llm_wiki
LLM Wiki is a cross-platform desktop application that turns your documents into an organized, interlinked knowledge base — automatically. Instead of traditional RAG (retrieve-and-answer from scratch every time), the LLM incrementally builds and maintains a persistent wiki from your sources。
LLM Wiki is a cross-platform desktop application that transforms documents into a persistent, interlinked knowledge base using large language models. Rather than traditional retrieve-and-generate on every query, it incrementally builds and maintains a structured wiki from your sources, preserving relationships and context over time. The application includes a local HTTP API and bundled MCP server that expose hybrid search, graph traversal, and chat capabilities to AI agents like Claude Code, Codex, and any MCP-compatible client. A companion agent skill can be installed with a single npx command to let your AI assistant query and cite your personal wiki directly.
Key Features
Use Cases
- 01Building a research knowledge base that incrementally indexes academic papers, extracts entities and concepts, and maintains cross-references automatically
- 02Querying your personal wiki from Claude Code or other AI agents using the MCP server or installable skill without leaving your IDE
- 03Capturing web articles via the Chrome extension and having them analyzed and integrated into your knowledge graph overnight
- 04Discovering unexpected connections between research topics through graph insights, surprising connections, and community detection
- 05Running deep research on identified knowledge gaps, with the LLM synthesizing web findings into wiki pages and extracting new entities
- 06Migrating a complete knowledge base across machines using ZIP export/import while preserving all relationships, embeddings, and conversation history
Related Skills
View moresuperpowers
An agentic skills framework & software development methodology that works.
skills
Skills for Real Engineers. Straight from my .agents directory.
skills
Public repository for Agent Skills
ponytail
Makes your AI agent think like the laziest senior dev in the room. The best code is the code you never wrote.
llm_wiki — FAQ
What is LLM Wiki and how does it differ from traditional RAG?+
LLM Wiki is a desktop application that builds a persistent, structured wiki from your documents using large language models. Instead of retrieving and generating answers from scratch each time, it maintains a living knowledge graph with entities, concepts, and cross-references that evolve as you add sources.
How do I install the LLM Wiki MCP server or agent skill?+
The MCP server is bundled with the desktop app; after enabling it in Settings → API + MCP, copy the generated configuration into your MCP client (Claude Desktop, etc.). For Claude Code or Codex, install the companion skill with `npx skills add https://github.com/nashsu/llm_wiki_skill.git --skill llm-wiki`.
Which AI clients work with LLM Wiki?+
The MCP server works with any MCP-compatible client including Claude Desktop. The agent skill works with Claude Code, Codex, and other skills-compatible runtimes. The HTTP API can be called by any tool or script that supports HTTP requests.
Do I need an API key or paid service?+
You need an LLM API key (OpenAI, Anthropic, Google, or any compatible provider) for document ingestion and chat. Optional features like vector search, web research (Tavily/SerpApi), and advanced PDF parsing (MinerU Cloud) require separate API keys, but core functionality works with just an LLM provider.
Is LLM Wiki free and open source?+
Yes, LLM Wiki is licensed under GNU GPL v3.0 and available as open source. Pre-built binaries are provided for macOS, Windows, and Linux. You pay only for the LLM API calls to your chosen provider.
Can I use LLM Wiki with local models?+
Yes, LLM Wiki supports Ollama and any custom OpenAI-compatible endpoint, so you can use local models. The configurable LLM timeout setting helps accommodate slower local inference. Vector search also supports any OpenAI-compatible embeddings endpoint.
How do I install llm_wiki?+
Open the source repository on GitHub and follow its README. llm_wiki is a skill — MCP Agents Market links you directly to the official repo.
Is llm_wiki free?+
llm_wiki is an open-source project hosted on GitHub. Check the repository for its license and any usage requirements.