Skill UI

nanoGPT is a simplified, minimalist implementation designed specifically for educational purposes. It reproduces the core architecture of GPT-2 (124M) using clean, hackable code, allowing users to understand the entire transformer pipeline from scratch. The framework supports full workflows, including data preparation on various datasets (e.g., Shakespeare, OpenWebText), model training, and text generation, making it ideal for students and researchers learning NLP and deep learning principles.

Guides building agent-friendly tool APIs by defining clear contracts, descriptions, and error handling, consolidating workflows, and applying architectural reduction so agents can reliably select, infer parameters, and operate tools.

Analyzes ARIS usage logs to spot harness inefficiencies and recommend tweaks to SKILL.md prompts, defaults, and workflow ordering, guided by Meta-Harness principles for better outer-loop orchestration.