I built a small project that solves a recurring pain point in LLM prompting: LLMs are great at generating text, but terrible at consistent formatting unless you explicitly spell out the structure every time. After seeing many teams repeatedly write long instructions like “Give me a JSON array with X fields and no prose,” I created a lightweight alternative.

AI Output Format Catalog is a collection of 116 standardized output-format tags (e.g., JSNARR, CMPTBL, FLWCHT, EMLFMT). Each tag maps to a specific predictable structure — JSON, tables, diagrams, code formats, email templates, etc.