한글 제목 (English Title)

Here's a thinking process:

1. Analyze the Request: - Role: StarCraft: Broodwar custom map reviewer. - Task: Write a review based *only* on the provided map data. - Output Rules: - Output must be JSON, Markdown, tables, lists, or headers - NO.

(Actually, it says "Output must be JSON, Markdown, tables, lists, or headers - NO." - wait, let me reread carefully: "Output must be JSON, Markdown, tables, lists, or headers - NO." - actually the text says: "Here's a thinking process: ... Output must be JSON, Markdown, tables, lists, or headers - NO.

- First write Korean review, 900-1100 characters (including spaces). - Then, after a blank line, translate the same content into English, without any section headers/subtitles like "Korean Review", "StarCraft Broodwar Map Review", etc. - No "Korean Review", "StarCraft Broodwar Map Review", "Map Review", " ", " ", "Translation" as headers.

- Output starts directly with the review body first sentence. - Total output must be 1800-3000 characters (including spaces). - Must naturally map: map structure, genre, gameplay style, player experience, recommended audience, pre-play check items. - Do NOT invent maker, triggers, win conditions, story, balance adjustments.

- Do NOT list map size, resolution, tile units, encoding, and repeat the same explanation. - Map Data Provided: (the data block) - Wait, the "Output must be JSON, Markdown, tables, lists, or headers - NO." seems like it might be a formatting instruction artifact. Let me read the actual prompt structure.

- Actually, the prompt says: "Here's a thinking process: ..." and then the actual user request is embedded. Let me re-read the user message carefully. - User message: "다음 를 자연스러운

로 번역해라. 제목, 머리말, 목록, Markdown, 부연 설명 없이 한국어 번역 본문만 출력해라. 원문에 없는 맵 정보나 평가를 추가하지 마라." - Wait, that's the first sentence? No, the user message actually starts with: "다음 를 자연스러운

로 번역해라. 제목, 머리말, 목록, Markdown, 부연 설명 없이 한국어 번역 본문만 출력해라. 원문에 없는 맵 정보나 평가를 추가하지 마라." - Then it says: "Here's a thinking process: ..." which seems like system/instruction text got pasted. Let me read the entire user message as provided. - Actually, the user message is: "다음 를 자연스러운

로 번역해라. 제목, 머리말, 목록, Markdown, 부연 설명 없이 한국어 번역 본문만 출력해라. 원문에 없는 맵 정보나 평가를 추가하지 마라. Here's a thinking process: ... [long instructions]" - This is confusing. It seems like the user pasted a thinking process template along with their request. The actual request might be embedded in the middle or end. Let me parse it carefully. - The very first line: "다음 를 자연스러운

로 번역해라. 제목, 머리말, 목록, Markdown, 부연 설명 없이 한국어 번역 본문만 출력해라. 원문에 없는 맵 정보나 평가를 추가하지 마라." - Then: "Here's a thinking process:" followed by a lot of instructions about output format, character counts, JSON/Markdown/headers rules, etc. - I think the user wants me to follow the "thinking process" instructions, but the initial line is the actual request?

Or maybe the "thinking process" is part of the prompt engineering and I should ignore the first line? Actually, in many LLM prompts, the "Here's a thinking process:" is part of the system prompt or user instruction. But the first line is in Korean: "다음 를 자연스러운

로 번역해라..." which translates to "Translate the following StarCraft map review into a natural Korean review...". But then it says "Here's a thinking process:" which might be the actual instruction block I need to follow. - Let me re-read the entire user message as given: "다음 를 자연스러운

로 번역해라. 제목, 머리말, 목록, Markdown, 부연 설명 없이 한국어