obsidian-llm-wiki-local

Why we included this project

This one is for Obsidian users who want their notes to link themselves up without ever leaving the machine. It runs a local model through Ollama by default, reads each Markdown file you drop into a folder, pulls out the concepts, and writes or updates wiki articles you can browse in the graph view. The loop is what makes it interesting: only articles tied to a changed note recompile, your hand edits survive, and when you reject a draft it remembers why and fixes it on the next pass. It can also answer questions straight from the compiled wiki, no embeddings or vector database involved, which keeps it lighter than a full RAG setup for personal notes. One thing to know before adopting it: the project is in maintenance mode and new work has moved to the Synto successor, though the current tool still works and the migration path is documented.