llm-wiki-agent
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.
llm-wiki-agent is an agent skill that enables coding agents to build and maintain a self-organizing personal knowledge base from source documents. Drop PDFs, markdown, or DOCX files into a raw folder, and the agent reads them, extracts entities and concepts, cross-references information, and writes structured, interlinked wiki pages that persist and compound over time. Unlike traditional RAG or note-taking tools, it pre-compiles knowledge into a navigable wiki with entity pages, concept pages, contradiction flags, and an interactive knowledge graph, eliminating the need to re-derive answers from raw chunks on every query.
Key Features
Use Cases
- 01Research synthesis: accumulate papers over weeks/months with auto-generated entity and concept pages, query for patterns, and visualize topic clusters
- 02Book annotation: ingest chapters incrementally to build character/theme wikis with cross-references, similar to fan wikis like Tolkien Gateway
- 03Personal knowledge management: track journal entries, articles, and podcast notes with automatic habit/pattern extraction and contradiction detection
- 04Business intelligence: ingest meeting transcripts, product docs, and customer calls to surface feature requests and decision rationales across sessions
- 05Competitive analysis: monitor companies, markets, or technologies by filing announcements and reports, then query for comparative summaries
- 06Team documentation: maintain living project wikis where the agent handles cross-referencing, orphan detection, and synthesis as new sources arrive
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-agent — FAQ
What is llm-wiki-agent?+
llm-wiki-agent is an agent skill that teaches coding agents like Claude Code, Codex, and Gemini CLI to read source documents, extract knowledge, and build a persistent interlinked wiki with entity pages, concept pages, and knowledge graphs. It turns your reading into a structured knowledge base that compounds over time without manual maintenance.
How do I install llm-wiki-agent?+
Clone the repository, cd into the directory, then open it with your coding agent (claude, codex, opencode, or gemini). The agent reads the schema file (CLAUDE.md, AGENTS.md, or GEMINI.md) and gains wiki-building capabilities. No API key or Python setup is required to start using it.
Which AI clients work with llm-wiki-agent?+
Claude Code, Codex, OpenCode, and Gemini CLI are officially supported. Any coding agent that can read a schema/config file and execute file operations can theoretically use it.
Do I need an API key to use llm-wiki-agent?+
No API key is needed when using it with Claude Code, Codex, or other supported agents. The standalone Python scripts in the tools/ directory require ANTHROPIC_API_KEY if you run them outside of an agent context.
Is llm-wiki-agent free?+
Yes, llm-wiki-agent is open source under the MIT License and free to use.
What file formats can llm-wiki-agent ingest?+
It supports markdown, PDF, DOCX, PPTX, XLSX, HTML, TXT, CSV, JSON, XML, RST, EPUB, and more. Non-markdown files are auto-converted via markitdown at ingest time with no separate conversion step needed.
How do I install llm-wiki-agent?+
Open the source repository on GitHub and follow its README. llm-wiki-agent is a skill — MCP Agents Market links you directly to the official repo.
Is llm-wiki-agent free?+
llm-wiki-agent is an open-source project hosted on GitHub. Check the repository for its license and any usage requirements.