“Copperminds” are folder-based wikis curated by an LLM Archivist. Three workflows drive them: store (ingest + structured write), tap (query against the compiled wiki), and polish (lint + rewrite). Ingest plugins cover plain text, Obsidian vaults, and PDFs with hybrid TOC/LLM/naive chunking. Multiple minds can be queried together and linked bidirectionally — giving you a graph of interconnected knowledge bases from a plain folder tree. Inspired by Karpathy’s wiki system and Cosmere copperminds.
Copper — AI-maintained knowledge bases
FEATURED
An LLM Archivist reads raw sources — notes, PDFs, Obsidian vaults — and compiles a structured markdown wiki. No embeddings, no RAG; just folders, text, and LLM workflows (store / tap / polish). Supports multi-mind queries and bidirectional linking across knowledge bases.
Python 3.12+LLM workflowspdfplumberTyper