Sharing and propogating team knowledge and practices through LLM and MCP automation.

A Request for Comments (RFC) detailing a comprehensive specification for implementing OAuth 2.1 authorization within the Model Context Protocol (MCP) when using Server-Sent Events (SSE) as the transport mechanism, ensuring vendor-neutral security for AI-to-tool communication.

Explores how the MCP knowledge base design approximates the effect of learning by manipulating context fed into a frozen LLM, sidestepping catastrophic forgetting by not updating LLM parameters, and leveraging a database as an evolving external knowledge store.

Illustrates the data ingestion and query flows for the Brainery system, detailing how data from sources like Discord, Memo Blog, and GitHub is processed via MCP and stored in TimescaleDB, and how it's queried by an LLM-powered chatbot.

Details the TimescaleDB-based schema for an append-only, migration-less system designed to capture observational data and facilitate the emergence of coined terms (like 'Vibe Coding') via LLM pattern detection, focusing on the observation_log hypertable and continuous aggregates.