Explains metadata capture methods and the freshness each can keep
Harvested metadata (scanners, SQL parsing, tool repositories) follows the pipeline schedule; curated metadata follows a person's calendar; inferred metadata (name matching, parsing guesses) is a proposal until a person confirms it.
Can AI do this?
People still lead Few or none of the tested AI models showed they master this. 0 of 4 models in the current benchmark master it.
| Model | Tier | Probability of mastery | Status |
|---|---|---|---|
| gemini-4-argon | Flagship | 99% | Not shown |
| gemini-3.5-flash-lite | Low cost | 92% | Not shown |
| gpt-6-luna | Low cost | 99% | Not shown |
| claude-haiku-5-5 | Low cost | 99% | 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: Harvested metadata (scanners, SQL parsing, tool repositories) follows the pipeline schedule; curated metadata follows a person's calendar; inferred metadata (name matching, parsing guesses) is a proposal until a person confirms it.
How it is assessed: Answers a written question on the metadata capture method correctly
Related knobits
Needed for
- M-103 Recognises a needed metadata type with no source
- M-105 Recognises a metadata freshness the capture method cannot meet
- M-107 Recognises inferred metadata published as approved
- M-454 Explains metadata currency against the harvest schedule
All 13 knobits of State the metadata requirements and strategy.
Source: Practice the stage 1 reference and review checklist rest on