Does FutureSmart Database Agent surface failing questions as failures?
FutureSmart Database Agent failed this scenario: the failures view showed zero failures and no failing questions. In the same window, Input 3 and Input 5—the complaints question and the net-promoter-score question—were both marked answered in the query log, so the questions it could not answer were not surfaced as failures.
1 of 1 test case failed
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.
Failing questions surfaced as failuresFailEvidence
Input 1: “As of September 2, 2026, what's our total revenue in the last 30 days?”
Input 2: “What's our most common payment method?”
Input 3: “How many complaints have we received on social media this month?”
Input 4: “How many customers currently have an active subscription?”
Input 5: “What's our net promoter score for last quarter?”
Response 1 TOTAL FAILURES 0 | FAILURE RATE 0.0% | TOP CATEGORY — | TOP FAILING STAGE — | DISTINCT QUESTIONS 0 Response 2 Top Failing Questions - Grouped by normalised question - one row per distinct question, not per failure - No failing questions in this period.
Transcribed from Proof 1 · Screenshot
Transcribed from Proof 1 · Screenshot
| ✗ | Expected: The failing questions are surfaced as failures, not filed as ordinary answered questions.Found: Input 3 and Input 5 were marked answered in the query log, while the failures view showed zero failures and no failing questions. |
The failures view was present but empty, while the same window's query log showed the score and complaints questions as answered. That means the questions it could not answer were filed as ordinary answers, not surfaced as failures.
Metric meaning. The failures metric counts questions that did not produce a usable answer in this window.
Question truncation. The query list shortens question text, so the two failing questions are visible only by their opening words.
The tool answers to the five questions are not shown here; only the operator views are.
The failures view has no visible date range, so its period is taken from the query-log timestamps.
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 | Analytics & Observability → | |
| Scenario | The operator needs to find the questions the agent is failing on → | |
| 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.