knowledge-management

From saved sources to traceable signal. A local-first research intake engine for Claude Code, Codex, and Obsidian.

Master mind Board Game implemented in Python

A local research engine for long-form writing. Verifies every quote against its source, refuses numbers that came from nowhere, and makes you pre-register the answer you expect before you search.

LLM-maintained second brain starter kit: Karpathy's LLM Wiki pattern plus a Raw/wiki split, session gates, and integrity scripts. Built by an AI clone and its human.

A Python CLI that turns images, PDFs, URLs, and text into structured Notion notes.

Hermes Knowledge Review 是一个面向个人、私有部署场景的 AI 知识审核与复习系统。 AI Agent 通过 MCP 提交知识草稿，用户在 Web 管理端审核、合并和维护；随后由系统发起复习、记录回答与评分，并通过统计帮助发现薄弱知识点。 它强调本地数据所有权、人工审核闭环和安全部署，适合将 AI 对话、学习笔记和工作沉淀变成长期可复用的知识库。

A personal knowledge flywheel for AI-assisted development - session memory, a living wiki, code graphs, and daily logs in one self-feeding loop

KK Nexus - Multi-Agent Knowledge OS Framework. Build your own multi-employee knowledge system with vault-based knowledge management, txtai semantic search, and WAL protocol.

Lumidoc is an AI-powered documentation platform that transforms codebases into structured, searchable, and developer-friendly documentation with automated analysis, generation, and maintenance workflows.

An evidence-gated belief system for AI agents. Every belief has a lifecycle. No falsification condition, no promotion.

Deterministic pipeline for turning fragmented conversations and documents into structured, traceable knowledge.

Claude Code skill: turns a YouTube video into a clean transcript and a structured conspectus. Fetches captions via yt-dlp and deduplicates auto-caption repetition deterministically.

Governed procedural memory for AI agents — turn corrections, source material, and proven workflows into review-gated, versioned recipes agents can safely retrieve, lock, execute, and improve.

Local-first personal knowledge base dashboard with demo mode, Markdown search, safety checks, and public showcase screenshots.

多 Agent 協作知識治理系統：LLM/深度學習知識庫精選 + 3M 三神祇架構（L0-L3 分級、handoff、規格仲裁）| Multi-agent knowledge governance: task tiering, handoff protocols, spec arbitration & full auditability

Source-first, auditable knowledge inflow pipeline and Obsidian dashboard

🌟 Lumiere: Multi-agent RAG system with semantic memory. Combines LangGraph, Qdrant vector search, and OpenAI for intelligent document Q&A, SQL data analysis, and context-aware conversations. Features long-term learning, critic validation, and full observability.

Store, query, and generate knowledge in the Open Knowledge Format (OKF) - markdown bundles with YAML frontmatter, trust signals, CLI, and LLM ingestion.

基于 RAG 技术的高质量知识管理系统