Capafy logo Capafy
開始使用
Research-to-Social｜研究轉社群

Research-to-Social｜研究轉社群

by: stockgood3
Turn your research into bilingual social drafts with traceable evidence. Includes a Python source-reference checker and reusable writing workflow. Requires your own AI client for writing; no model credits, web search, or auto-posting included. 將研究整理成中英文社群草稿與來源對照。包含 Python 引用檢查器與寫作流程；寫作使用買家自己的 AI 客戶端，不含模型額度、網路搜尋或自動發布。
#社群媒體
評分
需要更多評價
已售出
0
使用方式
下載

From research to evidence-backed social drafts

從研究資料到有依據的社群草稿

A downloadable Skill and Python checker for creators who already have source material. Your AI client writes the posts; the local checker verifies source IDs and exact quotations.

提供給已有研究素材的創作者：下載 Skill 指引與 Python 檢查器。買家自己的 AI 客戶端撰寫文案；本機程式檢查來源編號與引用文字是否吻合。

clipboard.webp

Watch the walkthrough

觀看操作介紹

Video walkthrough / 操作影片

What you receive

實際交付

  • Bilingual post and 30-second script workflow, claim/source table, and unresolved-check list, completed by your AI client.
  • 由買家 AI 客戶端完成中英文貼文、30 秒腳本、主張來源表與待查證清單的流程。
  • A Python drafting ledger checker, readable source, exact input schema and synthetic examples.
  • Python 寫作依據檢查器、可讀原碼、明確資料格式及合成範例。

Real example: source → check → draft

實際範例：來源 → 檢查 → 草稿

The included synthetic example distinguishes fact C1 from inference C2. The command produces a drafting ledger. examples/finished-copy.md shows a labeled bilingual writing sample; it is not a real customer result.

套件合成範例區分事實 C1 與推論 C2。執行指令產出寫作依據清單；examples/finished-copy.md 提供已標示的中英文文案範例，不是真實客戶成效。

Three steps

三步開始

  1. Unzip and open a terminal in the folder containing SKILL.md. Prepare sources, excerpts and claims using references/input.md.
    解壓後，在含 SKILL.md 的資料夾開啟終端機；依 references/input.md 準備來源、摘錄與主張。
  2. Run the included demo, then replace its JSON with your own structured inputs.
    先執行範例，再換成自己的 JSON 資料。
python3 scripts/check.py --input examples/demo.json
  1. Give your AI client SKILL.md and the authorized research inputs. Receive the bilingual drafts together with their source ledger.
    將 SKILL.md 與授權研究素材交給自己的 AI 客戶端，取得中英文草稿及來源清單。

Requirements and limits

環境與限制

Python 3.10+ recommended; command-line use is required. No GUI. Python checker uses the standard library and makes no model calls. Tested on macOS; clean Windows/Linux installation is unverified. Windows Python Launcher users may replace python3 with py -3. AI-client integrations and writing quality vary; universal compatibility is not promised.

建議 Python 3.10 以上，需要終端機操作，沒有圖形介面。檢查器只使用標準函式庫，不呼叫模型。已於 macOS 測試；Windows／Linux 全新安裝尚未驗證。Windows 可將 python3 換成 py -3。各 AI 客戶端的整合方式與寫作品質不同，不承諾全面相容。

Checks detect invalid source references and nonmatching quotations, not whether a claim is true or supported in meaning. No live search, image/video generation or automatic posting. The final writing step uses your own AI client and its paid quota, if any.

檢查能找出錯誤來源編號或不吻合的引文，不能判定主張真假或語意是否獲得支持。不含即時搜尋、圖片／影片生成或自動發布。最後寫作使用買家自己的 AI 客戶端及其額度。

Price and delivery

費用與交付

US$9 one-time purchase for version 0.1.0 Beta: Skill instructions, readable Python source and examples. No subscription, hosted service, model credits or manual research service is included. Buyer provides their computer and pays for any optional AI/data services.

US$9 一次買斷，交付 0.1.0 Beta 的技能指引、可讀 Python 原碼及範例。不含訂閱、託管服務、模型額度或人工研究；買家自備電腦並負擔選用 AI／資料服務費用。

Support: [email protected] — installation questions, suggestions and reproducible bugs; no guaranteed response time or custom integration. Send a synthetic example and redacted error, never keys or private client documents.

聯絡信箱：[email protected]，受理安裝問題、修改建議與可重現錯誤；不承諾固定回覆時間或客製整合。請附合成範例與去識別錯誤訊息，不要寄送金鑰或客戶私密文件。


View more