Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →stevesolun/domain-modeling
Clarify domain concepts, language, boundaries, and durable architectural decisions.
Passed 18 of 20 test cases with the skill, and 17 of 20 without it.
2 fixed19 passed either way1 still failing1 broke
Tokens, turns and time describe resource use, not a dollar price.
A test case the skill fixed. Same model, same prompt, with and without the skill.
We are building a dispatch module for a logistics platform and noticed confusing overlaps between 'Trip', 'Leg', 'Route', and 'Assignment' across our specifications and automated test suites. Please evaluate these entities using concrete operational scenarios, call out where codebase behavior departs from the intended domain model, and present a thorough domain summary. Include an explicit list of lingering ambiguities that need team resolution along with any suggested updates to context maps or ADRs.
| Checked | Without skill | With skill |
|---|---|---|
| The response explicitly lists lingering domain ambiguities. | Missed | Passed |
| The response explicitly details proposed artifact changes. | Missed | Passed |
The instructions your model reads when the skill loads.
Build a precise shared understanding of the concepts that matter to the current decision. Read existing glossaries, context maps, ADRs, code, and tests when they provide relevant evidence.
Ask the user to resolve a term only when their intent cannot be inferred safely and the distinction affects the outcome.
Update a glossary or context map when the user requests it or when a resolved term is durable, project-specific, and useful beyond the current conversation. Follow an existing repository format first; otherwise use the context format as a lightweight starting point.
Record an ADR only when a decision is costly to reverse, surprising without its context, and based on a meaningful tradeoff. Follow existing ADR conventions or use the ADR format. Do not create artifacts merely to complete the workflow.
Keep domain definitions separate from implementation plans. Report unresolved ambiguity and artifact changes explicitly.
test cases passed, vs 17 / 20 without
Published skills whose descriptions closely match this one, with their headline benchmark lift.
More skills from the same source and nearby in the registry, with their headline benchmark lift.