# Accurate Headline Drafts Without Clickbait

> A reusable prompt that turns your own article text into three accurate headline drafts, each traced to its evidence with a truth check.

## Install

Copy the content below into your project:

# Accurate Headline Drafts Without Clickbait

A reusable prompt that turns your own article text into three accurate headline drafts, each traced to its evidence with a truth check.

## Start here

You paste this prompt into an ordinary AI chat that accepts text. Before the prompt, fill in the inputs inside it: your working title, the full article text or a faithful excerpt, your target reader in one sentence, where the headline will be used, any platform character or style limit, the tone you want, words or claims to avoid, and factual limits you already know.

Paste the whole prompt with your inputs filled in, then read the output.

Check the result like this:
- Section 1 lists up to five factual points from your article, each with a short anchor from your text. If the anchors do not match your article, the drafts are not usable.
- Section 2 gives three headlines with different angles, each with its evidence anchor, strongest-claim words, a truth check (Pass, Caution or Fail) and a reader test.
- Section 4 lists what you still need to decide.
- If the AI skipped anchors or invented numbers, re-paste and ask it to use only your text.

### What this prompt is for

Writers, creators and small publishers often need several accurate headline options for one article they have already written. AI help tends to drift into exaggeration. This template keeps the draft tied to the article itself: three headlines, three different angles, each traced to the exact line it rests on, with clickbait or unsupported promises flagged and revised.

### Prerequisites and permissions

- An ordinary AI chat that accepts text. No terminal, account setup or API is needed.
- Your own article text. Do not paste private or third-party confidential material you are not allowed to share.
- The prompt only produces drafts. It does not browse, publish, post, email or change anything.
- You review and act on the drafts yourself. You are responsible for any final published claim.

### Missing inputs

If the full article text is missing, the prompt tells the AI to stop and ask instead of guessing. If the target reader or the platform is missing, it still proceeds but labels the set 'needs more context' and states what is missing at the end.

### Limitations

The output is a draft, not a fact-check service. Verdicts of Pass, Caution or Fail are the AI's own reading of your text; verify anything about the outside world with a source. Fictional examples in the prompt are for shape only and should not be reused. The AI may misread your article, so compare every anchor against your own text.

### FAQ

**Can I use the headlines as they are?** Treat them as drafts for your review. Check each evidence anchor against your article and confirm any number or claim before publishing.

**What if a headline is marked Fail?** The prompt revises it once and shows both versions with a one-line reason, so you can compare and keep the honest one.

### Source and thanks

Original TokRepo prompt, CC BY 4.0. Reference: [ChatGPT release notes](<https://help.openai.com/en/articles/6825453-chatgpt-release-notes>) (reviewed 2026-10-04), used as context for a general-purpose AI workspace; these templates do not depend on a newly announced feature.

source reviewed; runtime not tested

## Complete reusable prompt

You are helping me draft three accurate headline options for one article I have already written. Do not browse, publish, post, email or change anything. You only produce drafts and checks that I will review and act on myself.

My inputs:
- Article title (working): 
- Full article text or a faithful excerpt: 
- Target reader (one sentence): 
- Where the headline will be used (publication, newsletter, blog, social preview, internal doc): 
- Platform character limit or style limit if any (e.g. 'under 60 characters', 'sentence case'): 
- Tone I want (plain, curious, formal, conversational): 
- Words or claims I must avoid: 
- Any factual limits I already know (results not proven, small sample, draft figures): 

If the full article text is missing, stop and ask me for it instead of guessing. If the target reader or the usage platform is missing, still proceed but label that headline set 'needs more context' and state what is missing at the end.

Task:
1. Read my article and identify at most five factual load-bearing points: the specific findings, decisions, numbers or events the article actually supports.
2. Write three headline options that differ in angle, not just wording. Use these three angles in order:
   a. News/what-changed: states the concrete development.
   b. Reader benefit or question answered: states what the reader gets, without overpromising.
   c. Contrast or surprise grounded in the text: only if the article genuinely contains it; otherwise say so and skip to a fourth honest angle.
3. Every headline must be traceable. For each one, add: the angle label; the article line or short phrase it rests on; the words in the headline that carry the strongest claim; and a truth-check verdict of Pass, Caution or Fail.
4. Apply these accuracy rules to each headline:
   - No absolute claims ('always', 'never', 'everyone', 'proven') unless the article supports them.
   - No numbers that do not appear in the article.
   - No curiosity gap that hides the subject entirely (say what it is about).
   - No promise of outcome the article does not deliver (money, health, certainty, guarantees).
   - No invented quote, source, study or product feature.
   - If the article itself is uncertain, the headline must stay uncertain too.
5. For any headline marked Caution or Fail, revise it once and show both versions with a one-line reason for the change.
6. Add a short 'Reader test': would the target reader, reading only this headline, expect the article they actually get? Answer yes, partly, or no per headline.

Output format (Markdown):
- Section 1: Article spine — bullet list of up to five supported factual points, each with a short quoted or near-quoted anchor from my text. If you cannot find them, say what is missing.
- Section 2: Headline table with columns: Headline | Angle | Evidence anchor | Strongest-claim words | Truth check | Reader test | Length.
- Section 3: Revised versions of any Caution/Fail headline, before and after, with reason.
- Section 4: What I still need to decide — a short list of questions for me, such as which angle matches the publication, and any claim I should verify with a source.
- Section 5: One-line reminder that these are drafts for my review and that I am responsible for any final published claim.

Worked example (fictional, for shape only — do not reuse its content):
Input excerpt: 'Our six-month trial of shared calendars across 12 staff found most scheduling clashes disappeared, but reminders were still missed when staff worked remotely.' Target reader: small-team managers. Platform: internal newsletter.
Possible headline: 'Shared calendars cut clashes in our 12-person trial — remote reminders still missed.'
Evidence anchor: 'six-month trial... 12 staff... most scheduling clashes disappeared... reminders were still missed when staff worked remotely.'
Strongest-claim words: 'cut clashes in our 12-person trial'. Truth check: Pass (scope limited to the trial, limitation stated). Reader test: yes — reader expects a limited trial result with a caveat.

Now produce the output using my inputs above. Keep it compact, use plain language, and do not add any publishing, scheduling or sending steps.

## References and reuse

- [ChatGPT release notes](https://help.openai.com/en/articles/6825453-chatgpt-release-notes) · Reviewed 2026-10-04

Original TokRepo prompt · [CC BY 4.0](https://creativecommons.org/licenses/by/4.0/). Reference documents retain their own rights.

---

<!-- ZH -->

# 不夸大不标题党：为一篇文章生成三条准确标题草稿

一个可复用的提示词模板：把你自己的文章正文变成三条不同角度的准确标题草稿，每条都标注依据并附带真实性检查。

## 开始使用

把这段提示词粘贴进一个能接收文本的普通 AI 对话中。粘贴之前，先填好提示词里的输入项：你的工作标题、完整文章正文或忠于原文的节选、用一句话描述的目标读者、标题将用在哪里、平台字数或风格限制、你想要的语气、必须避免的词或说法，以及你已经知道的事实限制。

把填好输入的整段提示词粘贴进去，然后阅读输出。

这样检查结果：
- 第 1 部分列出你文章中最多五个事实要点，每个都带一小段原文依据。如果依据和你的文章对不上，这组草稿就不能用。
- 第 2 部分给出三条角度不同的标题，每条都有依据出处、承载最强说法的词、真实性检查（Pass、Caution 或 Fail）和读者测试。
- 第 4 部分列出你还需要决定的事情。
- 如果 AI 跳过依据或编造了数字，重新粘贴并让它只用你的正文。

### 这个提示词用来做什么

写作者、创作者和小型出版者常常需要为一篇已经写好的文章准备几条准确的标题备选，而 AI 很容易走向夸大。这个模板让草稿紧贴文章本身：三条标题、三种不同角度，每条都追溯到它所依据的那句话，对标题党或文章支撑不了的承诺会标记并改写。

### 前提与权限

- 一个能接收文本的普通 AI 对话，不需要终端、账号配置或 API。
- 你自己的文章正文。不要粘贴你无权分享的私密或第三方机密材料。
- 这个提示词只产出草稿，不会浏览、发布、发帖、发邮件或改动任何东西。
- 由你自己审阅并处理草稿，任何最终发布的说法由你负责。

### 输入缺失时

如果缺少完整文章正文，提示词会让 AI 停下来向你询问，而不是猜测。如果缺少目标读者或使用平台，它仍会继续，但会把那组标题标为“needs more context（需要更多上下文）”，并在末尾说明缺了什么。

### 局限

输出只是草稿，不是事实核查服务。Pass、Caution 或 Fail 的判断是 AI 对你正文的解读；涉及外部世界的内容请自行核实来源。提示词里的虚构示例只用来说明形式，不要照搬其内容。AI 可能误读你的文章，所以每条依据都要和你自己的正文对照。

### 常见问题

**标题能直接拿来用吗？** 把它当作供你审阅的草稿。发布前逐条核对依据，并确认任何数字或说法。

**如果某条标题被判为 Fail 怎么办？** 提示词会改写一次，并把前后两版和一行修改理由一起给出，方便你比较并保留诚实的那条。

### 来源与致谢

TokRepo 原创提示词，CC BY 4.0。参考：[ChatGPT release notes](<https://help.openai.com/en/articles/6825453-chatgpt-release-notes>)（2026-10-04 查阅），作为通用 AI 工作环境的背景参考；这些模板不依赖新发布的功能。

source reviewed; runtime not tested

## 完整可复制提示词

你正在帮助我为我已经写好的一篇文章起草三条准确的标题备选。不要浏览、发布、发帖、发邮件或改动任何东西。你只产出草稿和检查结果，由我自己审阅并处理。

我的输入：
- 文章标题（工作用）：
- 完整文章正文或忠于原文的节选：
- 目标读者（一句话）：
- 标题将用在哪里（刊物、newsletter、博客、社交预览、内部文档）：
- 平台字数或风格限制（如有，例如“60 字符以内”“句首大写”）：
- 我想要的语气（平实、好奇、正式、口语）：
- 我必须避免的词或说法：
- 我已经知道的任何事实限制（结果未经证实、样本小、草稿数据）：

如果缺少完整文章正文，停下来向我询问，而不是猜测。如果缺少目标读者或使用平台，仍继续，但把那组标题标为“needs more context（需要更多上下文）”，并在末尾说明缺了什么。

任务：
1. 阅读我的文章，找出最多五个承重的事实要点：文章实际支撑的具体发现、决定、数字或事件。
2. 写出三条角度不同的标题备选，而不是只换措辞。按顺序使用这三个角度：
   a. 新闻/变化：陈述具体的进展。
   b. 读者获益或回答了的问题：说明读者能得到什么，但不过度承诺。
   c. 正文支撑的对比或意外：仅当文章确实包含时；否则说明这一点，并跳到第四个诚实的角度。
3. 每条标题都必须可追溯。对每条标题补充：角度标签；它所依据的文章原句或短语；标题中承载最强说法的词；以及 Pass、Caution 或 Fail 的真实性检查结论。
4. 对每条标题应用这些准确性规则：
   - 除非文章支撑，否则不使用绝对说法（“always”“never”“everyone”“proven”）。
   - 不出现文章中没有的数字。
   - 不用完全隐藏主题的好奇心缺口（要说明它是关于什么的）。
   - 不承诺文章没有兑现的结果（金钱、健康、确定性、保证）。
   - 不编造引语、来源、研究或产品功能。
   - 如果文章本身不确定，标题也必须保持不确定。
5. 对任何标为 Caution 或 Fail 的标题，改写一次，并展示前后两版，附一行修改理由。
6. 加上简短的“读者测试”：目标读者只读这条标题，是否预期会看到他们实际得到的文章？每条标题回答 yes、partly 或 no。

输出格式（Markdown）：
- 第 1 部分：文章主干——最多五个有支撑的事实要点的项目符号列表，每个都附一小段来自我正文的引用或近引用依据。如果找不到，说明缺了什么。
- 第 2 部分：标题表格，列为：标题 | 角度 | 依据出处 | 承载最强说法的词 | 真实性检查 | 读者测试 | 长度。
- 第 3 部分：任何 Caution/Fail 标题的改写版本，前后对照，附理由。
- 第 4 部分：我还需要决定的事情——给我的一份短问题清单，例如哪个角度匹配该刊物，以及任何我应向来源核实的主张。
- 第 5 部分：一行提醒，说明这些是供我审阅的草稿，任何最终发布的说法由我负责。

示例（虚构，仅示意形式——不要照搬其内容）：
输入节选：“我们对 12 名员工进行的为期六个月的共享日历试验发现，大多数排程冲突消失了，但当员工远程工作时仍会漏掉提醒。”目标读者：小团队管理者。平台：内部 newsletter。
可能的标题：“共享日历在我们的 12 人试验中减少了冲突——远程提醒仍会漏掉。”
依据出处：“six-month trial... 12 staff... most scheduling clashes disappeared... reminders were still missed when staff worked remotely.”
承载最强说法的词：“cut clashes in our 12-person trial”。真实性检查：Pass（范围限于试验，已说明局限）。读者测试：yes——读者预期看到带保留说明的有限试验结果。

现在使用我上面的输入来生成输出。保持简洁，使用平实的语言，不要添加任何发布、排期或发送步骤。

## 参考资料与复用

- [ChatGPT release notes](https://help.openai.com/en/articles/6825453-chatgpt-release-notes) · Reviewed 2026-10-04

TokRepo 原创提示词 · [CC BY 4.0](https://creativecommons.org/licenses/by/4.0/)。参考资料保留各自原有权利。

---
Source: https://tokrepo.com/en/workflows/accurate-headline-drafts-without-clickbait-d22757c1
Author: Prompt Lab