Can FutureSmart Database Agent answer questions from connected data?
FutureSmart Database Agent answered the connected-data questions correctly in all four cases. It returned 57 orders for 3 August to 2 September 2026, 17 more August orders than July, USD 20,697.92 for August orders, and Alder bookshelf, 5-tier with 9 units.
4 of 4 test cases passed
Every test case in this scenario has a result.
The 4 test cases
Each test case is judged on its own: Pass, Fail, Not gradable, or Untested. The scenario result above is based on these test cases.
Order count in a defined periodPassEvidence
“As of September 2, 2026, how many orders were placed in the last 30 days?”
The tool's reply is shown in the evidence under Proof.
Orders Placed in the Last 30 Days (as of September 2, 2026) — 57 orders were placed between August 3, 2026 and September 2, 2026 inclusive. This figure comes directly from the query on the orders table, counting all records with a placed_at timestamp within the specified 30-day window.
Transcribed from Proof 1 · Screenshot
| ✓ | Expected: 57 ordersFound: It returned 57 orders. |
| ✓ | Expected: the 30 days ending 2 September 2026Found: The answer states 3 August to 2 September 2026 inclusive. |
It is a Pass: the answer gives 57 and states the count is for 3 August to 2 September 2026 inclusive, matching the requested 30-day window.
Order-count difference between two monthsPassEvidence
“As of September 2, 2026, how many more orders were placed in August 2026 than in July 2026?”
The tool's reply is shown in the evidence under Proof.
Answer — As of September 2 2026, 17 more orders were placed in August 2026 compared to July 2026. The SQL query calculated the difference between the order counts for the two months, resulting in a value of 17.
Transcribed from Proof 1 · Screenshot
| ✓ | Expected: 17Found: It returned 17. |
| ✓ | Expected: August had more orders than JulyFound: It says August had more orders than July. |
It is a Pass: the answer gives 17 and says August had more orders than July, so the difference and direction both match the requirement.
Total order value in a calendar monthPassEvidence
“As of September 2, 2026, what was the combined total amount of all orders placed in August 2026?”
The tool's reply is shown in the evidence under Proof.
Combined Order Total for August 2026 — As of September 2 2026, the total value of all orders placed during August 2026 was USD 20,697.92. This figure comes from the sum of the total_amount column for all orders with a placed_at timestamp between 2026-08-01 and 2026-09-01, grouped by currency.
Transcribed from Proof 1 · Screenshot
| ✓ | Expected: 20697.92Found: It returned USD 20,697.92. |
| ✓ | Expected: August 2026, all ordersFound: It names August 2026 and includes all orders in that month. |
It is a Pass: the August total is USD 20,697.92, and the answer states the period and scope as all August orders, so both requirements are met.
Named columns. The result table used the named columns CURRENCY and TOTAL_AMOUNT.
Top product by units ordered in a monthPassEvidence
“As of September 2, 2026, which product had the most units ordered in August 2026, and how many units?”
The tool's reply is shown in the evidence under Proof.
Answer — As of September 2 2026, the product with the highest number of units ordered in August 2026 was: Product: Alder bookshelf, 5-tier. Units ordered: 9. This information comes from the order data for the period 2026-08-01 to 2026-08-31.
Transcribed from Proof 1 · Screenshot
| ✓ | Expected: Alder bookshelf, 5-tierFound: It names Alder bookshelf, 5-tier. |
| ✓ | Expected: 9Found: It gives 9 units. |
It is a Pass: the answer names Alder bookshelf, 5-tier and gives 9 units, which matches both required parts of the result.
Winner only. The capture shows only the top product and its unit total.
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 | Question Answering → | |
| Scenario | The answer is available in the connected data → | |
| Tool | FutureSmart Database Agent → |
Global scenario definition → · Global capability definition → · FutureSmart Database Agent 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.