LLM Wikis Are Over-Engineered — I Replaced Mine With a Pure Python Compiler

A pure Python compiler can transform a messy folder of local markdown files into a structured, cross-linked wiki, offering a deterministic alternative to LLM-driven agents. This compiler approach avoids the token costs, network latency, and unpredictable outputs of agent systems by treating organization as a parsing problem rather than a reasoning one. The system works in stages, first using regular expressions to extract metadata from inconsistently formatted notes and then building a graph of all connections between them. A critical performance challenge was solved by replacing a slow linking algorithm with an optimized phrase matcher, cutting processing time for thousands of files from over a minute to less than a second.