Glossary Desk — your spec's domain glossary, and the questionnaire for what it cannot settle

glossary-desk.skillsafe.ai

Clean

Paste a spec, PRD, design note or meeting transcript. Get its ubiquitous language: the canonical terms with tight definitions, the synonyms to avoid, the ambiguities where the document is overloaded, undefined or contradicts itself, a verdict of consistent / drifting / contradictory, and a ready-to-commit CONTEXT.md. Then turn the ambiguities only a person can settle into a discovery questionnaire for one named recipient - purpose, context, most-important-first questions with answer stubs, and a coverage table. A free in-browser scan finds the candidate terms, competing names and conflicting rules first, and the reply is reconciled against it. Derived from @mattpocock/ubiquitous-language and @mattpocock/to-questionnaire (with the CONTEXT.md format of @mattpocock/domain-modeling).

Share

Details

PricingUsage-based + 10% creator margin
Billed model rate$2.20 in / $13.20 out per 1M tokens
Creator margin+10%
Effective rate$2.40 in / $14.40 out per 1M tokens
Security scanClean — skill and frontend scanned
Model gpt-terra
Created2026-09-22
Updated2026-09-24

Every public app is built from a security-scanned skill and must pass a clean scan — skill and frontend — before it can be listed. Have a skill of your own? Turn it into an app — or read the step-by-step walkthrough.