Synaplan
开源 AI 软件平台,面向企业:自托管 AI 代理,符合 GDPR 的聊天机器人小部件,RAG 知识库,WhatsApp 与 Outlook AI,MCP,多模型路由。支持 Docker、Kubernetes 或在 Germ 托管。
AI 翻译
尚未检测到更新。GitHub 发布版本后会显示在这里。
个人 GitHub 账号下公开仓库的所有者可以登录并验证所有权,自行管理介绍,无需管理员批准。
登录并生成验证文件,在 30 分钟内将其提交到此仓库的默认分支。其他分支或分叉仓库中的文件无效。然后在下方检查文件,即可无需管理员批准直接管理项目。请保留文件以维持管理权限。
如需登录,将前往登录页面。登录后请回到此处,再次点击所选的验证按钮。
在项目文件夹中运行 /nomorevibe verify,即可验证所有权并自行管理介绍。
一分钟完成验证正在加载完整 README。 以下为已保存的摘录。
A demo app built with Elements.
Customer booking, a crew job queue, comment threads that update live, invoices by email, and card payments.
• Agent: Claude Code, Opus 5.5 Medium
• Time: 14 min
• Cost: $6.64 at API rates, September 2026
Greenline needed customer booking, a crew queue that stays current, comment threads on each job, invoices by email when work is done, and card payments that mark a job paid. Each of those is a part of Elements, so the agent spent its 14 minutes on the lawn care business itself.查看完整 README ↗Greenline is a full-stack app built with Claude Code on Elements.
Customer booking, a crew job queue, comment threads that update live, invoices by email, and card payments.
Run this app on your machine. Use this demo as inspiration, or as the starting point for your own app. Your agent creates a new app with this one as the scaffold, and you prompt it to add whatever features you want.
Without a Stripe key, invoices are paid through the built-in test checkout: the Pay button opens a page in the app that shows the invoice and records the payment the same way a real one is recorded, so the paid status and live updates all work. No card is charged.
For real Stripe Checkout, create a free sandbox at dashboard.stripe.com/register, copy the secret key from Developers, API keys, and add it to config/env/development.env:
STRIPE_SECRET_KEY=<your sandbox secret key>
Production requires the key: the build fails without it and the app refuses to start with it empty. In production the app registers its own Stripe webhook endpoint the first time a customer starts a checkout, so there is nothing else to set up.
Greenline needed customer booking, a crew queue that stays current, comment threads on each job, invoices by email when work is done, and card payments that mark a job paid. Each of those is a part of Elements, so the agent spent its 14 minutes on the lawn care business itself.
Live jobs. Jobs are a LiveTable that the crew's queue, each customer's list and each job page all watch. Bookings, completions and payments reach every open page as they happen, so a new booking lands in the crew's queue and a payment shows on the customer's page live.
Live comment threads. Comments are a LiveTable opened per job. A comment appears for the crew and the customer at once, carries its author from the session, and schedules an email to the other side.
Booking and crew actions as function calls. The booking form and the crew's complete and invoice buttons call @rpc functions, with types checked from the page to the database.
Invoices from background jobs. Completing a job schedules its invoice in the same transaction, so every finished job has its invoice on the way. The job prices it from the service and emails the customer a payment link.
Card payments. Checkout charges exactly what the invoice says. The return page and the webhook both mark the job paid, and the payment is recorded once, whichever arrives first.
Data from SQL files. Two migrations define the business and seed a crew account, two customers, three services, and jobs in every status from scheduled to paid, with comment threads.
Elements is an integrated full-stack app framework and a project server. The project server checks the agent's work as it builds: it type-checks the server and browser code, analyzes the program, applies migrations and runs the tests, then reports the results to the agent. A security mistake, such as calling a server-only function from the browser, is a compile error. The agent reads the report and fixes its own mistakes before it tells you a feature is done. The same checks run on every feature you ask for next.