AIWeave
A specification and skill methodology for Go backend microservices that keeps code and documentation mechanically in sync, so an AI agent can rebuild an entire project from its documentation and skill files alone.
AIWeave is a methodology rather than a runnable tool: it defines a directory structure, documentation content rules, and step-by-step skill procedures for Go backend services built on Gin, GORM, cobra, and Sarama, aimed at both human architects and AI coding agents such as Claude Code. Its central claim is that a project's docs folder is a complete blueprint and its Claude skills folder is the AI's operating manual, precise enough on their own that an agent can rebuild the entire codebase from nothing but those two things.
It separates three orthogonal concerns: a design specification for how architectural decisions get made, a documentation specification for how those decisions get written down precisely enough to regenerate code, and a skill specification for how an AI executes against the documentation. The methodology supports both a one-time bootstrap mode for a new or heavily refactored project and a continuous incremental-sync mode that updates all affected documentation whenever a requirement is delivered.