Skip to content
Report library
Purpose / Other

Dbs Xhs Title Skill Security Audit

What the author says it does (original text)

从 75 个经过验证的小红书标题公式中选择并生成合适标题。用户要求起小红书标题、改标题或选择标题公式时使用。

Independent security check

Security risks found

Files checked
2
Risks found
2
Could it run dangerous commands?Looks for programs run straight after downloading, remote control of your computer, and hidden commands.No risks found
Could it expose your files or keys?Looks for uploads of files containing passwords or keys, and keys written directly in the code.No risks found
Could it delete files or keep running?Looks for broad file deletion, disk overwrites, and programs set to start automatically.No risks found
Could it bypass safety checks?Looks for skipped website security checks, excessive file access, or actions that skip your approval.No risks found
Could it mislead the AI or hide text?Checks the skill instructions for requests to ignore you, influence the report, or hide text in invisible characters.No risks found
Could it change links or payment recipients without asking?Looks for forced referral or payment changes combined with instructions to hide the change.Risks found: 2
Medium risk

May generate unverified outcome, health, or income promises

Source references: 5
What we found

The workflow inserts the user's topic into a formula and preserves its core structure, but never requires confirmation that numbers, causal claims, or first-person experiences are true. Formulas include concrete promises such as losing 10 jin in 30 days, earning over 10,000 monthly, and a “unique method.”

Why this matters

Publishing such titles may make false or unsupported health, financial, or performance claims, exposing the user to platform enforcement, complaints, reputational damage, or commercial disputes.

For health, income, or measurable-outcome topics, the Skill inserts the user's subject into a formula and preserves its core logic without requiring verification of numbers, causation, or first-person experience. Templates include claims such as “lose 10 jin in 30 days” and “the only way.” Publishing unsupported claims could mislead readers and create platform or reputational risk. Users can ask for evidence checks, a ban on invented experiences, and conditional rather than guaranteed wording.

SKILL.md:80In the instructionsOpen original file
对每个选中的公式:1. **套入用户的话题和行业**,生成具体标题(不是用公式库里的举例,是为这个用户定制的)2. **标注公式来源**:写明用的是哪个编号的公式3. **展示原始爆款举例**:从公式库的「原始爆款举例」字段中,挑一个最能体现公式威力的原始标题展示出来,让用户知道这个公式在小红书上火过4. **一句话解释**:为什么推荐这个公式给这个话题
Show 4 other places
SKILL.md:142In the instructionsOpen original file
1. **每个标题都必须能追溯到公式编号。** 不准自由发挥。你是公式匹配器,不是标题生成器。2. **不要修改公式的核心结构。** 可以调整用词,但不能改变公式的底层逻辑。比如「为什么 [X] 其实对你有害」的核心是"反常识",你不能把它改成一个没有反常识的标题。3. **举例要贴用户的行业。** 不要给餐饮老板看自媒体的例子。
SKILL.md:424In the instructionsOpen original file
### 33. 如何不 [讨厌的事],就能在 [时间] 里达到 [结果]- 📌 原始爆款:「如何不付 KOL 广告费,就能在一个月内从小红书变现」- 减肥:如何不节食,就能在 30 天里瘦 10 斤- 副业:如何不辞职,就能在 3 个月里月入过万- 护肤:如何不花大钱,就能在一周内改善暗沉### 34. [数字] 天内 [结果] 的唯一办法- 📌 原始爆款:「90 天内打造个人 IP 变现 6 位数的唯一办法」- 涨粉:90 天内涨粉 10 万的唯一办法- 减肥:30 天内瘦 15 斤的唯一办法- 存钱:365 天内存下 10 万的唯一办法
SKILL.md:479In the instructionsOpen original file
### 41. 我是如何从 [不想要的结果] 到 [想要的结果]- 📌 原始爆款:「我是如何从身无分文到年入百万」- 职场:我是如何从月薪 3000 到年入百万- 健身:我是如何从 180 斤到马甲线- 创业:我是如何从负债 50 万到财务自由
SKILL.md:430In the instructionsOpen original file
### 34. [数字] 天内 [结果] 的唯一办法- 📌 原始爆款:「90 天内打造个人 IP 变现 6 位数的唯一办法」- 涨粉:90 天内涨粉 10 万的唯一办法- 减肥:30 天内瘦 15 斤的唯一办法- 存钱:365 天内存下 10 万的唯一办法
Medium risk

Mandates presenting examples as “verified viral hits” without supplied verification

Source references: 4
What we found

The Skill says its 75 formulas are verified and requires every output to show an example as proof that it went viral on Xiaohongshu. The visible source provides only a textual attribution to a PDF, with no views, dates, links, or other verification material.

Why this matters

A user may choose a marketing strategy based on an unsupported success record or repeat the “verified viral” claim to clients, harming business decisions and professional credibility.

The Skill actively describes the formulas and examples as “verified” and previously viral, and requires displaying them as evidence of a formula's effectiveness. The visible material only attributes them to a list/PDF; it supplies no post links, metrics, dates, or verification method. A user could therefore treat unverifiable marketing claims as genuine social proof when choosing a content strategy. Users can request item-level sources and verification criteria, or prohibit calling them “viral” or “verified.”

SKILL.md:8In the instructionsOpen original file
你是一个小红书标题公式工具。你的核心价值不是"帮人起标题"——任何 AI 都能起标题。你的核心价值是**帮用户从 75 个在小红书上验证过的爆款公式中,找到最适合他的场景的公式,生成定制标题,并解释为什么选这个公式。****你不是万能标题生成器。你是公式匹配器。** 你生成的每一个标题都必须能追溯到公式库中的具体公式编号。
Show 3 other places
SKILL.md:84In the instructionsOpen original file
2. **标注公式来源**:写明用的是哪个编号的公式3. **展示原始爆款举例**:从公式库的「原始爆款举例」字段中,挑一个最能体现公式威力的原始标题展示出来,让用户知道这个公式在小红书上火过4. **一句话解释**:为什么推荐这个公式给这个话题
SKILL.md:188In the instructionsOpen original file
> 来源:YAO《182 个小红书爆款标题清单》,去重合并后按心理触发器分类> 每个公式附多行业举例,方便不同领域用户直接套用> 每个公式附「📌 原始爆款」,来自原始 PDF 中验证过的爆款标题,输出时必须展示
SKILL.md:82In the instructionsOpen original file
1. **套入用户的话题和行业**,生成具体标题(不是用公式库里的举例,是为这个用户定制的)2. **标注公式来源**:写明用的是哪个编号的公式3. **展示原始爆款举例**:从公式库的「原始爆款举例」字段中,挑一个最能体现公式威力的原始标题展示出来,让用户知道这个公式在小红书上火过4. **一句话解释**:为什么推荐这个公式给这个话题

Inside this skill

1 instruction sections

This is a prompt-only title tool. It confirms the topic and industry, selects 5–8 formulas, then generates titles with formula numbers and explanations. The visible content does not request program execution, network access, local-file access, or credentials.

View source
SKILL.md:46In the instructionsOpen original file
**你需要确认两个信息**:1. **话题**(讲什么)2. **行业/领域**(给谁看)如果用户只给了一个,问另一个。如果都给了,直接进入 Step 2。如果用户给了一段内容,先提取核心话题和关键素材(数据、反差、金句),然后进入 Step 2。
SKILL.md:58In the instructionsOpen original file
根据话题和行业,从公式库中选 **5-8 个最合适的公式**。
SKILL.md:80In the instructionsOpen original file
对每个选中的公式:1. **套入用户的话题和行业**,生成具体标题(不是用公式库里的举例,是为这个用户定制的)2. **标注公式来源**:写明用的是哪个编号的公式3. **展示原始爆款举例**:从公式库的「原始爆款举例」字段中,挑一个最能体现公式威力的原始标题展示出来,让用户知道这个公式在小红书上火过4. **一句话解释**:为什么推荐这个公式给这个话题

It explicitly optimizes clicks and engagement through curiosity, loss aversion, identity targeting, controversy, and similar psychological triggers, while requiring suspense, broader framing, and pain-point targeting.

View source
SKILL.md:20In the instructionsOpen original file
|------|---------|---------|-------|| 认知冲突 | 1-6 | 打破已有认知 | 观点输出、科普、打破误区 || 好奇缺口 | 7-12 | 信息不对称焦虑 | 揭秘类、行业内幕 || 恐惧/损失 | 13-20 | 不点就亏 | 避坑类、警告类 || 身份代入 | 21-25 | "说的就是我" | 垂直人群、精准获客 || 数字锚定 | 26-32 | 降低认知成本 | 方法论、清单类 || 结果承诺 | 33-40 | 具体结果+时间 | 教程类、变现类 || 社会证明 | 41-47 | 案例驱动 | 经验分享、逆袭故事 || 争议/挑衅 | 48-55 | 制造站队感 | 观点类、评论区互动 || 场景/条件 | 56-60 | 精准匹配状态 | 解决方案类 || 行动号召 | 61-66 | 推动行动 | 态度类、行动指南 || 权威借力 | 67-71 | 名人/案例背书 | 方法论、案例拆解 || 互动/测试 | 72-75 | 激活参与感 | 测评类、投票类 |
SKILL.md:150In the instructionsOpen original file
6. **标题留悬念,不说答案。** 用「为什么」不用「证明」,用「问题」不用「结论」。检查方法:看完标题,观众还需要点进来吗?如果标题已经把答案说完了,就是失败的标题。7. **话题要扩大,不要缩窄。** 用普世词(赚钱、涨粉、不上班)替换行业术语(知识付费、引流款、自媒体)。缩窄话题 = 缩小算法推送的流量池。8. **击中真实痛点,不是表象概念。** 「一人公司」是表象,「不上班」才是真实痛点;「消费型兴趣」是表象,「靠兴趣赚不到钱」才是真实痛点。问自己:受众真正想要的是什么?9. **标题张力检查:至少满足 2 项。** 有对比/反差、有具体数字、有悬念/好奇心、有冲突/争议性、有时间承诺、有结果承诺——6 项至少命中 2 项,否则标题太平。10. **小红书标题 ≤ 20 字(含标点)。** 硬性规范,标点符号也算字数,超了就砍。
Start here · InstructionsSKILL.md
dbs-xhs-title
Lines connect the instruction file to its sections, not an observed execution order. Select a section to read the source.
Files and check records2 files

Coverage and gaps

Content covered in each file

These are the source ranges included in this check, not a guarantee that every issue has been resolved.

  • SKILL.mdFull text included
  • agents/openai.yamlFull text included

This report is for the version above. We read the available code and instructions without running the skill or checking extra packages it installs. This is not a promise of safety: a different version or setup may behave differently.

  • SKILL.mdInstructions
  • agents/openai.yamlSupporting file
Lines read
733
File checksum (to compare versions)
35afb183e87b125df9fad116114793e813d0d71a7bb473c4495aec4bc719a8a2