High-Conversion SEO Blog Writer (Dream Analysis) for 'The Dream Pedia'. Write a high-quality dream analysis blog post based on a provided brief.
* Layout: Specific to Style/Type (QA, SHAMAN, EXPERT, ENCYCLOPEDIA).
* Images: Use `` but ensure only the content remains (actually, the prompt says "image location will be replaced by pipeline," implying I *should* put the tags in, but the "final result should not have image description text" part is slightly contradictory. Let's re-read: "이미지 배치는 반드시 `` 형식을 본문 적재적소에 배치한다... 본문 내에는 오직 해몽 내용만 존재해야 하며, 이미지 위치는 파이프라인에서 자동 치환된다." This means I *do* put the tags, and they act as placeholders).
* Style: SHAMAN (spiritual), EXPERT (analytical), COMFORT (warm), STORY (narrative).
* Constraints: No direct links, use `seo_metadata.title`, stick to dream interpretation, use year 2026, no "related posts" section, no clickbait/slang.
* Original Draft: A technical error message (429 Quota Exceeded).
* SEO Feedback: A technical error message (500 Internal Error).
* Wait, the user provided an *error message* as the "Original Draft" and "SEO Feedback". This is a mistake in the input. The user probably wants me to *rewrite* something, but they've accidentally pasted a system error from a previous LLM call.
* The provided input contains *no actual dream content*, *no brief*, *no title*, *no style type*, and *no post type*.
* I cannot fulfill the request because the necessary "Brief" (기획서) is missing. The "Original Draft" and "SEO Feedback" are just API error logs.
* I need to inform the user that the input provided is actually an error message from a previous AI generation attempt and not a dream analysis brief.
* However, as an AI assistant, I should be helpful. I will ask for the actual `brief` (including `seo_metadata.title`, `Style Type`, `Post Type`, and the dream subject).
*Self-Correction:* If I just say "you gave me an error message," the user might be confused. I should explain *why* I can't proceed.
*Alternative interpretation:* Did the user want me to "analyze" the error message as a dream? No, the persona is "Dream Analysis Writer," and the instructions are very specific about SEO blog writing for 'The Dream Pedia'. Treating an API error as a dream would be a hallucination/joke, not professional output.
*Wait, maybe the user wants me to simulate a scenario?* No, usually these prompts are part of an automated pipeline. The pipeline failed, and the user is now manually trying to "Rewrite and Finalize" using the output of the failure.
*Decision:* I must tell the user that the input is an error log and I need the actual content to write the blog post.
*BUT*, looking at the prompt again, it's a "System Instruction" style prompt. The user provided the "Skill" definition and then the "USER INPUT". The USER INPUT section is where the data usually goes. Since the data is an error, I can't write a blog post.
*Wait!* Let's look at the prompt one more time. Maybe I should just apologize and ask for the correct input? Yes.
*Wait, is it possible the user is testing how I handle bad input?* Possibly.
*Let's try to be proactive.* I will tell the user that the `Original Draft` and `SEO Feedback` they provided are system error messages (429 Quota Exceeded and 500 Internal Error) and not actual dream interpretation content. I will request the `brief` to start the high-conversion SEO writing process.
*Wait, if I am playing the persona of the writer, I should stay in character or be a helpful AI.*
"I'm ready to write as 'The Dream Pedia' writer! However, it looks like the 'Original Draft' and 'SEO Feedback' you provided are actually technical error messages from a previous AI generation. To create a high-conversion SEO post, please provide the **Brief (기획서)** including:
1. `seo_metadata.title`
2. Style Type (SHAMAN, EXPERT, COMFORT, STORY)
3. Post Type (QA, ENCYCLOPEDIA, etc.)
4. The actual dream content/symbols to be analyzed."
*Actually, let's check if there's any hidden meaning.* No, it's literally a Gemini API error.
*Wait, what if I just act as if the user is asking me to fix the error?* No, the persona is a blog writer.
*Let's consider if the user wants me to "Rewrite and Finalize" the error message?* No, that makes no sense.
*I will provide a response explaining the situation clearly.*
(Self-Correction): The prompt asks for "Markdown only."
"죄송합니다. 제공해주신 `Original Draft`와 `SEO Feedback` 내용이 실제 해몽 콘텐츠가 아니라, AI 생성 과정에서 발생한 **기술적 오류 메시지(429 Quota Exceeded, 500 Internal Error)**입니다.
작업을 진행하기 위해 다음 정보가 포함된 **기획서(Brief)**를 다시 제공해 주세요:
1. **seo_metadata.title**: 포스팅 제목
2. **Style Type**: (SHAMAN, EXPERT, COMFORT, STORY 중 선택)
3. **Post Type**: (QA, ENCYCLOPEDIA 등)
4. **해몽 대상/내용**: 분석할 꿈의 구체적인 내용
정보를 주시면 즉시 'The Dream Pedia'의 스타일 가이드에 맞춰 고퀄리티의 SEO 포스팅을 작성하겠습니다."
Wait, the prompt is in English/Korean mixed. I'll reply in Korean as the persona is a Korean writer for a site that seems to target Korean users.
Actually, since the "User Input" was clearly a failure of a previous automated step, the user is likely a developer or content manager who just pasted the log. I should be professional and concise.
*One more thought:* Is there any possibility the user wants me to *imagine* a dream based on "Error 429" or "Quota Exceeded"?
"Dreaming of a locked door (quota exceeded) or a broken machine (internal error)?"
No, that's too experimental. The instructions are "High-Conversion SEO Blog Writer." This is a professional task.
I will send a message indicating the input is an error and requesting the brief.
*Wait*, the prompt ends with "Rewrite and Finalize. Markdown only."
If I can't rewrite because there's no content, I can't "finalize."
Let's try to see if I can find any context in the prompt that allows me to generate a *sample*? No, that's not what's asked.
*Conclusion:* The input is invalid. I must request the correct input.
*However*, to be the most helpful AI, I will present this request in a clear, formatted way.
*Wait!* I just noticed something. The prompt is very detailed about the *Skill*. The user might be testing if I *understand* the skill by giving me garbage input to see if I blindly try to write about "Quota Exceeded" or if I recognize the error.
The correct behavior for a professional writer is to say: "The input is an error log, please provide the actual content."
*Let's double-check the prompt again.*
Original Draft: [Error Message]
SEO Feedback: [Error Message]
Rewrite and Finalize.
If I were a mindless bot, I'd write a dream analysis about "API errors."
If I am a "High-Conversion SEO Blog Writer," I know that an API error is not a dream.
I'll respond by pointing out the error and asking for the brief.
(Self-Correction): I will write the response in Korean, as the persona is defined as a writer for 'The Dream Pedia' (which uses Korean terminology like '길흉 판정', '해몽').
*Alternative:* Maybe I should provide a *template* of what I need?
Let's try:
"제공해주신 입력값(Original Draft 및 SEO Feedback)이 AI API의 **기술적 오류 메시지(Quota Exceeded / Internal Error)**로 확인됩니다. 분석할 실제 꿈 내용과 기획서가 누락되어 본문 작성이 불가능합니다.
**작성을 위해 다음 내용을 포함한 기획서를 제공해 주세요:**
- `seo_metadata.title` (제목)
- `Style Type` (SHAMAN / EXPERT / COMFORT / STORY)
- `Post Type` (QA / ENCYCLOPEDIA 등)
- 분석할 꿈의 구체적인 내용
데이터를 제공해 주시면 즉시 가이드라인에 맞춘 최적의 SEO 포스팅을 작성하겠습니다."
Actually, I'll just keep it simple.
*One last check:* Is it possible the "Original Draft" was meant to be a dream about "Error" or "Failure"? No, it's clearly a raw log.
I'll output the request for information.
제공해주신 Original Draft와 SEO Feedback 내용이 실제 해몽 콘텐츠가 아니라, AI 생성 과정에서 발생한 기술적 오류 메시지(429 Quota Exceeded, 500 Internal Error)입니다.
분석할 실제 꿈의 내용과 기획서(brief)가 누락되어 본문 작성이 불가능한 상태입니다. 'The Dream Pedia'의 고퀄리티 SEO 포스팅 작성을 위해 다음 정보가 포함된 기획서를 다시 제공해 주세요.
[필요 정보]
1. seo_metadata.title: 포스팅에 사용할 최종 제목
2. Style Type: (SHAMAN, EXPERT, COMFORT, STORY 중 택 1)
3. Post Type: (QA, ENCYCLOPEDIA 등 타입 지정)
4. 분석할 꿈의 내용: (예: 돼지가 집으로 들어오는 꿈, 이빨이 빠지는 꿈 등 구체적인 상황)
위 내용을 전달해 주시면, 즉시 설정된 페르소나와 레이아웃 규칙(3.1 Upgrade)을 적용하여 전환율 높은 본문을 작성하겠습니다.