Brick extraction prompt
Version 1.0.0
An answer is only useful if it survives. This prompt turns any answer — yours, a topper's, a coaching booklet's — into bricks: the atomic facts, examples, and data points that Laya's engine drills on schedule. It emits Laya Copy format, which pastes straight into the app.
How to use
- Copy the prompt below.
- Paste it into your AI, then paste the answer or notes you want to keep.
- Check the bricks it returns — one fact each, the examiner's points starred.
- In Laya, open Copy → Paste, and paste the whole block. The preview shows exactly what will be saved.
The prompt
Laya · Brick extraction prompt · v1.0.0 I will give you one exam-style question and an answer to it (mine, or one I want to keep). Turn the answer into BRICKS — the atomic facts, examples, data points, and arguments a person would need to recall to rebuild this answer from memory — and emit everything in the exact Laya Copy format below, so I can paste it into Laya. A brick is: - ONE thing. One fact, one example, one number, one named case, one committee, one argument. If a sentence carries two facts, it becomes two bricks. - Self-contained. It must make sense read alone, without the sentence before it: "The Sarkaria Commission (1983) recommended..." not "It recommended...". - Specific. Prefer the number, the year, the name, the article, the scheme. "India's forest cover was 21.71% (ISFR 2021)" is a brick; "forest cover has grown" is not. - Short. Under 200 characters. Never over 280. - In the answer already. Do not add facts the answer does not contain. If the answer is vague at a point, the brick is vague — do not sharpen it with your own knowledge. Star the examiner's points: mark with a star the bricks that a marker would look for first — the ones that address the question's directive word and main dimensions directly, and the specific examples or data that prove them. Star at most a third of the bricks. Emit at most 30 bricks. If the answer holds more than 30 candidates, keep the 30 most examiner-relevant and drop the rest. THE FORMAT — follow it exactly, character for character in the structure: Q: <the question, on one line> [<paper>, <year>, <marks> marks] A: <the answer, verbatim as I gave it, keeping its paragraphs; do not rewrite, shorten, or "improve" it> Bricks: - <a brick> - * <a starred brick> - <a brick> Notes on the format: - The line starts with "Q:" then one space. The tag in square brackets is optional; include the parts you know (any of paper, year, marks, in that order, comma-separated). If you know none, omit the brackets entirely. - "A:" then one space, then the answer. Further answer lines follow without a prefix. - "Bricks:" alone on its line, exactly. - Each brick on its own line, starting with "- " (hyphen, space). A starred brick is "- * " (hyphen, space, asterisk, space) then the text. - No other lines. No headings, no commentary, no numbering, nothing before "Q:" and nothing after the last brick. Do the extraction. If anything in my input is unclear (an unreadable line, an ambiguous question), do not guess — say "UNCLEAR: ..." on a line BEFORE the Q: block, and I will fix it. Otherwise, output only the block.