Jlc Support logo

Jlc Support

Community
John-owo
jlc-support

與 JLCPCB(及類似 PCB/PCBA 代工廠)的線上客服、工程師、業務窗口溝通。用於催辦審查或 DFM、就生產稿與貼裝確認向工廠提出技術澄清(比對與判讀交給 jlc-production-review)、算確認稿到期時間、訂單被取消後的追問、報價與優惠申請、付款後請對方負責到底、退款追問;先核對網站與信箱,再起草可獨立閱讀、含界線句的訊息,寄出與任何確認都等使用者同意。Use when drafting or sending messages to JLCPCB support or engineers, chasing a DFM or review, handling a cancelled order, asking the factory about production file changes, or asking for a quote or coupon. Triggers: "JLC 客服", "催 DFM", "寫信給 JLC", "訂單被取消", "問工廠生產稿", "確認稿期限", "JLCPCB support", "chase DFM", "production file question".

Overview

PublisherJohn-owo
Repositorypcb-agent-workflow
Skill namejlc-support
Stars
72
Forks
10
Bundled files
6
LicenseMIT
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • 6 bundled files

    Scripts, templates, and references the model can read while it works. Files are read-only and never executed.

  • Open source

    Published by John-owo on GitHub. Read the source before you install it.

Installation

Install the Jlc Support AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
    https://github.com/John-owo/pcb-agent-workflow/tree/main/skills/jlc-support
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/John-owo/pcb-agent-workflow.git /tmp/pcb-agent-workflow
mkdir -p .claude/skills
cp -r /tmp/pcb-agent-workflow/skills/jlc-support .claude/skills/jlc-support
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Jlc Support in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Jlc Support on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Jlc Support is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

JLC 客服溝通

目標有兩個:每則訊息都要換到能驗證的答覆;任何回覆都不能被當成你批准了什麼。

硬規則

以下動作每一次都要使用者在聊天中明確同意,不能從先前的同意類推:

  • 寄出信件或送出聊天訊息(草稿可以先寫)
  • 從網站下載工廠的生產稿或其他檔案
  • 上傳檔案、接受條款、提交訂單(包含「只送審不付款」)
  • 接受加費、補償或優惠方案
  • 確認生產稿(Confirm Production)或貼裝確認(Confirm Parts Placement)
  • 修改設計或製造檔

唯一的例外:使用者可以明確說出範圍和期限,事先授權「催辦」與「純技術澄清」兩種訊息,例如「到這張訂單出貨前,帶界線句的催辦可以直接送」。

  • 例外要寫進專案的 WORKLOG。
  • 例外內的每則訊息都要帶界線句,寄出後記錄時間與內容。
  • 例外永遠不包含上面清單的其他動作:確認、費用、補償、訂單、上傳、下載、設計修改。
  • 範圍不清楚或期限已過,就回到逐次確認。

付款、身分證號、清關資料由使用者本人在網頁填寫。agent 不讀取、不記錄、不轉述這些內容。

跟訂單有關的信,一律從訂單註冊信箱寄出,並回在原信串。寄信工具如果連到別的帳號,先停下來告訴使用者。

寫訊息前先核對

工具與退路見下方「工具搭配」。

  1. 打開網站訂單頁,看詳細狀態,不要只看頁首的徽章。頁首寫 In Production 時,詳細狀態可能還是 Data Preparation。
  2. 搜尋註冊信箱,垃圾信也要看;再讀上一次的客服對話。
  3. 記下所有編號(購物車儲存後可能改號,新舊編號都要記)、狀態、核對時間與時區。
  4. 客服說「信已寄出」或「已處理」時,先查信箱再回覆。這類說法曾被同一家公司的另一位客服推翻。

訊息結構

每則訊息依序包含:

  1. 主旨:<要求的動作> — <訂單編號>
  2. 問候
  3. 一行背景:編號、剛核對到的狀態、時間與時區
  4. 一句影響(例如假期將至、期限倒數)
  5. 編號問題,能寫成二選一就寫成二選一
  6. 退路:拿不到完成時間時,請對方給負責人和下次更新時間(含時區)
  7. 界線句,例如:This is not approval of DFM/production, design changes or extra fees. Please keep manual confirmation.
  8. 致謝與署名

寫作規則:

  • 第一則訊息要能獨立看懂。 線上客服會輪替,接手的人只看得到這一則。
  • 跨日寫絕對日期。 過了午夜才寄,就把「今天」改成「9 月 28 日」。
  • 一封信只談一件事。 退款、合規這類問題混在長信末段,連問三次都沒人回答。這類問題要開獨立信串;信尾有 Escalate Case 連結時,可以請使用者決定是否使用。
  • 大部分問題答完後,只追問剩下的那一點。
  • 把籠統答覆逼成明確確認。 對方說「別擔心」或「選 Yes 就好」時,回問二選一,例如「是 5 片全部,還是只有 2 片?」
  • 網站和客服說法衝突時,貼網站原文。
  • 核心問題解決前不接受補償。 在信中明寫「尚未接受 <補償> 作為解決方案」。

工程澄清

  • 寫明座標系、平移量和 mm 數值,例如 original (12.3, 4.5) mm single-board coordinates; CAM offset (+23, +17)。
  • 主動排除誤解:說明「並非指稱斷路」,或「這個讀數可能是游標位置或料庫原點」。
  • 請對方附上標註圖。
  • 不要為了讓顯示數值對齊而改設計;先確認那個數值代表什麼。
  • 重要要求同時寫進訂單備註和附件(裝配圖 PDF),工廠 CAM 端看得到備註。

管道

管道用途
線上聊天即時問答、催辦;結束前確認拿到一個能驗證的承諾
註冊信箱需要書面證據、附件、工程澄清;回在同一信串
業務或合作窗口窗口失效或要協調多個部門時才找。先報進度、說明影響,再提出請求
訂單備註+附件送單時就把工序要求寫進去,讓 CAM 端看得到

期限倒數

收到生產稿或貼裝確認信時,立刻算出到期的絕對時間(含時區),寫進 WORKLOG 並告訴使用者。用本 skill 的腳本計算,不要心算:

python -I scripts/deadline.py --received "<收信時間>" --tz +08:00 --hours 72

退出碼 0 代表還沒到期、1 代表已過期、2 代表輸入錯誤。輸出的「deadline」那幾行照抄進 WORKLOG。有些確認稿逾期會直接投產,在期限前必須讓使用者做出決定。

過去觀察到的期限見 references/deadlines.md。那是快照,以當下的信件與網站為準。

節奏與停止條件

  • 對方沒有承諾時間時,主動催辦至少相隔 4 小時;只查不催的監控可以更密。對方給了時間,就照那個時間。這是間隔的下限,不是自動寄出的許可:每則催辦仍要使用者同意,或落在硬規則的限定例外內。
  • 以下情況通知使用者:有實質進展、出現失敗、需要使用者決定、期限倒數。
  • 以下情況停止:對方最終拒絕;已取得報告並交給使用者;登入或瀏覽器中斷(通知一次後暫停);使用者說暫停。
  • 對方說「會寄信」「正在催」,不算結案。

工具搭配

原則:MCP 只用來看當下狀態,腳本用來算期限、做判定。MCP 回覆「成功」不代表訊息已送達或狀態已改變,要讀回確認。

步驟MCP/工具腳本(相對於本 skill 資料夾)限制
看訂單詳細狀態、客服對話瀏覽器:Claude in Chrome(Claude Code);ChatGPT browser extension(Codex 桌面版 @Chrome,Codex CLI 不支援)—只讀;不點 Confirm、Submit,不改選項,不下載
搜註冊信箱、讀信串信箱:Gmail connector(claude.ai 或 ChatGPT)—先確認連到訂單註冊帳號;寄出另依硬規則
起草訊息——草稿可以先寫;寄出等使用者同意
算確認稿到期時間—scripts/deadline.py結果寫進 WORKLOG
生產稿、貼裝確認的比對—交給 jlc-production-review skill本 skill 只負責把它的待問清單寫成澄清信
  • 上表只列實際用過的工具。其他瀏覽器或信箱工具未驗證,用之前先列出實際工具與 schema,不要猜參數。
  • 沒有工具時的退路:沒有瀏覽器或信箱工具(例如在 Codex CLI 裡),請使用者貼上訂單頁的詳細狀態、相關信件與上次對話,再起草。草稿的核對紀錄註明「狀態由使用者提供,未直接核對」。

交付草稿時附上

每份草稿前面先給使用者看這幾項,方便使用者決定:

  1. 核對紀錄:看了哪些來源(訂單頁、信箱、上次對話)、核對時間與時區;沒直接核對的註明。
  2. 訊息類型:需要逐次同意,或落在使用者事先授權的例外內(寫出是哪一條授權)。
  3. 建議管道:線上聊天、註冊信箱原信串,或業務窗口,以及理由。
  4. 期限:相關確認稿的到期時間(含時區)與剩餘時間。
  5. 草稿本文:英文,含界線句。

證據

  • 每次對話保存聊天全文 txt、寄出畫面截圖,記下對方角色與時間(含時區)。
  • 證據放在專案的私有目錄。截圖常含姓名、信箱、電話、地址,公開前必須去除。
  • agent 寫的摘要不算證據。

參考

Bundled files

The model reads these on demand while the skill is loaded. They are exposed as readable files and are never executed.

Frequently asked questions

What does the Jlc Support AI skill do?

與 JLCPCB(及類似 PCB/PCBA 代工廠)的線上客服、工程師、業務窗口溝通。用於催辦審查或 DFM、就生產稿與貼裝確認向工廠提出技術澄清(比對與判讀交給 jlc-production-review)、算確認稿到期時間、訂單被取消後的追問、報價與優惠申請、付款後請對方負責到底、退款追問;先核對網站與信箱,再起草可獨立閱讀、含界線句的訊息,寄出與任何確認都等使用者同意。Use when drafting or sending messages to JLCPCB support or engineers, chasing a DFM or review, handling a cancelled order, asking the factory about production file changes, or asking for a quote or coupon. Triggers: "JLC 客服", "催 DFM", "寫信給 JLC", "訂單被取消", "問工廠生產稿", "確認稿期限", "JLCPCB support", "chase DFM", "production file question".

Why use Jlc Support on TypingMind?

Because you install it once and use it with any model. Jlc Support is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Jlc Support in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/John-owo/pcb-agent-workflow/tree/main/skills/jlc-support. TypingMind reads its SKILL.md and bundles its files and installs it as a skill you can enable per chat.

Which AI models can use Jlc Support?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Jlc Support?

As many as you like. As long as a model supports skills, you can use Jlc Support with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Jlc Support AI skill free?

Yes. It is published on GitHub by John-owo under the MIT license. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇