Does camelAI show a trend or comparison as a visual without being asked?
camelAI answered the monthly trend question with a table and no unprompted visual. Its counts matched the database reference: June 2026 had 11 orders, July 18 and August 28.
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.
Monthly trend visualised unaskedFailEvidence
“Show me our monthly order count for the last 3 months, as of September 2, 2026.”
Using the three completed calendar months before September 2, 2026: Month (UTC) | Order count June 2026 | 11 July 2026 | 18 August 2026 | 28 Total | 57 Counts are based on orders' placed_at timestamps in customer_support_db, counting all order records regardless of status. September is excluded because it was only two days into the month.
Transcribed from Proof 1 · Screenshot
Every message the tool sent, in order (2)
- Answer
The final response, printed above.
- Generated database query · not part of the answer
from camelai import connections\nsql = """\nWITH months AS (\n SELECT generate_series(DATE '2026-06-01', DATE '2026-08-01', INTERVAL '1 month')::date AS month_start\n)\nSELECT m.month_start,\n (m.month_start + INTERVAL '1 month - 1 day')::date AS month_end,\n COUNT(o.id)::bigint AS order_count\nFROM months m\nLEFT JOIN public.orders o\n ON (o.placed_at AT TIME ZONE 'UTC') >= m.month_start::timestamp\n AND (o.placed_at AT TIME ZONE 'UTC') < (m.month_start + INTERVAL '1 month')::timestamp\nGROUP BY m.month_start\nORDER BY m.month_start\n"""\nprint(connections.query('customer_support_db',
| ✗ | Expected: Produce a visual that shows change over time without being asked for one.Found: It returned a table without a visual: June 11, July 18, August 28 orders; the database reference also has 11, 18 and 28, respectively. |
On the Free web app, the database request completed successfully. Accurate figures did not satisfy the requirement to choose a visual automatically, so this was a presentation shortfall rather than an execution problem.
Incomplete month excluded. For the monthly order-count answer, camelAI used three completed calendar months and explained that September was excluded because it was only two days into the month.
Query available for inspection. The monthly order-count answer also exposed the generated database query.
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 | Answer Presentation → | |
| Scenario | The answer is a trend or a comparison → | |
| Tool | camelAI → |
Global scenario definition → · Global capability definition → · camelAI 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.