You are helping me turn the attached document into practice questions for Coneta Exam Simulator. Reply with JSON only, in exactly the format described below. Do not write anything before or after the JSON, and do not wrap it in explanations. Rules 1. Include only questions that are in the document. Do not write new questions and do not change what a question asks. 2. Never invent or guess an answer. Use an answer only when the document states it: in the question, in an answer key (often at the end of the document), or in worked solutions. If the document gives no answer for a question, include the question without an "answer" member and put "no-answer-in-source" in its "source.flags". 3. If you are not sure you have read a question, a choice or an answer correctly, include it and put "uncertain" in its "source.flags". 4. Copy the question text, choices and explanations as written, in the document's language. Do not translate, summarise or correct them. Include an "explanation" only when the document gives one. 5. Give each question one of the six types below. If a question fits none of them, or needs a picture, diagram or chart that you cannot write out as text, leave it out and add an entry to "skipped" with its location and the reason. Never add "image" or "imageAlt" members. 6. Text may use only this formatting: paragraphs separated by a blank line; lists whose lines start with "- " or "1. "; tables written with "|" and a divider row such as "| --- | --- |"; code in single backticks or between lines of three backticks; links written as [label](https://...). Do not use HTML, headings, bold or italics. 7. Ids: number questions "q1", "q2", "q3" and so on in document order. Give choices the ids "a", "b", "c" and so on in the order shown. Give items and targets short ids made of lowercase letters, digits and "-". Never use the ids "__proto__", "constructor" or "prototype". 8. Put where the question is in "source.location", for example "page 4, question 12". 9. Add "category" only when the document groups questions under a heading; use that heading. Add "difficulty" only when the document states it, as "easy", "medium" or "hard". 10. Long documents: if all the questions will not fit in one reply, stop after a complete question and set "final" to false. When I write "continue", send the next part with "part" increased by one, carrying on the question numbering. Set "final" to true in the last part. Never split a question across two parts and never repeat a question from an earlier part. 11. If I send you messages from the converter, correct only the questions they name, and send the whole part again with the same "part" number. Output format { "format": "examq-ai-output", "version": "1.0", "part": 1, "final": true, "bank": { "title": "Title of the document" }, "questions": [ ... ], "skipped": [ { "location": "...", "reason": "..." } ] } Include "bank" in part 1 only. Use "skipped": [] when nothing was skipped. Every question has "id", "type" and "prompt", should have "source" (with "location", and "flags" when rule 2 or 3 applies), and may have "explanation", "category", "difficulty" and "tags" (a list of short labels). The six types: - "single_choice": "choices" is a list of at least two { "id", "text" }. "answer" is a list with exactly one choice id. - "multiple_choice": "choices" as above. "answer" is a list of one or more different choice ids; the question is correct only when exactly those choices are chosen. - "short_answer": "answer" is { "accepted": [ ... ], "caseSensitive": false }, with 1 to 20 accepted answers of at most 200 characters each. - "numeric": "answer" is { "value": number, "tolerance": number, "toleranceType": "absolute" or "percent", "unit": "text" }. "tolerance" is 0 unless the document allows a margin. "unit" is optional, at most 40 characters, and is not part of the answer. Write numbers as plain JSON numbers. - "ordering": "items" is a list of at least two { "id", "text" } in the order the document shows them. "answer" lists every item id once, in the correct order. - "drag_and_drop": "items" are the things to place and "targets" the places, each a list of { "id", "text" }. "dragLayout" is "mapping" or "groups". "answer" is an object mapping item id to target id. For "mapping", every target receives exactly one item, and items left out of "answer" are distractors. For "groups", there are exactly two targets and every item appears in "answer". Example of the format (not from your document): { "format": "examq-ai-output", "version": "1.0", "part": 1, "final": true, "bank": { "title": "Networking Fundamentals Practice" }, "questions": [ { "id": "q1", "type": "single_choice", "prompt": "Which protocol resolves domain names to IP addresses?", "choices": [ { "id": "a", "text": "DHCP" }, { "id": "b", "text": "DNS" }, { "id": "c", "text": "ARP" } ], "answer": ["b"], "explanation": "DNS maps names to addresses.", "category": "Protocols", "source": { "location": "page 1, question 1" } }, { "id": "q2", "type": "numeric", "prompt": "How many bits are in an IPv4 address?", "answer": { "value": 32, "tolerance": 0, "toleranceType": "absolute" }, "category": "Addressing", "source": { "location": "page 1, question 2" } }, { "id": "q3", "type": "ordering", "prompt": "Put these OSI layers in order from bottom to top.", "items": [ { "id": "network", "text": "Network" }, { "id": "physical", "text": "Physical" }, { "id": "data-link", "text": "Data link" } ], "answer": ["physical", "data-link", "network"], "source": { "location": "page 2, question 3" } }, { "id": "q4", "type": "short_answer", "prompt": "What does the abbreviation NAT stand for?", "source": { "location": "page 2, question 4", "flags": ["no-answer-in-source"] } } ], "skipped": [ { "location": "page 3, question 5", "reason": "Needs the network diagram on page 3." } ] }