技能 thought-mining

📦

thought-mining

Name: thought-mining
Author: yunshu0909

內容修訂 r1 中風險 ⚙️ 外部命令

將原始想法轉化為可發布的草稿

零散的想法很難轉化為聚焦的文章。此技能會引導一場對話，用來捕捉洞見、選擇主題、驗證事實、改善文筆，並審閱最終草稿。

支援: Claude Codex Code(CC)

📊 69 尚可

Maintain this Skill? Claim it 📊 69 尚可

使用我的智慧代理安裝

將此請求複製給你的智慧代理，其中包含權威 Skill 頁面和清單。

智能體請求

Review the Skillstore skill "thought-mining" from https://skillstore.io/skills/yunshu0909-thought-mining.md and its manifest at https://skillstore.io/api/skills/yunshu0909-thought-mining/manifest. Verify the artifact. You may proceed after verification, subject to the environment's own policy.

你的智慧代理仍應顯示其計畫，並依安全政策請求必要確認。

Agent 可讀資源

當 AI Agent、爬蟲或腳本需要乾淨脈絡、而不是讀取完整頁面時，請使用這些連結。

Markdown 詳情 GET /skills/yunshu0909-thought-mining.md 已簽名 manifest GET /api/skills/yunshu0909-thought-mining/manifest 已簽名 lockfile GET /api/skills/yunshu0909-thought-mining/lockfile

測試它

正在使用「thought-mining」。 I have many notes about AI workflows but no clear article angle.

預期結果:

從對話中捕捉到的一組具名洞見。
文章最有力想法的候選清單。
一個可供使用者確認或修改的核心觀點提案。

正在使用「thought-mining」。 Here is my draft. I think the logic may be weak in the middle.

預期結果:

逐節邏輯審閱。
針對不清楚句子的具體措辭改善建議。
關於應發布或修改的最終建議。

安全審計

中風險

v4 • 2026/7/8 開啟版本化報告

The static command findings are false positives caused by markdown inline-code markers and a fenced file tree. The entropy findings are false positives from readable Chinese markdown prose, not binary or encrypted content. A separate filesystem safety issue remains because the workflow creates local files from user-provided topic names.

已掃描檔案

560

分析行數

複核項目

已忽略的誤報

已確認的安全問題 (2)

中

stages/01-mining.md:19-20 stages/02-topic.md:31-33

Unconstrained Local File Creation

The workflow tells the agent to create insight and writing-record files using the user-provided topic. Without filename sanitization or confirmation, a topic containing path separators could write to an unintended local path.

The file creation instructions are explicit and derive filenames from the topic placeholder. The risk depends on agent behavior, so confidence is high but not absolute.

低

stages/03-validation.md:13-16

External Search May Expose Sensitive Topics

The validation stage instructs the agent to search using the topic and core concepts. It does not require user confirmation before sending potentially private writing ideas to external search tools.

The network-search instruction is clear, but the data exposure depends on the user topic and available tools. This is a privacy risk rather than evidence of malicious exfiltration.

風險因素

⚙️ 外部命令 (6)

SKILL.md:22 SKILL.md:23 SKILL.md:24 SKILL.md:25 SKILL.md:26 SKILL.md:48-62

審計單位: codex 檢視審計紀錄 →

分享與引用此報告

分享版本化評估報告、中性徽章、嵌入卡片與引用。Skillstore 只報告證據，不判斷該 Skill 是否安全。

開啟版本化報告

複製報告連結

https://skillstore.io/skills/yunshu0909-thought-mining/audits/4?utm_source=security_passport&utm_medium=share&utm_campaign=versioned_report

Markdown 徽章

[![Skillstore security assessment](https://skillstore.io/badges/skills/yunshu0909-thought-mining/security.svg)](https://skillstore.io/skills/yunshu0909-thought-mining?utm_source=security_passport_badge)

HTML 徽章

<a href="https://skillstore.io/skills/yunshu0909-thought-mining?utm_source=security_passport_badge"><img src="https://skillstore.io/badges/skills/yunshu0909-thought-mining/security.svg" alt="Skillstore security assessment" loading="lazy"></a>

嵌入卡片

<iframe src="https://skillstore.io/embed/skills/yunshu0909-thought-mining.html" title="Skillstore Security Assessment" sandbox="allow-popups allow-popups-to-escape-sandbox" loading="lazy" referrerpolicy="no-referrer" width="420" height="180"></iframe>

學術引用 (APA · BibTeX · CFF)

APA 引用

yunshu0909. (2026). thought-mining security audit report (audit version 4) [Author version unspecified]. Skillstore. https://skillstore.io/skills/yunshu0909-thought-mining/audits/4

BibTeX 引用

@techreport{yunshu0909-yunshu0909-thought-mining-2026,
  author = {yunshu0909},
  title = {thought-mining security audit report (audit version 4)},
  institution = {Skillstore},
  year = {2026},
  number = {4},
  url = {https://skillstore.io/skills/yunshu0909-thought-mining/audits/4},
  note = {Author version unspecified}
}

CITATION.cff

cff-version: 1.2.0
message: "If you use this Skill, cite its author and this versioned security audit report."
title: "thought-mining security audit report (audit version 4)"
version: "unspecified"
type: report
authors:
  - name: "yunshu0909"
date-released: "2026-07-08"
url: "https://skillstore.io/skills/yunshu0909-thought-mining/audits/4"
identifiers:
  - type: other
    value: "skillstore:yunshu0909-thought-mining:audit:4"
    description: "Skillstore immutable audit report identifier"

Skillstore 評分

評分依據證據置信度: 高

架構

可維護性

內容

社群

規範符合度

你可以打造什麼

發展文章主題

透過引導式提問，將粗略觀察轉化為清楚的論點與文章方向。

梳理產品思考

捕捉產品洞見、比較可能角度，並為讀者選擇最有價值的訊息。

審閱技術草稿

檢查草稿是否符合先前洞見、驗證主張，並具備清楚的最終訊息。

試試這些提示

開始捕捉想法

Use the thought-mining workflow. I want to write about [topic]. Ask me questions first and record my scattered ideas as insights.

找出核心觀點

Review the insights we have collected. Help me identify the strongest 3 to 5 points and ask me the one-sentence question to find my core viewpoint.

起草前驗證主張

Move to the validation stage. Search authoritative sources for the key claims in my topic, then tell me what is accurate, uncertain, or missing.

執行最終發布審閱

Use the final review checklist on this draft. Check topic clarity, logic, factual accuracy, strong lines, ending strength, and give exact revision suggestions.