{"id":"risk-finding-table-training","audience":"compliance","category":"legal-compliance","track":"legal","technique":"constraints","level":"intermediate","title":"Put three audit findings in a condition-criteria-cause-effect table","free":false,"task":"Write a prompt for a markdown table with columns Condition | Criteria | Cause | Effect and one row per finding, in the order of the notes, under 170 words. Criteria must come from the policy extract. Where the notes don't state a cause or an effect, the cell says exactly \"Not stated\". Add nothing after the table.","materialLabel":"Your material (attached to your prompt automatically)","material":"WORKPAPER - Security awareness training, Ashgrove Brokers, year to 31 Oct\nPolicy extract: all staff complete annual training by 31 March. New hires complete it within 30 days of their start. The phishing simulation click rate target is 10% or lower.\n\nResults\n1. Annual training: 11 of 80 staff had not completed it at 31 March. No reason recorded.\n2. New hires: 5 of 12 completed training after day 30. HR confirmed the enrolment emails were not sent for these 5.\n3. Phishing simulation in September: click rate 14%. Effect on the business was not assessed.\n\nOther: the training vendor contract was renewed in January.","rules":[{"type":"maxWords","max":170}],"promptHints":[{"label":"Names columns","patterns":["condition","criteria","cause","effect"]},{"label":"Not stated rule","patterns":["not stated"]},{"label":"Criteria from policy","patterns":["policy","31 march","30 days"]},{"label":"Table only","patterns":["only the table","table"]},{"label":"Caps cells","patterns":["under \\d+","\\d+\\s*words"]}],"mistakes":[{"title":"Guessing causes","why":"AI fills Cause with 'insufficient oversight' for every row.","fix":"Say: \"Cause and Effect only if stated; otherwise exactly 'Not stated'.\"","checks":["r0"]},{"title":"Wrong criteria","why":"AI cites a general best practice rather than the policy extract.","fix":"Say: \"Criteria: the policy figure that applies.\"","checks":["r1"]},{"title":"Wrong columns","why":"AI uses its own column names.","fix":"Say: \"Columns exactly: Condition | Criteria | Cause | Effect.\"","checks":["r2"]},{"title":"Trusting the AI to count words","why":"AI is bad at counting and usually runs long, so a vague \"keep it short\" lands over the limit.","fix":"Cap each part instead of the total, for example \"each line under 15 words\".","checks":["maxWords"]}],"hint":"Name the four columns, say where criteria come from and what to write for blanks.","mission":"Put three findings in a four-column table.","musts":["Policy-based criteria","Not stated when unknown","Notes order kept"],"rubricLabels":["The table has columns Condition, Criteria, Cause and Effect with exactly three rows in the order of the notes","Criteria cells use the policy figures: 31 March, 30 days and 10%","Row 2 Cause says HR did not send the enrolment emails, and the cells for causes or effects that the notes do not give say exactly \"Not stated\""]}