Back to directory

LLM Wiki Agent

A coding-agent skill that reads documents dropped into a folder and builds a persistent, cross-referenced markdown wiki that grows richer with every new source.

LLM Wiki Agent

LLM Wiki Agent is a coding-agent skill: source documents dropped into a raw folder are read by the agent, which extracts knowledge from them and builds a persistent, interlinked wiki that compounds with every new source added.

It works with Claude Code, Codex, OpenCode or Gemini CLI, reading each tool's own configuration file, and needs no API key or separate Python setup. It understands natural-language requests to ingest a source, query the wiki, check for contradictions across sources, or build and inspect the underlying knowledge graph, and Claude Code additionally exposes these as slash commands. Non-markdown formats such as PDF, DOCX, PPTX, XLSX, HTML and EPUB are automatically converted to markdown at ingestion time. The wiki accumulates structured pages for entities such as people, companies and projects, and for key concepts, each updated whenever a new source references them.

It is written in Python and released under the MIT license.

More in Agent skills

See all 581
  • Provenance gate

    github.com

    A deterministic authorization gateway that sits between an AI agent and its tools, tracing where each parameter of a side-effect call came from, and blocking the call or asking a human when the source does not check out.

  • A markdown framework of system rules that restates Isaac Asimov's Four Laws of Robotics as a set of prioritized loss calculations for large-language-model agents to follow, published as a discovery file for agents and crawlers to read.

  • Vitruvius

    github.com

    An engineering research agent that runs a discover, read, synthesize, verify and review loop over technical questions, reading sources directly rather than inferring from titles, and publishing its own scored verifier benchmark, weaknesses included.