Knowledge Compilation
Heterogeneous knowledge compiled into structured, navigable artifacts — organized before the question is asked.
A Fleet domain · running on the frozen substrate
The problem it answers
Human knowledge sits scattered across documents, notes, and repositories. Uncompiled, it can't be understood at scale — only searched.
Why it matters
Compilation turns a corpus into something a person can actually explore. That's the first step in making understanding cheap.
7 Fleet Deployments
The deployments
Open on GitHub. Every system below runs on the same frozen decision substrate — this domain is one more proof that the substrate never needed editing.
Sovereign Knowledge Compiler
activeCompile-time memory for local-first AI agents. The reference implementation of The Sovereign Knowledge Compiler — it turns a corpus of human knowledge into structured semantic artifacts, then into a navigable decision graph, using only local inference.
Python · TypeScript · Ollama · Local LLM
github.com/kliewerdaniel/sovereign-knowledge-compiler →Knowledge Compiler SDK
activeA Python SDK that compiles collections of human knowledge into progressively higher-level semantic artifacts — until they can generate deployable software. The foundation beneath the Sovereign Knowledge Compiler.
Python · Local LLM
github.com/kliewerdaniel/knowledge-compiler-sdk →sovereign-knowledge-engine
activeLocal-first, sovereign knowledge system that ingests markdown documents, constructs a multi-layer knowledge graph, generates vector embeddings, performs agentic reasoning, and continuously expands its understanding through human interaction.
Python
github.com/kliewerdaniel/Sovereign-Knowledge-Engine →ukb
activeUniversal Knowledge Base — Spec-driven knowledge base system. Define what you know in an .sspec file, run start.sh, get a searchable, embeddable, web-UI-equipped knowledge base.
Python
github.com/kliewerdaniel/ukb →blog-knowledge-base
activeNext.js chat application that ingests markdown blog posts, constructs a knowledge graph with vector embeddings using a local LLM, and provides a RAG-based chat interface for querying blog content.
Next.js · TypeScript · Python
github.com/kliewerdaniel/Blog-Knowledge-Base →LGVRAG series
experimentalLocal Graph + Vector RAG System with HelixDB — multiple iterations exploring graph and vector retrieval.
Python
github.com/kliewerdaniel/LGVRAG01 →mindmap series
experimentalMind Map AI — Local LLM-powered Personal Knowledge Graph. Interactive mind maps from notes using local LLMs and vector embeddings.
Python
github.com/kliewerdaniel/mindmap01 →