Traces a logical-model anomaly to its structural cause
Links an insertion, update or deletion anomaly to the defect that produces it, such as a repeating group or a transitive dependency, so the repair removes the cause.
Can AI do this?
People still lead Few or none of the tested AI models showed they master this. 0 of 10 models master it.
| Model | Tier | Probability of mastery | Status |
|---|---|---|---|
| gemini-4-argon | Flagship | 45% | Not shown |
| gemini-3.1-pro-preview | Flagship | 43% | Not shown |
| gpt-6-astra | Flagship | 44% | Not shown |
| claude-fable-5-1 | Flagship | 44% | Not shown |
| gemini-3.8-flash | Mid tier | 42% | Not shown |
| gpt-6.1-sol | Mid tier | 45% | Not shown |
| claude-sonnet-5-5 | Mid tier | 42% | Not shown |
| gemini-3.5-flash-lite | Low cost | 49% | Not shown |
| gpt-6-luna | Low cost | 42% | Not shown |
| claude-haiku-5-5 | Low cost | 43% | Not shown |
Mastered means a probability of at least 90% that the model has this knobit, estimated from its graded answers. “Not shown” means the answers do not settle it either way. How the models are tested.
What it looks like in practice
Expected result: Each anomaly paired with the dependency, group or missing entity that causes it
How it is assessed: Names, for each reported anomaly, the structural cause it comes from, and fixes the cause rather than the symptom
Related knobits
Builds on
Usually followed by
- K-533 Decomposes an entity along a functional dependency without loss
- K-535 Replaces a repeating group with a dependent entity
All 56 knobits of Derive the logical data model.
Source: Stage 5 reference (references/05-logical.md) and the skill-tests logical-model task