Synaplan
開源 AI 平台,供企業使用:自行託管的 AI 代理、符合 GDPR 的聊天機器人小工具、RAG 知識庫、WhatsApp 與 Outlook AI、MCP、多模型路由。Docker、Kubernetes 或託管於 Germ
AI 翻譯
尚未偵測到更新。GitHub 發布版本後會顯示在這裡。
個人 GitHub 帳號下公開儲存庫的擁有者可以登入並驗證所有權,自行管理介紹,無需管理員核准。
登入並產生驗證檔案,在 30 分鐘內將其提交至此儲存庫的預設分支。其他分支或分叉儲存庫中的檔案無效。然後在下方檢查檔案,即可無需管理員核准直接管理專案。請保留檔案以維持管理權限。
如需登入,將前往登入頁面。登入後請回到此處,再次點擊所選的驗證按鈕。
在專案資料夾中執行 /nomorevibe verify,即可驗證所有權並自行管理介紹。
一分鐘完成驗證製作者的介紹僅有上面的一句話。 更多詳情請閱讀儲存庫 README。 儲存庫 README ↗
正在載入完整 README。
檢視完整 README ↗A clickable, no-backend prototype of the Permit to Work, training, Safety Passport and HSSE reporting system for Shell retail stations in Oman, delivered with Artelia. Prepared by FullCircleWali for stakeholder review.
Everything runs in the browser. Data is stored in localStorage, seeded with
Shell demo data on first load. Nothing is sent to a server, no real payments
are taken and no emergency services are contacted.
# 1. In this folder
git init
git add .
git commit -m "Gemsafety prototype"
# 2. Create an empty repo on github.com (no README), then:
git remote add origin https://github.com/<you>/<repo>.git
git branch -M main
git push -u origin main
git push to main redeploys automatically.Run locally:
npm install
npm run dev # http://localhost:3000
The camera (face recognition, course proctoring, video call) only works on
https://orlocalhost, which Vercel provides.
The login page has two tabs, Username and Civil ID. Under the form
is a Demo accounts list — click any person and their username/password
(or Civil ID + expiry) is filled in. Every demo password is Shell@2026.
| Person | Role | Username | Civil ID / expiry |
|---|---|---|---|
| Ahmed Al Rashdi | Permit Holder | ahmed.holder | 12345678 / 2028-05-31 |
| Sara Al Kindi | Permit Holder | sara.holder | 12876540 / 2027-02-28 |
| Rashid Al Maskari | Permit Holder (imported) | rashid.maskari | 87654321 / 2029-08-15 |
| Hamed Al Siyabi | Trainee (part-way) | hamed.trainee | 13300912 / 2030-04-01 |
| Noor Al Lawati | Trainee (ready to register) | noor.trainee | 13455120 / 2028-12-19 |
| Salim Al Amri | Trainee (registration pending) | salim.trainee | 11873402 / 2027-10-05 |
| Mariam Al Farsi | Permit Issuer | mariam.issuer | 10456721 / 2027-11-02 |
| Khalid Al Habsi | Permit Issuer | khalid.issuer | 10988230 / 2029-01-20 |
| Fahad Al Balushi | Fuel Station (Shell Al Khuwair) | alkhuwair.station | 11203344 / 2027-06-30 |
| Layla Al Hinai | Fuel Station (Shell Sohar) | sohar.station | 11407788 / 2028-09-12 |
| Yusuf Al Balushi | Administrator | admin | 09812345 / 2028-03-14 |
"Your account already exists": go to Create an account and enter Civil
ID 87654321, expiry 2029-08-15. Rashid was imported from the legacy
register, so Gemsafety tells him his account already exists, lists the
certificates already on his record, and signs him in with his Civil ID.
Branding — Shell and Artelia logos on the login page, sidebar, top bar, welcome tour, Safety Passport and every PDF (permit, certificate, journey plan, passport). Shell stations and Shell demo data throughout.
Guided tour & tips — first sign-in for each user shows a welcome card and a step-by-step spotlight tour of their menu. The ? button in the top bar shows role tips and replays the tour.
Training
4242 4242 4242 4242 (or 5555 5555 5555 4444); real card numbers
are rejected and only the last four digits are kept./verify/<passport no>), stamps for every course, PDF download.Face recognition — warning screen → live camera with face oval → capture → match result. Enrol from My profile or the Safety Passport. Matching is simulated in the prototype (production: Amazon Rekognition with Face Liveness, per the quotation).
HSSE
Admin — dashboard with live charts, Import existing workers (Users → Import: loads a sample legacy register with certificates), training modules grouped Shell / Artelia / Local with prices, course revenue.
Fixes from the walkthrough — camera previews now show video (F-01), the presenter note is removed from the documents step (F-02).
Sidebar → Reset demo data restores the seed and signs everyone out.
No database, no server-side auth, no real payments, no real face matching. The production build (PostgreSQL, real auth/RBAC, object storage, AmwalPay, Amazon Rekognition) is described in the quotation.