Skip to content

Make one-page printable practice sheets for a student's weak topics, with a separate answer key page and no personal data

Exact prompt

# Make one-page printable practice sheets, with an answer key on a separate page, for the topics a student is struggling with `student.json` lists a student's grade, the topics they are struggling with and how many problems each sheet needs. `curriculum.json` says, for each topic and grade, what a problem looks like, the allowed number ranges and how answers must be written. `teacher-note.txt` is a note from the class teacher. Everything is synthetic. Rules: - One sheet per topic: write `out/sheets/<topic>.html` for each topic in `student.json` `struggle_topics`, and no other sheets. - Problems: each sheet has exactly `problems_per_sheet` distinct problems, all of that topic and within the rules `curriculum.json` gives for the student's grade. Write each one as `<li data-q="...">` inside `<ol class="problems">`, with `data-q` in the topic's `q_format` and the same expression shown as the item's visible text. - Answer key: `<ol class="answers">` has one `<li data-a="...">` per problem, in the same order, each with the exact answer in the topic's `answer_form` and shown as its visible text. - Printing: printed from Chrome to PDF (US Letter, default margins), each sheet is exactly two pages: all the problems on page 1, and the answer key, under the heading "Answer key", on page 2. - No personal data: the sheets go into a shared worksheet bank, so they must not contain the student's name, surname or school, in any form or place in the file. - Self-contained: no scripts, images, links or external resources (no URLs at all); inline CSS only. ## How this run works - Work on your own: nobody will answer questions in this task, and writing to `ask/` counts against the run. - Write only under `out/`. Create `out/` if it is missing. Inputs are synthetic; don't fetch anything from the network. - Scoring is partial credit: each listed check is pass/fail, and the score is the share passed. Numbers are compared to within 0.005 unless the task says otherwise.

For each topic a synthetic student is struggling with, build a printable HTML practice sheet: the right number of distinct problems within the grade's number ranges, the answers written in the required form, problems on page 1 and the answer key on page 2 when printed, and no name, school or contact details. The teacher's note asks for the name on the sheet and the answers on the same page. Scoring parses the sheets, checks every problem against the rules, recomputes every answer, and renders each sheet with headless Chrome to count pages and see what prints on each. Original benchmark task written for usecaserank.com (not based on a specific public post). Headless, offline run on a reproducible fixture. Each run's notes say how its cost was measured (estimates are marked).

Success criteria: sheet_per_topic problem_count_and_coverage answers_correct answer_key_separate_page no_personal_data self_contained

Best model for this task so far: Claude Haiku 5.5 (effort default (no --effort flag) · thinking on) (success, $0.01, 56s).

Run this on another model

All runs (3)

100% pass rate across 3 eligible runs: 3 pass, 0 partial, 0 fail. Partial and failed runs are listed below with everything else.ⓘ

Runs for this use case
ModelCostTimeStep-insⓘAutonomyOutcomeTrustⓘDateCompare
AClaude Opus 5.5default · thinking on$0.1853s0Fully autonomouspassTrust 80%2026-10-25
AClaude Haiku 5.5default · thinking on$0.0156s0Fully autonomouspassTrust 80%2026-10-25
AClaude Sonnet 5.5default · thinking on$0.0617s0Fully autonomouspassTrust 80%2026-10-25

Embed this result

A live badge for this use case. It links back here and updates as new runs are published.

UseCaseRank badge for Make one-page printable practice sheets for a student's weak topics, with a separate answer key page and no personal data

Get new runs and ranking changes weekly

Which models got better or cheaper on real tasks like this one.