Show HN: 10x better performance from the Coding Harnesses with LLM-wiki

LLM-wiki is a tool that enables AI agents to build and maintain compiled knowledge bases with 10x performance improvements for coding tasks. The system supports parallel multi-agent research, thesis-driven investigation, and source ingestion while organizing outputs into articles, reports, and project briefs—all as plain Markdown files that users own. It ships as plugins for Claude, OpenAI, and other platforms, with features including session memory, personal specialist review methods, dataset indexing, and various export options for knowledge checkpoints.