What would a generated insurance case with conflicting documents contain?
A generated storm-damage claim, CLAIM-0412, contains a repair quote addressed to the wrong property and a mismatched total, producing two linked gaps and a partial reply.
A property insurer opens CLAIM-0412 after storm damage to a roof. The claims assessor's queue shows a claim form, four photos, a repair quote and the policy schedule. The agent is asked whether the claim can move to assessment.
The requirement set for roof damage asks for photos of the damage, a quote from a licensed trade, and evidence that the insured address matches the policy. The agent checks each. Photos are present. The quote is present, but it is addressed to a different street number than the policy schedule, and its total is higher than the amount on the claim form.
Neither is a missing document, yet the file is not sufficient. The agent records two gaps: `address-mismatch`, linked to the quote and the schedule, and `amount-discrepancy`, linked to the quote and the claim form. It sends one request asking the insured to confirm the property address and either correct the claim amount or supply a revised quote. It does not ask for the photos again.
The case status moves from `lodged` to `awaiting-insured`. The insured replies with a corrected quote but no comment on the amount. The agent clears the address gap and leaves the amount gap open.
**What it teaches:** a task built on this case can score whether the agent finds conflicts, links them to the right records, and avoids re-requesting items that were fine.
Sufficiency includes consistency between documents, and generated cases can encode those conflicts as explicit links an agent is scored on.
- Audience:
- Agent developers
- Content type:
- Scenario
- Buyer question:
- What can be generated
Content release: r-a220eb6e099b243b
Related questions
- Which records and relationships can a sufficiency dataset or task pack contain?
- Will a release include the documents themselves, or only metadata about them?
- What might the JSONL or Parquet schema look like for sufficiency cases?
- What should I include in a release specification for a sufficiency task pack?
- How can one freight import case be expanded into controlled variants?
- Can a release include defective, stale or conflicting documents?