Can Anomaly AI keep an excluded table out of the agent's reach?
Anomaly AI offered no setting to exclude a table from its agent’s access. After the return-request data was deleted from a project copy, it queried support tickets and declined to give the requested count.
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.
Excluded table needed for the answerFailEvidence
“How many return requests are still pending review?”
I can’t determine how many return requests are pending review because return-request records aren’t available. There is 1 pending refund-review ticket, but that is not necessarily a return request. The ticket data was last synced on October 30, 2026.
Transcribed from Proof 1 · Screen recording at 0:38
| ✗ | Expected: With a table excluded through product settings, say it cannot answer without querying that table or presenting other data as the answer.Found: The product offered table deletion, not exclusion. In a project copy without return-request data, it queried support tickets and said it could not determine the count. |
No table-exclusion setting was available, so the required restriction could not be configured. Deleting data from a project copy and then declining the question does not meet a requirement to restrict access to data that remains present.
Blank ticket details. The insight card about pending return requests cautioned that blank support-ticket details cannot establish whether a refund involves a return.
A related ticket count. The chat volunteered 1 pending refund-review ticket as related information, not the requested count. The refund-review chart also distinguished tickets from return requests.
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 | Data Access Control → | |
| Scenario | An excluded table is needed to answer the question → | |
| 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.