BigHugger
GH Repository · SamurAIGPT

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.

stars
3,528
30-day movement
+186/day
Related entries
60
Connections
5
Pythonsecond-brainnote-takingautomationopen-sourcemuapiclaude-codeknowledge-basepythonai-toolsllmwikiobsidianknowledge-graphmarkdownresearchcodexgenerative-airaggeminipersonal-knowledge-managementai-agent

llm-wiki-agent is a self-maintaining personal knowledge base: you drop in sources, and a coding agent such as Claude Code, Codex, OpenCode, or Gemini CLI reads them, extracts knowledge, and keeps a persistent interlinked wiki updated. It ships as configuration for those agents (AGENTS.md, CLAUDE.md, GEMINI.md files) in Python and needs no API key.

Use it to have your coding agent turn raw sources into an organized, interlinked markdown wiki without manual curation.

Use it to

  • Build a personal wiki from dropped-in source documents
  • Maintain interlinked notes automatically with Claude Code or Codex
  • Create a second brain in markdown compatible with tools like Obsidian
  • Extract and organize research knowledge via an agent CLI

For Researchers and note-takers using coding agent CLIs

Role
rag
Language
Python
Licence
MIT
Forks
402
Open issues
3
Last push
2026-09-14
topicsknowledge-basewikiclaude-coderagmarkdownpersonal-knowledge-management