A model-invoked engineering skill that actively builds and sharpens the project’s domain model and ubiquitous language — challenging terms against the glossary, stress-testing with edge-case scenarios, and updating the GLOSSARY.md shared-language glossary (artifact-domain-model) and ADRs (artifact-adr) inline. A multi-context repo indexes several glossaries through GLOSSARY-MAP.md. It triggers on “discussing codebase terminology, writing or editing a GLOSSARY.md, or recording or editing an ADR”.

Directly attacks failure mode #2 (verbosity): a precise shared vocabulary lets the agent say more with fewer words, name code consistently, and navigate the codebase faster (pattern-context-engineering). The same vocabulary feeds mp-wait-what and mp-pr, which both write in the GLOSSARY.md language.

See Also

Version history

v1.3.0 (2026-10-04)

  • The domain-doc convention was renamed from CONTEXT.md / CONTEXT-MAP.md to GLOSSARY.md / GLOSSARY-MAP.md everywhere the skills read and write it. Existing files must be git mv’d, since the skills only look for the new names.
  • The trigger was broadened from “pin down domain terminology or a ubiquitous language / record an architectural decision” to discussing codebase terminology or editing a GLOSSARY.md or ADR directly.

v1.2.0 (2026-08-05)

  • The deprecated ubiquitous-language skill (which wrote UBIQUITOUS_LANGUAGE.md) was deleted from the repo; this skill is its replacement.

v1.1.0 (2026-07-08; ingested 2026-07-09)

  • Absorbed the deprecated ubiquitous-language DDD-glossary skill: building and hardening the ubiquitous language became part of domain-modeling itself.

Initial ingest (2026-06-28)

  • Paged as the skill that sharpens the CONTEXT.md domain model and ADRs.