nodrix

I’ll translate this chunk faithfully and keep the markdown structure intact, then return only the English copy.## What is nodrix? nodrix is an open-source IoT backend that deploys to the user’s own Cloudflare account. It provides a single-tenant runtime for ingesting device telemetry, storing state, visualizing data in dashboards, and triggering automations without requiring a separate broker or server.

The product is designed to let hardware send data over HTTPS or WebSocket, then turn that data into variables that can be read back through an API, displayed in realtime dashboards, or used in automation flows. It runs on Cloudflare primitives such as Workers, Durable Objects, D1, R2, and KV, so each deployment lives in the customer’s account rather than in a shared vendor environment.

Key Features

• Single-tenant deployment to a Cloudflare account — each install runs in the user’s own account, with data stored on their own Cloudflare resources.
• Telemetry ingestion over HTTPS or WebSocket — devices can POST JSON or open a WebSocket connection, which keeps integration simple for hardware and edge clients.
• Auto-created variables — incoming metrics become variables on first sight, avoiding manual schema setup or broker configuration.
• Realtime dashboards and controls — drag-and-drop widgets can display live readings and send commands back to variables.
• Visual automation builder — users can trigger flows from thresholds, time, sunrise/sunset, or events, then branch and act by setting variables, calling integrations, or emitting events.
• Clean read API — the platform exposes latest state, recent time-series, and variable listings behind a bearer token, with edge-cached reads for downstream apps.
• Authentication and access options — email/password login is included, with optional Google or GitHub sign-in and an owner-gated MCP server for AI clients.
• No SDK requirement — any client that can speak HTTPS can send telemetry and read data back from the worker URL.

How to Use nodrix

Start by deploying nodrix to your Cloudflare account, which provisions the required Worker and storage resources. Then claim the instance by creating the first account, create a project, and mint a project token for devices or clients.

From there, point hardware at the /v1/telemetry endpoint to send metrics, build dashboards by binding widgets to variables, and create automations for actions you want to run on thresholds or schedules. If another application needs the data, read it through the REST API or connect an AI client through the MCP server when that access is enabled.

Use Cases

• Device telemetry collection — gather sensor values from hardware that can send HTTPS or WebSocket requests, then store and inspect the readings in one place.
• Live operational dashboards — monitor metrics such as temperature, humidity, or device status in realtime with dashboard widgets bound to variables.
• Rule-based automation — trigger actions when a value crosses a threshold, when a clock time is reached, or around sunrise and sunset.
• Data sharing with other apps — expose current state and recent timeseries to a React app, Grafana setup, or other consumer that reads via API.
• Edge-connected control surfaces — use toggles, sliders, gauges, or screens connected to a Raspberry Pi or similar device to read and control variables locally.

FAQ

Does nodrix require an MQTT broker? No. The product description says hardware can send data over plain HTTPS or WebSocket, so no broker is required for the core flow.

Where does the data run and store? Each deployment lands in the user’s own Cloudflare account and uses Cloudflare primitives such as Workers, Durable Objects, D1, R2, and KV.

Can the platform be used without writing an SDK integration? Yes. The site says no SDK is required; a client can send telemetry to the worker URL and read data back through the API.

Does nodrix support AI clients? Yes, via an MCP server. The site notes that access to automations is owner-gated, off by default, and audit-logged.

What authentication options are mentioned? Email and password are included out of the box, with optional Google or GitHub sign-in.

Alternatives

• Hosted IoT platforms — these typically centralize data on the vendor’s infrastructure and are managed as a shared service rather than a single-tenant deployment in your own account.
• MQTT broker plus custom backend — a traditional approach where a broker handles ingestion and separate services are built for storage, dashboards, and automation.
• Cloudflare-native custom build — teams can assemble their own IoT backend using Cloudflare Workers and storage products directly, but with more implementation work than using nodrix.
• General dashboard and automation stacks — tools focused on visualization or workflow automation can complement or replace parts of the stack, but they do not provide the same end-to-end IoT ingestion and variable model.