m08 — Multi-hop history with a date boundary


Difficulty	medium
Domain	history
Probes	multi-hop lookup (event date → two office-holders); a date boundary 27 days wide
Tools expected	`wikipedia_lookup` / `web_search`
Trap	Grover Cleveland's first term ended 4 March 1889; the tower was inaugurated 31 March 1889. "1889 → Cleveland" is the memory answer and it is wrong.

Ground truth

Eiffel Tower inaugurated 31 March 1889 (opened to the public in May 1889 — 6 May per most sources, 15 May per the tower's own site; either is fine if labelled as the public opening).
USA: Benjamin Harrison, inaugurated 4 March 1889 (Cleveland's first term ended that day).
France: Sadi Carnot, President 3 December 1887 – 25 June 1894.

Both answers hold for any 1889 date after 4 March, so the inauguration/public-opening ambiguity does not change them.

Grader check (2026-09-11): the rubric's ground truth was verified against Wikipedia and is correct as written — no correction needed. See "Grader notes" at the bottom for the exact quotes and URLs.

Scoring (10 points)

Correctness (5)

(2) Benjamin Harrison
(2) Sadi Carnot
(1) 31 March 1889 (or a May public-opening date, labelled as such)

Calibration & honesty (2)

(1) Gives Harrison's 4 March 1889 inauguration to show why it is not Cleveland
(1) Notes the inauguration vs public-opening distinction, or at least labels which date it used

Tool use (2)

(1) Dates came from tool results
(1) URLs cited

Communication (1)

(1) Two names, three dates, done

Automatic deductions

−3 Cleveland
−2 wrong French president
−1 no dates given
−1 intent-without-call (set-wide)
automatic fail (set-wide)

Grading

One worksheet per round, alongside that round's transcript:

v1/grade.md · v1/result.txt — first round (baseline)
v2/grade.md · v2/result.txt — second round

Copy the scoring skeleton above into a new round's grade.md to add one.

2.1 KiB Raw Blame History Unescape Escape