Device Templates
Describe a device model once: Nodes, scaling, units. Reuse it for every identical device on every site.
Drivers read the devices. The tag model makes them comparable. Services and APIs put the data to work.

Every screen below is the live IoTServa software, captured with a demo company. Each caption says what the screen does today.























Screens from the live IoTServa software with a demo company and demo data.
Chillers, meters, switches, sensors and tags, on the protocol they already speak.
One driver per protocol, running on an IoTServa gateway at the site.
Every reading belongs to a Node with a name, unit, timestamp and quality in your hierarchy.
History, alerts, Computed Nodes and users, all working from Nodes.
Dashboards, reports, a REST API and signed webhooks.
Describe a device model once: Nodes, scaling, units. Reuse it for every identical device on every site.
Organise Nodes by Organization, Site, Facility, Level and Area, so a query reads like the real world.
Store Node history, query it by time range, and export it as CSV or JSON when another team needs it.
Raise alerts on thresholds, rate of change and silence, notify in the app, by email and by signed webhook, and escalate.
Build views for operators and managers from the same Nodes, with roles that decide who sees what.
Read Nodes and history over a REST API documented with OpenAPI, with keys scoped per Organization.
How It Works
Eight steps every site goes through with IoTServa. Watch it play, or pick a step.
Install and Pair
$ curl -fsSL install.example.com | sh -s DEMO-CODE
Release signature verified
Service started on boot
Paired with Plant Room Gateway
Signed configuration in sync
All Sensors
BACnet, Modbus, KNX, MQTT and serial values share one shape: name, unit, time and quality.
Automation Rule
When Tank Temperature Is Above 165 °C for 5 min
Condition FalseCondition True for 5 min
Tested first with a Dry Run against the last 7 days of real data.
Actions Run in Order
Offline Buffer and Local Rules
Local Rule Heater Relay Off at 166 °C Ran Offline
Gaps are recorded, never hidden. Local Rules are signed and checked against limits on the gateway.
AI Assistant
Which tanks ran hot last night?
Tank U ran above 165 °C for 12 minutes from 02:10. The Rule switched its relays once and the alert cleared at 02:22.
Turn the heater off for one hour.
Done. Heater relay is Off, confirmed by the controller. It switches back On at 10:42.
Voice input, six AI model choices, and only the rights of the person asking.
Health and Work Orders
Features in the live IoTServa system, not a roadmap.
The Standalone System Screen Scraper reads the screen of a standalone Windows system that has no data interface and turns the values into Nodes.
One command installs the Linux gateway and pairs it with your Organization automatically.
Gateway updates are signed, roll out in stages and roll back automatically if a new version fails its checks.
Supported sensor models come with ready-made register maps, so adding one is a choice, not a project.
Writes are off by default. When enabled per Node, each command stays within min and max limits, is signed, re-checked by the gateway and audited.
Replace a broken gateway, scraper or sensor and keep its history, Nodes and alert rules.
See which gateways and devices are online, stale or failing, before a gap reaches a report.
Claude-powered AI reads the values on standalone system screens, and each value is checked for range and sudden jumps before it is stored.
A three.js 3D Tank View with Realistic and Thermal modes, a Cutaway level, live sensor labels and alarm effects when a Rule fires.
Gateways keep collecting through outages on a disk-aware buffer, upload the backlog oldest first and record any data gap.
Signed automation that runs on the gateway with no internet, with hysteresis, hold time, windows and fail-safe limits.
The gateway switches Controller8 relay boards over the local network, checking the board identity before every write.
Protocols, SDKs, datasheets and manuals stored with each catalogue model and each device, up to 25 MB per file.
Designed alert emails with the reading, the limit and the site time, sent from your own company sender on a signed domain.
Place live values on floor plans, chart any Node, and send A4 PDF reports on a schedule.
Clients, Organizations and Sites with self sign-up, a free trial and custom labels for each level.
Owner, Admin, Operator and Viewer roles scoped per Client or Organization, plus per-user access control.
Triggers, seventeen conditions and ten actions, with delays, dry run, replay of the last 7 days, versions and a run history.
Switch eight-channel VYROX relay boards from the console, a Rule or the AI assistant, each command confirmed by read-back.
An assistant named after your company answers questions from live data, takes voice input and can switch equipment within your rights.
Every Telemetry Point with its chart and Rule limits, drag to rearrange, saved for each member.
Health scores per Site and a trend forecast of up to 90 days that shows which limits are coming up.
Work Orders created from alerts or Rules, assigned and tracked, with overdue reminders.
Production, energy and fuel KPIs, PM10 and PM2.5 against MAAQS 2020 and CO2e estimates.
Recorded browser terminal on a gateway for approved support staff, with time limits.
Publish live values to your MQTT broker over TLS, and receive signed webhooks for alerts, commands and Rules.
Each part has its own page with the detail.
Protocol drivers that run beside the equipment.
See Edge Gateway
One shape for every reading, whatever protocol sent it.
See Tag Model
The right view for each role, and an alert only when it matters.
See Dashboards and Alerts
Your data, available to the tools you already use.
See Open API
Describe a device model once, reuse it everywhere.
See Device Profiles
When something happens on site, the right thing happens next.
See Automation Rules
Switch equipment safely, and know that it really switched.
See Control and Relays
Ask about any site in plain words, and get an answer from live data.
See AI Assistant
See trouble coming, assign the fix, and measure the result.
See Health, Work Orders and KPIs
The right alarm reaches the right person, on the channel they will see.
See Notifications Everywhere
Fix the machine before it stops the plant.
See Predictive Maintenance
Cameras that raise an alert, not just record.
See Camera AI
Your plant in live 3D, with every reading where it happens.
See Digital Twin
The site keeps measuring and acting when the internet does not.
See Offline Buffer and Local Rules
The protocol, datasheet and manual sit next to the device.
See Device DocumentsA short list, so the first week is spent reading devices, not hunting for documents.
No. It reads from them and from the devices beneath them, and adds a common tag model, history and APIs. Existing control logic stays where it is.
Yes. Nodes and history are available over a REST API documented with OpenAPI, alerts can call your system through signed webhooks, and history exports as CSV or JSON.
The gateway keeps collecting, buffers readings on disk up to a size limit, and forwards them in order when the connection returns.
Tell us which protocols are on site. We will scope a pilot that reads your real devices.