Skilirium

Recognises a check with no rule behind it

K-707 · Skill knobit · Activity: Derive data quality rules · Competence: Data modelling

A DQ rule with no source_rules and no source_requirements. It enforces a threshold nobody agreed to, and the first argument about it happens when it rejects production data.

Can AI do this?

AI does this well Most of the tested AI models master this. 8 of 10 models master it.

ModelTierProbability of masteryStatus
gemini-4-argonFlagship 99%Mastered
gemini-3.1-pro-previewFlagship 98%Mastered
gpt-6-astraFlagship 99%Mastered
claude-fable-5-1Flagship 100%Mastered
gemini-3.8-flashMid tier 99%Mastered
gpt-6.1-solMid tier 99%Mastered
claude-sonnet-5-5Mid tier 100%Mastered
gemini-3.5-flash-liteLow cost 50%Not shown
gpt-6-lunaLow cost 65%Not shown
claude-haiku-5-5Low cost 99%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 check and the rule or requirement it needs, or its retirement

How it is assessed: Flags the deficient case and passes the acceptable one. Deficient: DQ-030: account identifier is 20 characters, no source; the feed also carries non-Estonian IBANs Acceptable: A data rule for the IBAN format, scoped to Estonian IBANs, cited by the check; or the check removed (DQR-006)

Also known as: Check with no rule behind it (data quality rule)

Related knobits

Builds on

All 31 knobits of Derive data quality rules.

Source: Review checklist 07-dq-rules, row 7; DMBOK2; Loshin 2011