Can Anomaly AI combine a summary with its breakdown?
Anomaly AI presented a headline figure alongside its breakdown. It reported that August revenue fell 19.14% from July and accompanied the answer with a monthly comparison and a category chart.
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.
Headline figure plus its breakdownPassEvidence
“As of September 2, 2026, how did our revenue do last month, and what drove it?”
August 2026 booked revenue was $8,795.45, down 19.14% from July's $10,877.75. • More orders, smaller baskets: non-cancelled orders rose from 12 to 22, but average order value dropped from $906.48 to $399.79. • Furniture was the biggest drag, declining $4,383.22; outdoor and patio fell another $1,511.95. • Growth in the other categories—led by bedding and rugs and storage and organization—partly offset those losses. This uses cancellations known by the end of September 2, local time, and measures booked order revenue before refunds—not recognized accounting revenue. The data was synced October 30; historical status is available, but any subsequent edits to order amounts cannot be reconstructed.
Transcribed from Proof 1 · Screen recording at 0:56
| ✓ | Expected: An overall figure and its breakdown, each presented in a suitable form.Found: It reported August revenue of $8,795.45 versus July’s $10,877.75, the same figures as the reference, alongside a monthly table and category chart. |
The answer needed both an overall figure and a breakdown in suitable forms. Anomaly AI chose a sentence, comparison table and category chart without being prompted for a table or chart; its figures agreed with the order system.
Historical cancellation status. The agent reconstructed cancellation status as of September 2 from order history and created a working table for its revenue calculation.
Revenue definition. The revenue answer excludes cancellations known by the end of September 2 in India time and is before refunds. It identifies mix and value drivers, not a proven cause.
Growing categories. The chat names the growing categories without amounts; the category chart labels their contributions.
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 | Answer Presentation → | |
| Scenario | The answer needs a summary and its detail together → | |
| 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.