llm-wiki-agent

llm-wiki-agent is a personal information management tool that creates a permanent and interconnected knowledge base by processing source documents. Claude works with AI agents like Codex or Gemini to read resources, extract information, and keep a structured wiki up to date. The system includes index.md, source summaries, auto-generated entity and concept pages, and graph.json data structure.

LL

SamurAIGPT · Wiki & Notes / Notion Alternatives

STARS3.7K
FORKS426
COMMITS0
CONTRIBUTORS0
VERSION—
UPDATED5 days ago
PythonBilgi TabanıAIDocumentationMIT

Common commercial products in this category: Notion, Confluence, Evernote

What can it do?
  • You can import source documents collectively or as singles.
  • You can query by synthesizing responses from wiki pages.
  • You can run a review to detect missing links and contradictions.
Heads up
  • • No published release
User reviews
  • ↓ NegativeGitHub Issue

    [Feature] Graph-driven wiki maintenance: feed structural insights back into wiki ops

    “## Problem The knowledge graph (`tools/build_graph.py`) currently operates as a **read-only visualization endpoint**: ``` wiki pages → build_graph.py → graph.json + graph.html → user looks at the graph → end ``` The graph computes valuable structural insights — community…”

  • • NeutralGitHub Issue

    [Bug/UX] Two critical issues in build_graph.py: Cache state-loss for inferred edges & Visual text clustering for labels

    “Hi there, thanks for this brilliant LLM-OS implementation! During my robust ingestion test covering 40+ pages, I encountered two critical issues in the `build_graph.py` script. I've resolved them locally and wanted to share the feedback here. **1. Catastrophic Cache State-Loss…”

GitHub

My note

Alternative pages

More in: Wiki & Notes / Notion Alternatives