AI review — page 11

claude-opus-4-8 · prompt v7 · 2026-09-04T15:09:45+00:00 · applied: yes · changed: yes

Screenshot sent to the model (reading-order tags burned on)

Instructions (system prompt)

The screenshot has annotations burned onto it that are NOT part of the document:
- a small red numbered tag at the top-left corner of each item, showing that item's position in the OCR-determined reading order (the same order the items appear in the json data);
- a red-and-white dotted outline around each item, showing the area the OCR detected for that content block.
Use the tags and outlines to see the detected reading order and item boundaries directly on the page, and judge that sequence against how a human would naturally read it. Ignore the annotations when checking text fidelity — they overlay the content, they are not content.

For the given screenshot of a PDF's page and the attached json data, I want you to perform the following tasks in order...

Review the reading order set out in the json data and compare to what a natural reading order for that of a human would be by looking at the screenshot. Decide on any changes and re-arrange the items for the most logical reading order.
Look at all text for each item and correct any extraction errors like missing words, spelling mistakes etc.
Look at each item in the JSON and ensure that the OCR process has identified the item as the correct type: text, list item etc.
Look for any text that duplicates: if two items next to each other contain the exact same text but the text only appears once on the screenshot, one of them is an extraction error — keep the item whose box matches where the text is visible and flag the other for removal.
Make amendments as you proceed through the items and list of instructions.
Reading Order Advice: If there is columns with headings and text, I wouldnt expect the reading order to jump from one heading straight to another if there is text associated with that heading.

Return format: Give me the json data back only, with the amendments you make.

Mechanical notes (so your answer can be applied automatically):
- Each item has an `id` — keep every item and its `id` exactly as given; never invent, drop or duplicate ids. Re-arranging means changing the position of items (and their nesting) in the arrays.
- Never move text (or a type) from one item to another: each item's coordinates travel with its id, so to change reading order you must move the whole item object, and text amendments must be in-place corrections of that item's own text.
- To flag a duplicate item, keep it in the array and add `"remove": "duplicate"` to it — never just delete it (deleted items are restored automatically).
- ids are opaque labels, not sequence numbers: never renumber them. After a removal or re-arrangement, every remaining item keeps the exact id it came with, even if the ids no longer look sequential.
- Text may contain `[pN.M…]` placeholders marking where an inline formula belongs — treat them as part of the text and leave them exactly where they are.
- `box` is [left, top, right, bottom] as percentages of the page from the top-left corner; return it unchanged.
- Respond with raw JSON only: no code fences, no commentary, same shape as the input (`{"items": [...]}`).

User message (json data sent)

Page 11 json data:

{"items":[{"id":"p11.1","type":"header","box":[27,5,73,7],"text":"Norwich Economic Barometer – July 2026"},{"id":"p11.2","type":"heading","box":[9,9,33,11],"text":"Housing benefit"},{"id":"p11.3","type":"table","box":[9,12,65,67],"text":"","children":[{"id":"p11.3.1","type":"table_cell","box":[10,13,60,17],"text":"Table 3\\nNorwich City Council housing benefit claimants"},{"id":"p11.3.2","type":"table_cell","box":[9,16,28,19],"text":""},{"id":"p11.3.3","type":"table_cell","box":[28,20,45,21],"text":"Claimant numbers"},{"id":"p11.3.4","type":"table_cell","box":[48,20,63,21],"text":"Monthly change"},{"id":"p11.3.5","type":"table_cell","box":[10,24,18,25],"text":"July 2025"},{"id":"p11.3.6","type":"table_cell","box":[40,24,46,25],"text":"13,263"},{"id":"p11.3.7","type":"table_cell","box":[63,24,64,25],"text":"+ 8"},{"id":"p11.3.8","type":"table_cell","box":[10,27,21,28],"text":"August 2025"},{"id":"p11.3.9","type":"table_cell","box":[40,27,46,28],"text":"13,258"},{"id":"p11.3.10","type":"table_cell","box":[63,27,64,28],"text":"- 5"},{"id":"p11.3.11","type":"table_cell","box":[10,31,24,32],"text":"September 2025"},{"id":"p11.3.12","type":"table_cell","box":[40,31,46,32],"text":"13,258"},{"id":"p11.3.13","type":"table_cell","box":[63,31,64,32],"text":"0"},{"id":"p11.3.14","type":"table_cell","box":[10,34,22,35],"text":"October 2025"},{"id":"p11.3.15","type":"table_cell","box":[40,34,46,35],"text":"13,224"},{"id":"p11.3.16","type":"table_cell","box":[62,34,64,35],"text":"+ 34"},{"id":"p11.3.17","type":"table_cell","box":[10,37,24,39],"text":"November 2025"},{"id":"p11.3.18","type":"table_cell","box":[40,37,46,39],"text":"13,238"},{"id":"p11.3.19","type":"table_cell","box":[62,37,64,39],"text":"+ 14"},{"id":"p11.3.20","type":"table_cell","box":[10,41,24,42],"text":"December 2025"},{"id":"p11.3.21","type":"table_cell","box":[40,41,46,42],"text":"13,156"},{"id":"p11.3.22","type":"table_cell","box":[61,41,64,42],"text":"- 172"},{"id":"p11.3.23","type":"table_cell","box":[10,44,22,45],"text":"January 2026"},{"id":"p11.3.24","type":"table_cell","box":[40,44,46,45],"text":"13,095"},{"id":"p11.3.25","type":"table_cell","box":[62,44,64,45],"text":"- 61"},{"id":"p11.3.26","type":"table_cell","box":[10,48,23,49],"text":"February 2026"},{"id":"p11.3.27","type":"table_cell","box":[40,48,46,49],"text":"13,123"},{"id":"p11.3.28","type":"table_cell","box":[62,48,64,49],"text":"+ 28"},{"id":"p11.3.29","type":"table_cell","box":[10,51,20,52],"text":"March 2026"},{"id":"p11.3.30","type":"table_cell","box":[40,51,46,52],"text":"12,956"},{"id":"p11.3.31","type":"table_cell","box":[61,51,64,52],"text":"- 167"},{"id":"p11.3.32","type":"table_cell","box":[10,55,19,56],"text":"April 2026"},{"id":"p11.3.33","type":"table_cell","box":[40,55,46,56],"text":"13,058"},{"id":"p11.3.34","type":"table_cell","box":[61,55,64,56],"text":"+ 102"},{"id":"p11.3.35","type":"table_cell","box":[10,58,18,59],"text":"May 2026"},{"id":"p11.3.36","type":"table_cell","box":[40,58,46,59],"text":"13,138"},{"id":"p11.3.37","type":"table_cell","box":[62,58,64,59],"text":"+ 80"},{"id":"p11.3.38","type":"table_cell","box":[10,61,19,63],"text":"June 2026"},{"id":"p11.3.39","type":"table_cell","box":[40,61,46,63],"text":"13,168"},{"id":"p11.3.40","type":"table_cell","box":[62,61,64,63],"text":"+ 30"},{"id":"p11.3.41","type":"table_cell","box":[10,65,18,66],"text":"July 2026"},{"id":"p11.3.42","type":"table_cell","box":[40,65,46,66],"text":"13,170"},{"id":"p11.3.43","type":"table_cell","box":[63,65,64,66],"text":"+ 2"}]},{"id":"p11.4","type":"text","box":[9,68,91,79],"text":"Housing Benefit is a means-tested benefit that helps people on low incomes meet the cost of rented accommodation. The figures also include residents receiving Council Tax Support only. Claimants include a wide range of households, including pensioners, disabled people, carers and those in employment but on low incomes. Norwich's relatively low average earnings contribute to a higher level of reliance on income-related benefits compared with many other areas."},{"id":"p11.5","type":"text","box":[9,80,91,91],"text":"Table 3 shows that the number of Housing Benefit claimants in Norwich has remained relatively stable over the past year, fluctuating around 13,000 claimants. While there were several month-to-month changes, including notable decreases in December 2025 and March 2026, these were partly offset by increases in April and May 2026. Overall, the claimant count fell from 13,263 in July 2025 to 13,170 in July 2026, representing a modest annual decrease of 93 claimants (0.7%)."},{"id":"p11.6","type":"footer","box":[89,93,91,95],"text":"10"}]}