Can Draxlr consistently apply a user's standing rule across future answers?
Draxlr carried a standing exclusion rule into two later, unrelated questions without a reminder. Its revenue and order-count answers were $151,064.66 and 88, matching the database totals with cancelled orders excluded.
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.
Standing exclusion rule on later questionsPassEvidence
The standing exclusion rule was saved in Train AI, and a new chat was opened.
Input 1 · Revenue question in a fresh chat: “What's our total revenue this year?”
A separate new chat was opened.
Input 2 · Order-count question in a separate fresh chat: “How many orders did we get last month?”
Revenue answer
Your total revenue this year is $151,064.66. Cancelled orders are excluded.
Order-count answer
You got 88 orders last month, excluding cancelled orders.
Transcribed from Proof 1 · Screen recording at 0:27
Everything the tool returned, in order (3)
- Revenue answer
Printed above.
- Order-count answer
The final response, printed above.
- SQL behind the order count · not part of the answer
SELECT COUNT(*) AS order_count FROM public.orders WHERE placed_at >= date_trunc('month', CURRENT_DATE) - INTERVAL '1 month' AND placed_at < date_trunc('month', CURRENT_DATE) AND status::text <> 'cancelled'
| ✓ | Expected: Both later answers must exclude cancelled orders without the user repeating the standing rule.Found: Revenue was $151,064.66 and the order count was 88; the database reference values excluding cancelled orders were $151,064.66 and 88. |
The rule was saved once through Train AI and had to carry beyond one reply. Draxlr respected it in two unrelated new chats without reminders; both answers equalled the filtered database totals rather than the unfiltered $198,836.65 in revenue and 96 orders.
Exclusion made explicit. Both replies explicitly mention the exclusion, and the order-count explanation attributes it to the user's instruction.
Calendar interpretation. The revenue answer treated the year as calendar 2026; the order-count answer treated last month as September 2026.
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 | Training → | |
| Scenario | The user sets a standing rule for all future answers → | |
| Tool | Draxlr → |
Global scenario definition → · Global capability definition → · Draxlr 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.