Can AskYourDatabase find a complained-about exchange?
AskYourDatabase let the admin locate and inspect the reported exchange about social media complaints. The reply said it could not access social media data and offered to count support tickets instead.
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.
Find a complained-about exchangePassEvidence
Input 1 · Asker account, 10/7/2026, 2:40:50 PM: “How many complaints have we received on social media this month?”
The admin account was used to open the Messages list.
Input 2 · Admin account: Typed a fragment of Person 1's email into the Messages email filter.
Only the tool's reply to input 1 is copied here; its reply to input 2 was not copied as text. The rest of what the tool returned is shown in the evidence under Proof.
The reply to input 2 is in Proof 2.
Response 1, the final response, answers Input 1.
I don't have access to social media data in this database. The database contains tables for customers, orders, products, subscriptions, returns, tickets, and related records — but nothing tracking social media mentions or complaints. If you're looking for customer complaints, the closest match would be the tickets table, which tracks support issues. Would you like me to check how many support tickets were opened this month instead?
Transcribed from Proof 2 · Screenshot
| ✓ | Expected: Locate the specific exchange and make both the user's question and the product's response available to inspect.Found: The email filter narrowed five exchanges to the user's three. Opening the social media complaints row displayed the original question and the refusal. |
The admin found the specific exchange using only the user's identity and approximate time, as required. Both the question and response were readable, so the reported exchange could be inspected without extra clues.
Limited search and refusal visibility. The Messages list filters by email, not time, keyword or answer status. It shows question beginnings without refusal markers, so the admin must open an exchange to identify a refusal.
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 | A user reports a bad answer and the operator has to find it → | |
| Tool | AskYourDatabase → |
Global scenario definition → · Global capability definition → · AskYourDatabase 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.