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。
KnightHacks 2026 · NextEra Energy SPOT Management Dashboard track. An operator console that tells you what a SPOT inspection mission actually found, not just that it "completed".
All data is synthetic and fictional, supplied by NextEra Energy for KnightHacks 2026.
Requires Node 20+ (CI uses 22). Run from the repository root:
For real-API mode, copy .env.example to .env.local before npm run dev:查看完整 README ↗A control room for SPOT, the four-legged inspection robot. Built at KnightHacks 2026 for the NextEra Energy "SPOT Management Dashboard" challenge.
Try it online: https://knight-hacks2026.vercel.app
Power plants send robots like Boston Dynamics' SPOT on walking routes to read gauges, take heat and sound measurements, and photograph equipment. When a route ends, the robot usually reports one word: "completed". That word hides a lot. A "completed" route can still have skipped stops, readings that can't be trusted, and serious problems with the equipment it looked at.
Watchdog reads everything the robot recorded and tells the person on duty, in plain language:
| Screen | What it shows |
|---|---|
| Live | A cartoon map of the plant with SPOT walking its route, stop by stop, as it happened. A side panel lists what it is doing and anything that needs attention. You can also try "what if" situations, like a heat wave or a box blocking the charging dock, and check whether today's weather is safe for an outdoor route. |
| Missions | Every route the robot walked, with search and filters. Routes marked "completed" that still had problems are flagged. |
| Mission report | One route explained: what went wrong, why, how serious it is, and buttons for the operator to choose what to do (send a technician, keep watching, or dismiss). |
| Equipment page | The history of one piece of equipment: its readings over time, past problems, and a sample inspection photo. |
| Insights | The big picture across all routes: where problems cluster on the map, which stops fail most, how battery level relates to failures, and how two robot software versions compare. |
All plant data is fictional. NextEra Energy supplied it for the hackathon; no real plant, robot or equipment is shown.
You only need this if you want to run your own copy. The link above works without installing anything.
npm, which installs the rest.Open a terminal (on Windows: PowerShell; on Mac: Terminal) and run:
git clone https://github.com/jbear05/KnightHacks2026.git
cd KnightHacks2026
npm ci
npm run dev
The first command downloads what the app needs and takes a few minutes. When the second one prints Ready, open http://localhost:3000 in your browser. To stop it, press Ctrl + C in the terminal.
In this mode a yellow "Mock data mode" banner at the top reminds you that you are looking at sample data. Every screen works, but choices you make (like "send a technician") are forgotten when you reload the page.
This makes the app read from a database and remember the operator's choices. It takes about ten minutes and needs two extra things:
postgresql://. Team members can use the team's existing Neon database, which already has the data loaded, and skip the migrate and seed lines below.SPOT_KnightHacks_Challenge_Kit Part One from the kit inside the KnightHacks2026 folder.Start the data server (the "backend") in one terminal:
cd backend
npm ci
cp .env.example .env
On Windows PowerShell use copy .env.example .env instead of cp. Open the new file backend/.env in a text editor and paste your database link after DATABASE_URL=. Then:
npm run migrate
npm run seed -- "../SPOT_KnightHacks_Challenge_Kit Part One/02_DATA"
npm run dev
migrate creates the tables and seed loads the challenge data; you only run them once per database. Leave npm run dev running: the backend is now at http://localhost:3001.
In a second terminal, from the KnightHacks2026 folder, point the app at the backend and start it:
cp .env.example .env.local
npm run dev
(Again, copy instead of cp on Windows.) Open http://localhost:3000. The yellow "Mock data mode" banner is gone, and operator choices now survive a reload. To go back to sample data, delete .env.local and restart npm run dev.
| What you see | What to do |
|---|---|
npm or node is "not recognized" | Node.js isn't installed, or the terminal was open before you installed it. Install it and open a new terminal. |
| Port 3000 or 3001 "already in use" | The app is already running in another terminal. Stop it with Ctrl + C, or use the one that is running. |
| Pages show "couldn't load" errors after step 4 | The backend isn't running, or backend/.env has the wrong database link. Check the first terminal for an error. |
Missing dataset file when seeding | The challenge folder isn't in the right place, or its name is different. It must be named exactly SPOT_KnightHacks_Challenge_Kit Part One. |
| Role | Person |
|---|---|
| Design & Frontend | Jair Garcia (@jbear05) |
| Map & Simulation | Ilias Bahau@IliasCoder |
| Backend & Database | Alfredo Cuevas @FredoC7 |
| Data & AI | Rayan Hannani @Rayan15-07 |
ARCHITECTURE.md explains how the pieces fit together, with a section for each role. The backend has its own guide in backend/README.md, and the map engine in map-sim/README.md.