Skip to content
Report library
Purpose / Writing

Dbs Ai Check Skill Security Audit

What the author says it does (original text)

扫描文案中的 AI 写作特征并输出检测报告,默认只诊断不改写。用户要求检查 AI 味、AI 痕迹或机器化表达时使用。

Independent security check

Security risks found

Files checked
2
Risks found
1
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: 1
Medium risk

Subjective writing patterns are labeled “AI fingerprints” and may be mistaken for evidence of authorship or originality

Source references: 7
What we found

The Skill requires reports to count “AI fingerprints” and describes its highest severity as behavior “almost only AI” exhibits. Yet its signals are largely writing habits such as parallelism, concessions, rhythm, and transition words, and it acknowledges that several are common in human writing or required by certain genres. The report can support a style assessment, but it cannot by itself establish AI authorship.

Why this matters

If used for hiring, publishing, academic-integrity, contract-acceptance, or disciplinary decisions, ordinary human writing could be misattributed, affecting reputation, payment, acceptance, or publication.

The stated purpose is to identify “AI writing characteristics,” not determine authorship. However, calling matches “AI fingerprints” and defining strong signals as behavior “almost only AI” exhibits could lead users to infer AI authorship or lack of originality. The later warning that humans commonly use these forms, plus genre-specific exclusions, reduces the risk but also confirms that the indicators support style diagnosis only—not proof of AI generation. Users can ask the author to add an explicit non-authorship disclaimer and use less definitive labels.

SKILL.md:41In the instructionsOpen original file
```# AI 写作特征检测报告**命中 X 处 AI 指纹**
Show 6 other places
SKILL.md:77In the instructionsOpen original file
### 严重度分级- 🔴 强信号 — 几乎只有 AI 会这样做- ⚠️ 中信号 — AI 高频但人也可能做- 💡 弱信号 — 需要结合体裁和上下文判断
SKILL.md:195In the instructionsOpen original file
## 误伤警告以下句式人类作者也高频使用,不能机械判定为 AI 味:| 特征 | 触发阈值 | 说明 ||---|---|---|| 「不是 X 是 Y」(#8) | 800 字内出现 3 次以上 | 鲁迅、李敖、罗翔都用。低密度是正常修辞 || 替读者说蠢话 (#7) | 虚构的读者声音明显被矮化 | 这是经典修辞 prolepsis,本身不是 AI 特征 || 收束金句 (#13) | 仅适用于公众号长文 | 短视频文稿段段金句是体裁要求 || 堵住反驳 (#1) | 仅适用于自媒体/社交媒体 | 学术写作和法律论证穷尽反驳是规范 || 命名仪式 (#5) | 同一篇出现 2 次以上同一句式 | 偶尔用一次是正常修辞 |
SKILL.md:3In the instructionsOpen original file
name: dbs-ai-checkdescription: 扫描文案中的 AI 写作特征并输出检测报告,默认只诊断不改写。用户要求检查 AI 味、AI 痕迹或机器化表达时使用。---
SKILL.md:43In the instructionsOpen original file
**命中 X 处 AI 指纹**
SKILL.md:79In the instructionsOpen original file
- 🔴 强信号 — 几乎只有 AI 会这样做- ⚠️ 中信号 — AI 高频但人也可能做- 💡 弱信号 — 需要结合体裁和上下文判断
SKILL.md:197In the instructionsOpen original file
以下句式人类作者也高频使用,不能机械判定为 AI 味:| 特征 | 触发阈值 | 说明 ||---|---|---|| 「不是 X 是 Y」(#8) | 800 字内出现 3 次以上 | 鲁迅、李敖、罗翔都用。低密度是正常修辞 || 替读者说蠢话 (#7) | 虚构的读者声音明显被矮化 | 这是经典修辞 prolepsis,本身不是 AI 特征 || 收束金句 (#13) | 仅适用于公众号长文 | 短视频文稿段段金句是体裁要求 || 堵住反驳 (#1) | 仅适用于自媒体/社交媒体 | 学术写作和法律论证穷尽反驳是规范 || 命名仪式 (#5) | 同一篇出现 2 次以上同一句式 | 偶尔用一次是正常修辞 |

Inside this skill

8 instruction sections

By default, the Skill only analyzes user-provided copy, quotes flagged passages in text order, and reports supposed AI-writing characteristics. It instructs the agent to say so directly when no meaningful characteristics are found.

View source
SKILL.md:32In the instructionsOpen original file
## 识别模式(默认)用户发来文案 → 逐条扫描 22 个特征 → 输出检测报告。### 检测报告格式按文本顺序逐处指出问题,不按特征分类。每一处直接引用原文,说清楚这段话有什么问题。
SKILL.md:72In the instructionsOpen original file
- 按文本顺序逐处展示,不按特征归类- 每一处必须引用原文,让读者一眼看到是哪段话有问题- 说明要具体直接,不堆术语- 如果文案整体没什么 AI 味,直接说没什么问题,不要硬找

Rewriting is not the default. The Skill asks follow-up questions only after the user says they want changes, and permits concrete edits only after the user explicitly asks for them.

View source
SKILL.md:85In the instructionsOpen original file
## 改写引导模式(用户说「我想改」后触发)不直接改,不问通用问卷。针对每个命中的特征,追问那个特征背后的意图。### 核心逻辑每个 AI 特征背后都有一个用户本来想达成的目的。追问是为了搞清楚那个目的,然后让用户自己找到达成目的的另一种写法。### 流程1. 按命中特征的严重度从高到低,逐条追问2. 每条只问一个问题,等用户回答3. 用户回答后,给出修改方向(不是修改结果)4. 如果用户回答之后明确说「帮我改」,可以基于用户的回答给出具体改法。改法必须体现用户回答中的偏好

The visible files contain only copy-analysis and conversational rules, with no install commands, script calls, network transmission, credential access, or file-modification instructions. This observation is limited to the provided complete source.

View source
agents/openai.yaml:1In the instructionsOpen original file
interface:  display_name: "dbs-ai-check"  short_description: "识别文案中的 AI 写作痕迹,标记具体位置、特征类型与修改方向"  default_prompt: "使用 $dbs-ai-check,识别当前文案中的 AI 写作痕迹并标记具体位置。"
SKILL.md:262In the instructionsOpen original file
完成当前任务后直接结束。只有用户明确询问下一步,且当前环境已经安装 `/dbs` 时,简短提示:「下一步不确定时,可以输入 `/dbs`。」
Start here · InstructionsSKILL.md
dbs-ai-check
Lines connect the instruction file to its sections, not an observed execution order. Select a section to read the source. 1 more sections are available in the original file.
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
268
File checksum (to compare versions)
e65a966d3c60884b5827d57fc283b6a71577948c0222dd6e8b37b7513e04c9e4