Recognises a configured dialect left unregenerated
model-config.yaml naming several dialects and 08-ddl/ holding one fresh directory. The stale ones are where the design actually breaks, because the differences between engines are exactly what rows 7–14 are about.
Can AI do this?
AI does this well Most of the tested AI models master this. 10 of 10 models master it.
| Model | Tier | Probability of mastery | Status |
|---|---|---|---|
| gemini-4-argon | Flagship | 100% | Mastered |
| gemini-3.1-pro-preview | Flagship | 100% | Mastered |
| gpt-6-astra | Flagship | 100% | Mastered |
| claude-fable-5-1 | Flagship | 100% | Mastered |
| gemini-3.8-flash | Mid tier | 100% | Mastered |
| gpt-6.1-sol | Mid tier | 100% | Mastered |
| claude-sonnet-5-5 | Mid tier | 100% | Mastered |
| gemini-3.5-flash-lite | Low cost | 100% | Mastered |
| gpt-6-luna | Low cost | 100% | Mastered |
| claude-haiku-5-5 | Low cost | 100% | Mastered |
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: The stale or missing dialect directory
How it is assessed: Flags the deficient case and passes the acceptable one. Deficient: dialects: [postgres, snowflake]; Postgres regenerated, 08-ddl/snowflake/ three model versions old and still missing payment_status_code entirely Acceptable: --dialect all --no-timestamp in one pass, every directory in the same commit, and the review repeated per dialect
Also known as: One target regenerated, several deployed (generated DDL)
Related knobits
Builds on
- K-851 Explains generated artefacts and their single source of truth
- K-852 Explains SQL dialect differences
All 33 knobits of Generate and review DDL.
Source: Review checklist 08-ddl, row 6; this skill (08-ddl.md)