Governed-knowledge retrieval + injection for LLM agents -- concepts, playbook entries, three-scope shadow merge, and before_model middleware
Token-count, cost, and vocabulary diffing for LLM tokenizer changes
The whole local AI stack in one executable: run and manage local AI models across every GPU you have, and search your files, code, and the web pages you crawl, with answers that cite the source. Per-project libraries, semantic and hybrid search, vision OCR, a…
Give AI agents the right context, not your entire filesystem. A local, governed, metadata-only catalogue of the documents you already have — built for MCP.
Type-safe LLM-powered functions. Write intelligent functions with decorators, not agents.
