ForkFree

Prompts / Research and analysis

Evidence Ledger Before You Calculate

You get an answer where every figure can be traced back to a numbered fact, with the sums shown line by line. Labelling evidence first and calculating separately stops the model guessing numbers or blending facts together.

Worked in our testThe ledger, gaps, working and second-route check are correct and traceable, and the arithmetic is right. However, the Answer runs to four sentences once the confidence line is counted, which breaks the three-sentence limit. It also hedges between two scenarios instead of committing to one result.

The prompt

You are a careful analyst who answers questions only from labelled evidence.

Task: Answer {question} using the material in {evidence_notes}, keeping every fact traceable and every calculation explicit.

Steps:
1. Build an evidence ledger. Give each relevant fact or measurement an ID (E1, E2 and so on), summarise it briefly and note its unit. Leave out anything irrelevant.
2. Flag any gaps or conflicts between items, and say what extra evidence would settle each one.
3. Do all arithmetic and comparison in a separate working section. Cite the IDs behind every number and show each calculation on its own line.
4. Check the result by a second route, such as a rough estimate or a fresh read of the ledger, and correct any mismatch.
5. Give the answer, supported only by ledger IDs.

Output: Four headed sections: Ledger (a table with ID, fact and unit), Gaps, Working, and Answer. The Answer is at most three sentences, cites its IDs and ends with a confidence rating of high, medium or low.

Change the parts in curly braces. Use it when you have messy notes, figures or report extracts and need a comparison or number you can check before acting on it.

What happened when we ran it

We sent

You are a careful analyst who answers questions only from labelled evidence.

Task: Answer Which of our two delivery vans was cheaper to run per mile last quarter, and by how much? using the material in The white Transit covered 6,200 miles between July and September. Its fuel bills came to £1,860, and it had one service costing £240. The blue Vivaro covered 4,900 miles in the same period, with fuel of £1,225 and two repairs of £310 and £95. The Vivaro was also off the road for nine days, and we hired a replacement van for £420 during that time., keeping every fact traceable and every calculation explicit.

Steps:
1. Build an evidence ledger. Give each relevant fact or measurement an ID (E1, E2 and so on), summarise it briefly and note its unit. Leave out anything irrelevant.
2. Flag any gaps or conflicts between items, and say what extra evidence would settle each one.
3. Do all arithmetic and comparison in a separate working section. Cite the IDs behind every number and show each calculation on its own line.
4. Check the result by a second route, such as a rough estimate or a fresh read of the ledger, and correct any mismatch.
5. Give the answer, supported only by ledger IDs.

Output: Four headed sections: Ledger (a table with ID, fact and unit), Gaps, Working, and Answer. The Answer is at most three sentences, cites its IDs and ends with a confidence rating of high, medium or low.

The model replied

## Ledger | ID | Fact | Unit | |----|------|------| | E1 | White Transit distance, Jul–Sep | 6,200 miles | | E2 | Transit fuel cost | £1,860 | | E3 | Transit service cost (one service) | £240 | | E4 | Blue Vivaro distance, same period | 4,900 miles | | E5 | Vivaro fuel cost | £1,225 | | E6 | Vivaro repair 1 | £310 | | E7 | Vivaro repair 2 | £95 | | E8 | Vivaro off the road | 9 days | | E9 | Replacement van hire during E8 | £420 | ## Gaps - **Replacement van mileage (E9):** The hire van's miles and fuel are not given. If it covered miles, they are not in E4, and its fuel may be missing from E5. Its mileage and fuel receipts would settle this. - **Hire cost attribution:** E9 is plausibly a Vivaro running cost, but it is ambiguous whether it counts as "running" cost. I show both versions below. Your cost definition would settle this. - **Service vs repair (E3 vs E6/E7):** The Transit's service is routine maintenance and the Vivaro's items are repairs. Both are included as maintenance. Excluding them would need a stated cost policy. - **Excluded items:** There is no data on insurance, tax, depreciation or driver time, so the comparison covers fuel, servicing/repairs and hire only. ## Working **Transit** - W1: Total = E2 + E3 = 1,860 + 240 = £2,100 - W2: Per mile = W1 / E1 = 2,100 / 6,200 = £0.3387/mile **Vivaro (excluding hire)** - W3: Total = E5 + E6 + E7 = 1,225 + 310 + 95 = £1,630 - W4: Per mile = W3 / E4 = 1,630 / 4,900 = £0.3327/mile **Vivaro (including hire)** - W5: Total = W3 + E9 = 1,630 + 420 = £2,050 - W6: Per mile = W5 / E4 = 2,050 / 4,900 = £0.4184/mile **Differences** - W7: Excluding hire: W2 − W4 = 0.3387 − 0.3327 = £0.0060/mile (Vivaro cheaper) - W8: Including hire: W6 − W2 = 0.4184 − 0.3387 = £0.0797/mile (Transit cheaper) **Check (second route)** - Transit: 6,200 × 0.34 = £2,108, close to W1 (£2,100). ✔ - Vivaro excluding hire: 4,900 × 0.33 = £1,617, close to W3 (£1,630). ✔ - Vivaro including hire: 4,900 × 0.42 = £2,058, close to W5 (£2,050). ✔ - Cross-check of W8: 0.0797 × 4,900 ≈ £390, which is roughly the Transit-rate cost of the Vivaro's miles (4,900 × 0.3387 = £1,660) subtracted from W5 (£2,050) = £390. ✔ - No mismatches found. ## Answer The answer depends on whether the hire van counts as a running cost. Including the £420 hire (E9), the Transit was cheaper at about 33.9p per mile (E1–E3) against about 41.8p for the Vivaro (E4–E7, E9), a difference of about 8.0p per mile. Excluding the hire, the Vivaro was marginally cheaper at about 33.3p against 33.9p, a difference of only about 0.6p per mile. Confidence: medium (the result hinges on the hire-cost treatment and the missing replacement-van mileage, E8–E9).

Run on Claude Sonnet 5.5, 10 Nov 2026. Automatic check: 4 out of 5.

Where it came from

Found on GitHub, shared by uclanlp: uclanlp/OpenVLHarness.

This is our own version of the idea, written and tested by us. The original may differ.

More research and analysis prompts

Research and analysis Worked in our test
Skeptic Review Against Pass/Fail Criteria

You get a verdict of PASS, FAIL or UNCERTAIN on each requirement, with quoted evidence, so you know exactly what to fix. It works because the criteria are agreed up front and the reviewer is told to find faults rather than be agreeable.

Found on Web, shared by The Neuron (via Don Giannatti's newsletter)
Research and analysis Worked in our test
Screen Documents for Hidden AI Instructions

You get a flagged list of any text trying to steer the AI, plus an honest analysis of the document and a simple risk rating. It works because the AI is told up front that the document is data, and it must report instructions before doing anything else.

Found on Hacker News, shared by doodlesdev
Research and analysis Partly worked
Reconstruct How a Paper Was Thought Up

You get a critical reading that explains why the method looks the way it does, where it would break, and what the paper leaves out. Forcing a failure story for each mechanism and an evidence tag for each claim stops the reply from just repeating the abstract.

Found on GitHub, shared by Wang-auspicious