Skill UI

This guide details the best practices for packaging, documenting, and releasing research artifacts (code, models, data) for top-tier ML conferences. It covers two stages: preparing a non-identifiable archive for anonymous review, and managing the public release runway, focusing on reproducibility, licensing, and sustained maintenance.

Restructure oversized single-file SKILL.md or scattered project rules into a well-organized Skill directory with progressive rigor, routing, validation, and drift-resistant maintenance for AI agent harnesses.