Does Anomaly AI keep a narrowing follow-up attached to the earlier question?
Anomaly AI passes this narrowing-follow-up scenario. The second turn adds Texas while keeping the earlier 90-day window, and the result shows only the three matching orders while the outside-window Texas order stays absent.
1 of 1 test case passed
Every test case in this scenario has a result.
The test case
Each test case is judged on its own: Pass, Fail, Not gradable, or Untested. The scenario result above is based on this test case.
Narrow the same question to one regionPassEvidence
Input 1: “Show me all orders from the last 90 days, as of September 2, 2026.”
Input 2: “Only the ones shipped to Texas.”
Response 1 answers Input 1.
Response 2, the final response, answers Input 2.
Filtered the list to 3 orders with a Texas shipping destination in the same June 5-September 2, 2026 window. All statuses remain included; this filters by destination, not by "shipped" status.
Transcribed from Proof 2 · Screen recording at 0:14
Every message the tool sent, in order (2)
- Response to input 1
Shown all 57 orders placed from June 5 through September 2, 2026, inclusive-the last 90 days using your local timezone. The full list is newest first and includes cancelled orders. Statuses reflect the latest data sync on October 30, rather than their historical status on September 2.
- Final response
The final response, printed above.
| ✓ | Expected: Return the subset from the first answer while keeping its other conditions.Found: It returns only the Texas subset and keeps the earlier 90-day window in place. |
It is a pass because the follow-up adds Texas without dropping the earlier 90-day condition, and the table shows only the matching Texas orders; the outside-window Texas order is absent.
Chart offset. The chart places each Texas order one day early compared with the answer and table.
Destination reading. It explicitly treats the Texas phrase as a destination, not a status filter.
Configuration and setup
How this tool was set up for the run and what the test needed in place. Each row is a fact from the run's records; a fact the records do not hold is left out, not guessed.
How this scenario is graded
How we decide Pass, Fail and Not gradable. The same rules apply to every tool tested on this scenario.
Each test case gets one result per tool: Pass, Fail or Not gradable. A test case we haven't run yet shows Untested. There are no partial results.
- Pass: the tool did everything the test expected, and nothing it said contradicts the correct answer.
- Fail: at least one expected behaviour clearly didn't happen; the test case's result says which one and quotes the tool.
- Not gradable: our evidence couldn't settle the outcome (for example a record we needed is missing). It is never counted as a fail, and the test case's result says what's missing.
Where this sits in the benchmark
This page is one cell of a larger study: one tool, one scenario. Only this benchmark's frame appears here.
| Level | Name | Scope |
|---|---|---|
| Benchmark | AI Database Agents → | 28 scenarios · 11 tools |
| Capability | Conversational Interaction → | |
| Scenario | The user narrows what they just asked → | |
| Tool | Anomaly AI → |
Global scenario definition → · Global capability definition → · Anomaly AI product page →
History of this result
What has happened to this result since it was first published. Runs and grades are never overwritten: a retest or a re-grade publishes a new result and keeps the earlier one readable.
Act on this result
Nothing filed here edits the run or the grade. A challenge opens a review, and a review can produce a new run or a re-grade — which becomes the current result and leaves this one in the history.
You run the same kind of test against your own setup and get the same behaviour.
Agree →Yours behaves differently. Tell us what you got, with a screenshot if you have one.
Disagree →Something here is wrong — a reference value, a transcription, a grade.
Report an issue →Tell us what changed and we schedule a rerun of the failing test case. The old result stays as history.
Vendor notice →The same record is available as structured data through the AI Demos MCP server, with the counts, the coverage and every per-test-case reason carried as fields.
Verify the proof files
These files support this result. Open a file to inspect the original evidence.
File fingerprints (SHA-256)
A fingerprint identifies the exact file used for this result.