claude Desktop, no terminal, no code.
Schema-as-code memory for AI agents in Obsidian: typed cards, entity dedup, link repair, update-in-place, and Ebbinghaus decay. Plain Markdown you own — a Claude Code skill.
Self-hosted memory server for AI agents. Your brain is a git repo, served over MCP and REST.
Karpathy's LLM Wiki idea as a product — an AI that builds and maintains a markdown wiki from your notes and sources. MCP server + web UI, runs on free local models (Ollama), no API key needed. MIT.
DeepSeek V4-Flash Vision RAG 让 AI 真正"看懂" 一份 PDF，然后你对它提问：它告诉你答案、答案在第几页， 并把那一页的原图展示出来给你核对。 基于 DeepSeek 视觉大模型 deepseek-v4-flash-vision-exp 的 PDF 深度问答与检索 （vision RAG）agent skill。支持文字版 PDF，也支持扫描版；能看懂 图表、表格、代码块、公式，而不只是认字。
DeepSeek V4-Flash Vision Video RAG 让 AI 真正"看懂" 一段视频，然后你对它提问：它告诉你答案、答案发生在 第几分几秒，并切出那一段的可播放片段和关键帧给你核对。 基于 DeepSeek 视觉大模型 deepseek-v4-flash-vision-exp 的视频理解与问答 （video RAG）agent skill。先按时间轴抽帧阅读、建立索引（一次性），再对问题做 本地粗筛 → 视觉精排 → 深读回答；回答带 [MM:SS] 时间戳引用，自动生成 自包含 HTML 预览页（内嵌可播放片段 + 关键帧 + 答案），双击浏览器即看。
Bedrock Knowledge Base and Agents for Retrieval Augmented Generation (RAG)
A local-first AI memory system with hybrid search, MCP integration, and a knowledge graph.
Find your files with natural language and ask questions.
LLM-managed personal knowledge base. Auto-extracts from git repos, Claude Code sessions (via ccrider FTS5), and iMessage bookmarks. Cross-project, qmd-indexed, runs on cron
Markdown vector search MCP server for Claude Code. Natural language search for markdown files using multilingual-e5-small embeddings.
Agentic memory for CTI in Python — STIX knowledge graphs, threat-actor alias resolution, offline-first RAG, MCP server for Claude Code and LangChain agents
Stop re-explaining your research to your AI agent. Persistent, LLM-maintained wikis that compound over time. Drop PDFs, URLs, YouTube - your agent remembers forever. Based on Karpathy's LLM Wiki pattern.
Karpathy-style personal wiki for LLMs - markdown + frontmatter, no vector DB, no RAG. Built for Claude Code, works with any file-editing AI agent.
A local dual-layer memory pattern for AI agents: a compact, human-readable markdown index paired with semantic retrieval from a local vector store, queried before each message. For cross-project recall where flat memory files or vector-only RAG fall short. Local-first. Reference implementation.
The whole local AI stack in one executable: it runs and manages local AI models across every GPU, and it's a search engine you can talk to, with cited answers from your files, code, and the web. MCP server for coding agents, web crawler, TUI, CLI, REST API, Python library. No Ollama or LM Studio needed, works with both.
Source code graph RAG (GraphRAG) for C/C++ development based on clang/clangd
Semi-automated research assistant and local knowledge base for paper analysis, ideation, coding, experiments, writing, and publication workflows.
Long-term, cross-project memory for AI coding agents. Your own Obsidian vault as the source of truth. Daemonless and without opaque databases, your memory belongs to you.
Local-first AI PKM for coding conversations: import Claude Code/Cursor/Codex, distill notes, semantic search, tag graph, MCP memory.
AI-powered second brain for Claude Code that builds itself. Extract knowledge from every session—past and present—into auto-organized Markdown. Local-first, queryable, learns your patterns.
一款跨平台智能笔记 Agent 应用，支持多格式笔记、本地知识库、AI 智能助手、向量语义检索 。
Zissa Wiki — an LLM-maintained research knowledge base in plain markdown, maintained by any AI agent (Claude, Codex, Gemini, DeepSeek, Kimi, Grok, Mistral or any chat app). Implements Karpathy's LLM Wiki pattern; checks every quote against its source.
A dual-path memory system for proactive agents. Facts and procedures in a local knowledge graph, exposed as a CLI, MCP tools and skills. No API key.
Agent-driven proactive memory CLI for AI agents — autonomously recall, maintain, and evolve persistent, source-grounded knowledge across sessions.
LLM using long-term memory through vector database
Enterprise-ready vector database toolkit for building searchable knowledge bases from multiple data sources. Supports multi-project management, automatic ingestion from Confluence/JIRA/Git, intelligent file conversion (PDF/Office/images), and semantic search. Includes MCP server for seamless AI assistant integration.
Give Claude semantic memory of your Obsidian vault — local semantic search over Smart Connections embeddings via MCP. Multi-vault, block-level, 100% private.
Shared, persistent memory for AI agents. Self-hosted MCP server with semantic search, vector RAG, and live updates. Works with Claude, Cursor, Codex, and any MCP client.
This MCP server provides tools for listing and retrieving content from different knowledge bases.
蒸留蔵 — distilled long-term memory for agents: recall by meaning, writing gated by evidence, one kura per agent mode. Ships as a DeepSeek Harness plugin and an MCP server.
A compile step for knowledge bases. Gives your agent a concept graph of your content — under 20s to index, 8ms queries on device.
A compile step for knowledge bases. Gives your agent a concept graph of your content — under 20s to index, 8ms queries on device.
Dragon Brain — persistent long-term memory for AI agents via MCP (Model Context Protocol). Knowledge graph (FalkorDB) + vector search (Qdrant) + CUDA GPU embeddings. Works with Claude, Gemini CLI, Cursor, Windsurf, VS Code Copilot. 30 tools, 1121 tests.
Autonomous knowledge base plugin for Claude Code - captures reserch, ideas, and decisions into an interlinked wiki with reserch-on-miss, semantic search, and a Wikipedia-style web UI. Knowledge compounds as you work.
Source-grounded MCP memory runtime and Obsidian-compatible vault template for LLM agents.
A reusable template for building personal LLM-maintained knowledge bases. Implements Karpathy's LLM Wiki pattern.
Agentic memory for CTI in Python — STIX knowledge graphs, threat-actor alias resolution, offline-first RAG, MCP server for Claude Code and LangChain agents
AI에게 '정리해줘'라고 하면 흩어진 자료를 옵시디언(원본)에 노트로 정리하고, 관련 노트끼리 연결·분류·요약하고, 선택적으로 노션에 색인하는 Claude Code 플러그인 + 이식형 MCP 코어. 보관함 자동탐지·정리·자동 링크/MOC·자동 분류·요약(원자 노트)·검색(가짜 인용 방지)·볼트 건강검진(중복·깨진 링크·고아 탐지, 삭제 없는 안전 수정)·작업 로그. 사람 승인 게이트·원문 보존 알림·경로/프롬프트 인젝션 방어·무의존.
Agent-ready DevOps, security, infrastructure, and compliance knowledge base with 80+ skills across Kubernetes, Terraform, AWS/Azure/GCP, AI platform operations, container hardening, SOC2/ISO27001, and incident response—plus ready-to-run scripts, templates, and playbooks for SRE, platform, and security teams.
Claude Code skill for building persistent, interlinked knowledge bases from source documents. Knowledge is compiled once and kept current — never re-derived per query. Based on Karpathy's LLM Wiki pattern.
Your markdown vault, compiled into a 6-persona MCP team for Claude Code, Codex, OpenCode, and Gemini CLI. Headless-first. Cites, doesn't guess.
Your AI forgets. A wiki doesn't. Turn chat-pasted text/links into a verified, translated, searchable wiki knowledge base — verify links, filter ads, summarize in your language, save to a wiki, and alert on time-sensitive items. Zero dependencies (curl + Python stdlib), platform-independent agent skill for Claude Code / Codex / Cursor / Hermes.
基于 Karpathy llm-wiki 方法论的 Claude Code 个人知识库构建 Skill
Local-first personal knowledge base CLI — hybrid FTS + pgvector search, a GraphRAG entity graph, and LLM enrichment over your notes, transcripts, Slack & Gmail. Runs fully local on Ollama (no cloud, no API keys); queryable by any AI agent through the built-in MCP server. Postgres-backed, publishes an Obsidian-style Quartz wiki.
A compounding knowledge base maintained by LLM agents — inspired by Karpathy's LLM Wiki pattern
File-first knowledge base for humans and AI agents — self-hosted Markdown notes with a web editor and a built-in MCP server.
知识驱动的昇腾训练/推理诊断 skill 套件 — Ascend training/inference diagnosis skill suite (5 skills + 3-tier knowledge base, Agent Skills standard)
| Tool | Stars | Language | License | Score |
|---|---|---|---|---|
| ragflow | ★ 90.9k | Python | Apache-2.0 | 73 |
| codegraph | ★ 73.1k | C | MIT | 82 |
| LightRAG | ★ 40.0k | Python | MIT | 75 |
| OpenViking | ★ 39.2k | Python | AGPL-3.0 | 82 |
| book-to-skill | ★ 33.0k | Python | MIT | 79 |
| cognee | ★ 31.3k | Python | Apache-2.0 | 81 |
| MaxKB | ★ 22.9k | Python | GPL-3.0 | 79 |
| deepwiki-open | ★ 17.9k | Python | MIT | 73 |
| openwiki | ★ 16.9k | TypeScript | MIT | 80 |
| claude-obsidian | ★ 14.8k | Python | MIT | 85 |
| Yuxi | ★ 7.2k | Python | MIT | 73 |
| airweave | ★ 6.6k | Python | MIT | 63 |
| code-graph-rag | ★ 5.2k | Python | MIT | 75 |
| obsidian-second-brain | ★ 4.6k | Python | MIT | 72 |
| m_flow | ★ 4.5k | Python | Apache-2.0 | 75 |
| pipeshub-ai | ★ 3.8k | Python | Apache-2.0 | 73 |
| knowhere | ★ 3.6k | Python | Apache-2.0 | 72 |
| llm-wiki-agent | ★ 3.6k | Python | MIT | 77 |
| ReMe | ★ 3.5k | Python | Apache-2.0 | 71 |
| obsidian-wiki | ★ 3.5k | Python | MIT | 69 |
| arscontexta | ★ 3.5k | Shell | MIT | 57 |
| memsearch | ★ 2.6k | Python | MIT | 69 |
| karpathy-llm-wiki | ★ 2.4k | Python | MIT | 67 |
| m_flow | ★ 2.4k | Python | Apache-2.0 | 68 |
| mcp-crawl4ai-rag | ★ 2.3k | Python | MIT | 58 |
| llm-wiki-compiler | ★ 2.2k | TypeScript | MIT | 74 |
| edgeever | ★ 2.0k | TypeScript | AGPL-3.0 | 74 |
| mcp | ★ 1.9k | TypeScript | CC-BY-4.0 | 72 |
| mex | ★ 1.8k | TypeScript | MIT | 69 |
| llmwiki | ★ 1.6k | Python | Apache-2.0 | 72 |
| synthadoc | ★ 1.5k | Python | AGPL-3.0 | 66 |
| chunkhound | ★ 1.4k | Python | MIT | 71 |
| llm-wiki | ★ 1.4k | Python | MIT | 73 |
| repobrain | ★ 1.3k | Python | MIT | 67 |
| ApeRAG | ★ 1.3k | Python | Apache-2.0 | 56 |
| antigravity-workspace-template | ★ 1.3k | Python | MIT | 66 |
| OpenContext | ★ 1.2k | JavaScript | MIT | 64 |
| arkon | ★ 1.2k | Python | — | 61 |
| chubbyskills | ★ 1.2k | Python | MIT | 69 |
| KnowNote | ★ 1.2k | TypeScript | GPL-3.0 | 72 |
| jonex | ★ 1.1k | Python | — | 68 |
| MODULAR-RAG-MCP-SERVER | ★ 1.1k | Python | — | 51 |
| minima | ★ 1.0k | Python | MPL-2.0 | 49 |
| autollm | ★ 1.0k | Python | AGPL-3.0 | 44 |
| LaunchStack | ★ 890 | TypeScript | Apache-2.0 | 65 |
| codegraph-rust | ★ 885 | Rust | — | 40 |
| second-brain-os | ★ 813 | HTML | MIT | 67 |
| automem | ★ 803 | Python | MIT | 67 |
| second-brain-cloudflare | ★ 800 | TypeScript | MIT | 66 |
| codealmanac | ★ 791 | TypeScript | Apache-2.0 | 70 |
| memora | ★ 729 | Python | MIT | 66 |
| MindOS | ★ 677 | TypeScript | MIT | 65 |
| RAGLight | ★ 672 | Python | MIT | 66 |
| swarmvault | ★ 649 | TypeScript | MIT | 71 |
| crypto-rag | ★ 648 | Python | MIT | 76 |
| llm-wiki-skill | ★ 644 | TypeScript | — | 59 |
| docsagent | ★ 623 | TypeScript | — | 63 |
| haiku.rag | ★ 616 | Python | MIT | 70 |
| MinerU-Document-Explorer | ★ 612 | TypeScript | MIT | 53 |
| pi-llm-wiki | ★ 599 | TypeScript | MIT | 68 |
| MisakaNet | ★ 520 | Python | Apache-2.0 | 68 |
| obsidian-llm-wiki | ★ 458 | TypeScript | Apache-2.0 | 65 |
| graphrag-toolkit | ★ 444 | Python | Apache-2.0 | 69 |
| beever-atlas | ★ 444 | Python | Apache-2.0 | 63 |
| Molio | ★ 431 | TypeScript | — | 65 |
| mcp-apple-notes | ★ 415 | TypeScript | — | 46 |
| mcp-local-rag | ★ 405 | TypeScript | MIT | 71 |
| Kwipu | ★ 374 | Python | MIT | 70 |
| agent-second-brain | ★ 367 | Python | MIT | 71 |
| llm-wiki | ★ 362 | Python | MIT | 58 |
| mcp-documentation-server | ★ 342 | TypeScript | MIT | 69 |
| mcp-logseq | ★ 338 | Python | MIT | 69 |
| session-knowledge | ★ 335 | Python | MIT | 76 |
| Ori-Mnemos | ★ 328 | TypeScript | Apache-2.0 | 66 |
| llm-wiki-compiler | ★ 325 | HTML | MIT | 56 |
| arag | ★ 323 | Python | — | 47 |
| tw-legal-rag | ★ 323 | Python | — | 68 |
| MegaMemory | ★ 313 | TypeScript | MIT | 55 |
| file-system-like-github | ★ 296 | TypeScript | — | 68 |
| Revornix | ★ 292 | TypeScript | — | 62 |
| pharos | ★ 292 | Python | MIT | 69 |
| rust-docs-mcp-server | ★ 291 | Rust | MIT | 47 |
| google-surf-mcp | ★ 290 | TypeScript | MIT | 74 |
| knowledge-rag | ★ 288 | Python | MIT | 75 |
| BYO-LLM-WIKI | ★ 280 | Python | — | 59 |
| marginalia | ★ 247 | Python | AGPL-3.0 | 72 |
| PharosRAG | ★ 243 | Python | MIT | 68 |
| BYO-WIKI | ★ 237 | Python | — | 60 |
| serverless-rag-demo | ★ 224 | TypeScript | MIT-0 | 55 |
| skills | ★ 223 | Python | MIT | 62 |
| LeanKG | ★ 220 | Go | Apache-2.0 | 65 |
| erag | ★ 214 | Python | — | 41 |
| llm-wiki | ★ 212 | HTML | MIT | 73 |
| llm-wiki | ★ 208 | — | — | 66 |
| graphmind | ★ 207 | Rust | MIT | 63 |
| student-llm-wiki | ★ 196 | — | MIT | 65 |
| OnCo | ★ 194 | TypeScript | MIT | 65 |
| flexible-graphrag | ★ 188 | Python | Apache-2.0 | 61 |
| FirstData | ★ 183 | Python | MIT | 70 |
| wiki-skills | ★ 174 | Python | MIT | 67 |
| awesome-ascend-skills | ★ 170 | Python | — | 62 |
| file-system-brain-mcp | ★ 168 | TypeScript | — | 64 |
| Axon.MCP.Server | ★ 167 | Python | — | 52 |
| llm-wiki-newsroom | ★ 167 | Python | MIT | 70 |
| zettelkasten-mcp | ★ 164 | Python | MIT | 44 |
| Nexus-skills | ★ 163 | Python | — | 55 |
| akb | ★ 161 | Python | — | 61 |
| okf | ★ 154 | Ruby | Apache-2.0 | 65 |
| mcp-pinecone | ★ 149 | Python | MIT | 44 |
| scholar-rag-agent | ★ 149 | Python | — | 64 |
| mcptube | ★ 148 | Python | — | 47 |
| llm-wiki | ★ 148 | Shell | MIT | 75 |
| pdf-mcp | ★ 147 | Python | MIT | 66 |
| DocMason | ★ 146 | Python | Apache-2.0 | 61 |
| echoes-vault-codex | ★ 143 | Python | MIT | 75 |
| Petrichor | ★ 142 | TypeScript | Apache-2.0 | 62 |
| third-brain-v7-skills | ★ 141 | Python | MIT | 70 |
| kaas | ★ 137 | Python | MIT | 71 |
| ragrabbit | ★ 136 | TypeScript | MIT | 42 |
| my-wiki-skill | ★ 130 | JavaScript | MIT | 59 |
| knowledge-graph-system | ★ 128 | Python | Apache-2.0 | 65 |
| llm-wiki-memory | ★ 127 | JavaScript | MIT | 70 |
| okf-gem | ★ 126 | Ruby | Apache-2.0 | 65 |
| remindb | ★ 124 | Go | MIT | 53 |
| my-wiki | ★ 124 | JavaScript | MIT | 55 |
| third-brain-v5-skills | ★ 123 | HTML | MIT | 69 |
| claude-mega-brain | ★ 122 | Python | MIT | 73 |
| heimdall | ★ 121 | JavaScript | MIT | 61 |
| apple-rag-mcp | ★ 120 | TypeScript | MIT | 62 |
| mfs | ★ 119 | Python | Apache-2.0 | 60 |
| agent-brain | ★ 119 | Python | MIT | 65 |
| gno | ★ 115 | TypeScript | MIT | 62 |
| vouch | ★ 112 | Python | MIT | 62 |
| karpathy-wiki | ★ 112 | Python | MIT | 60 |
| braindb | ★ 109 | Python | Apache-2.0 | 64 |
| Flamehaven-Filesearch | ★ 106 | Python | MIT | 57 |
| metronix-memory | ★ 106 | Python | Apache-2.0 | 63 |
| open-hikmah | ★ 105 | TypeScript | GPL-3.0 | 54 |
| mindbase-llm-wiki | ★ 105 | TypeScript | MIT | 69 |
| OpenDocuments | ★ 103 | TypeScript | MIT | 63 |
| karpathy-wiki | ★ 103 | Shell | MIT | 65 |
| quanta-quest | ★ 102 | TypeScript | — | 44 |
| llm-wiki-plugin | ★ 102 | TypeScript | MIT | 72 |
| llmwiki-cli | ★ 101 | TypeScript | MIT | 69 |
| wiki | ★ 101 | Python | Apache-2.0 | 72 |
| Awesome-OKF | ★ 100 | Python | MIT | 67 |
| docsmint | ★ 99 | TypeScript | Apache-2.0 | 69 |
| aws-agentic-document-assistant | ★ 98 | Jupyter Notebook | MIT-0 | 40 |
| matryca-plumber | ★ 98 | Python | Apache-2.0 | 61 |
| Klear-Team-Brain | ★ 98 | JavaScript | Apache-2.0 | 65 |
| loci | ★ 98 | Python | MIT | 68 |
| modular-context-obsidian-plugin | ★ 97 | TypeScript | — | 60 |
| GitHub-Stars-AI-Tools | ★ 97 | TypeScript | — | 70 |
| the-curator | ★ 97 | JavaScript | — | 57 |
| claude-paperloom | ★ 94 | JavaScript | Apache-2.0 | 51 |
| DeepRefine-Skill | ★ 92 | Python | MIT | 67 |
| claude-bedrock | ★ 91 | HTML | MIT | 49 |
| commonplace | ★ 91 | Python | CC-BY-4.0 | 64 |
| ai-second-brain-skills | ★ 90 | — | MIT | 58 |
| notebooklm-wiki-pipeline | ★ 90 | Python | MIT | 51 |
| obsidian-knowledge-brain | ★ 90 | Python | MIT | 70 |
| semiont | ★ 89 | TypeScript | Apache-2.0 | 48 |
| enzyme | ★ 86 | TypeScript | — | 67 |
| Corpus2Skill | ★ 85 | Python | MIT | 67 |
| Socratic | ★ 81 | JavaScript | Apache-2.0 | 53 |
| wenlan | ★ 79 | Rust | Apache-2.0 | 59 |
| streamient | ★ 79 | JavaScript | AGPL-3.0 | 57 |
| ecommerce-ai-skills | ★ 79 | Python | CC0-1.0 | 67 |
| scholar-rag | ★ 77 | Python | MIT | 48 |
| empiricalwiki | ★ 77 | Python | MIT | 58 |
| pin-llm-wiki | ★ 76 | Python | Apache-2.0 | 65 |
| Kompl | ★ 76 | TypeScript | Apache-2.0 | 61 |
| mykg | ★ 75 | Python | MIT | 60 |
| nia | ★ 73 | — | MIT | 57 |
| indxr | ★ 72 | Rust | MIT | 59 |
| Agent-Fusion | ★ 72 | Kotlin | MIT | 50 |
| llm-wiki-manager | ★ 71 | Python | MIT | 75 |
| tela | ★ 70 | Go | AGPL-3.0 | 58 |
| cmds-llm-wiki | ★ 70 | Python | — | 66 |
| llm-wiki-skills | ★ 69 | — | MIT | 67 |
| autograph | ★ 68 | Python | MIT | 61 |
| brainoutside | ★ 68 | Python | MIT | 61 |
| mindbase | ★ 68 | TypeScript | MIT | 63 |
| deepseek-v4-flash-vision-rag | ★ 67 | Python | — | 68 |
| obsidian-vault-intelligence | ★ 66 | TypeScript | MIT | 54 |
| deepseek-v4-flash-vision-video-rag | ★ 66 | Python | — | 66 |
| bedrock-kb-rag-workshop | ★ 65 | HTML | MIT-0 | 56 |
| memory-vault | ★ 65 | Python | MIT | 66 |
| Archive-Agent | ★ 64 | Python | GPL-3.0 | 61 |
| scribe | ★ 64 | Go | MIT | 60 |
| devrag | ★ 63 | Go | — | 56 |
| zettelforge | ★ 63 | Python | MIT | 56 |
| Tapestry | ★ 62 | Python | MIT | 56 |
| llm-wiki-kit | ★ 62 | Python | MIT | 64 |
| karpathy-claude-wiki | ★ 62 | Python | MIT | 53 |
| zer0dex | ★ 62 | Python | Apache-2.0 | 72 |
| lilbee | ★ 61 | Python | MIT | 59 |
| clangd-graph-rag | ★ 60 | Python | Apache-2.0 | 58 |
| BITE | ★ 60 | Python | MIT | 59 |
| agentcairn | ★ 60 | Python | Apache-2.0 | 61 |
| ChatCrystal | ★ 58 | TypeScript | Apache-2.0 | 68 |
| remember | ★ 58 | JavaScript | MIT | 61 |
| open-note | ★ 58 | Dart | — | 59 |
| zissa-wiki | ★ 58 | Python | MIT | 66 |
| memoose | ★ 58 | Python | Apache-2.0 | 63 |
| llm-wiki-cli | ★ 56 | Rust | Apache-2.0 | 70 |
| llm-agent | ★ 55 | Python | — | 44 |
| qdrant-loader | ★ 55 | Python | Apache-2.0 | 56 |
| smart-connections-mcp | ★ 54 | TypeScript | MIT | 72 |
| montycat-mcp | ★ 54 | Python | MIT | 64 |
| knowledge-base-mcp-server | ★ 53 | TypeScript | Unlicense | 60 |
| distill-kura | ★ 53 | Python | MIT | 69 |
| enzyme | ★ 52 | TypeScript | — | 55 |
| enzyme | ★ 51 | TypeScript | — | 59 |
| Dragon-Brain | ★ 51 | Python | MIT | 62 |
| LLM-Wiki | ★ 51 | Python | MIT | 58 |
| osk-system | ★ 51 | Python | — | 57 |
| llm-wiki-template | ★ 50 | Python | — | 66 |
| zettelforge | ★ 50 | Python | MIT | 62 |
| SoDam-WikiMate | ★ 50 | JavaScript | Apache-2.0 | 60 |
| DevOps-Security-Agent-Skills | ★ 49 | Shell | MIT | 51 |
| scraps | ★ 46 | Rust | MIT | 60 |
| llm-wiki | ★ 42 | Shell | MIT | 61 |
| obsidian-llm-wiki | ★ 36 | TypeScript | GPL-3.0 | 61 |
| wiki-knowledge-agent | ★ 19 | Python | MIT | 63 |
| llm-wiki-skill | ★ 18 | Shell | — | 51 |
| second-brain | ★ 14 | Python | MIT | 63 |
| llm-wiki | ★ 10 | Python | MIT | 71 |
| notarium | ★ 5 | TypeScript | AGPL-3.0 | 54 |
| ascend-sleuth | ★ 5 | Python | MIT | 53 |
The top knowledge base & rag tools in 2026 are ragflow, codegraph, LightRAG. Agent Skills Hub ranks 230 options by GitHub stars, quality score (6 dimensions including completeness, examples, and agent readiness), and recent activity. The list is rebuilt every 8 hours from live GitHub data.
ragflow (90.9k stars) is the most adopted choice for general knowledge base & rag workflows, written in Python. codegraph (73.1k stars) is a strong alternative and uses C instead. Pick by your existing stack: match the language and runtime your team already uses to minimize integration cost. If unsure, start with ragflow — it has the deepest community and the most examples online.
Avoid pre-built knowledge base & rag tools when (1) your use case requires deep customization that the tool's plugin system doesn't support, (2) you have strict compliance requirements that ban third-party dependencies, (3) the tool's maintenance is inactive (last commit >6 months ago), or (4) your data volume is small enough that a 50-line custom script is cheaper than learning the tool. For most production workflows above 100 requests/day, the time savings from a maintained tool outweigh the customization loss.
Knowledge Base & RAG focuses specifically on build ai-powered knowledge bases with retrieval-augmented generation (rag) — ingest documents, search semantically, and answer questions. Semantic Search is a related but distinct category — see https://agentskillshub.top/best/semantic-search/ for those tools. The two often appear in the same agent pipeline but solve different problems: choose knowledge base & rag when your primary goal is the specific task, and semantic search when the workflow is broader.
For most teams, yes. ragflow has 90.9k stars worth of community testing, handles edge cases you haven't thought of, and ships with documentation. Build your own only when (1) your requirements are deeply non-standard, (2) you have a security/compliance reason to avoid OSS dependencies, or (3) the maintenance burden is small enough (<200 lines of code) that you'll save time long-term. The break-even point is usually around 2-3 weeks of dev time saved.
Most knowledge base & rag tools listed are open source under permissive licenses (MIT, Apache 2.0). A handful offer paid managed/cloud versions on top of free self-hosted core. Always check the LICENSE file on each tool's GitHub repository before commercial use — some use AGPL or non-commercial restrictions that may not fit your deployment model.
Grades come from a rule-based scan built on the SlowMist agent-security taxonomy, covering 11 red-flag categories including credential harvesting, data exfiltration, and curl | sh installers. It is a first-layer scan, not a manual audit — we say so rather than overstate it.
The scale of the problem is documented independently: Liu et al. (2026), in a study of 31,132 agent skills, report that 26.1% contain security vulnerabilities. Our own full-catalog census is published as a citable open dataset.
Sources & who's responsible: