MEATER smart home integration, free and open source
MEATER wireless meat thermometers in Gladys: temperatures, remaining time, cook state.
Published by guim31
- Devices
- Version 1.0.1
- Requires Gladys >=5.1.0
- ★ 0
Follow your MEATER wireless meat thermometers in Gladys: internal and ambient temperatures, target, remaining time and state of the cook, and scenes that start at the right moment of the cook.
Not affiliated with Apption Labs / MEATER. MEATER is a trademark of Apption Labs. Developed without the hardware: feedback welcome, on the Gladys forum or in the issues.
Overview
The integration reads your probes through the official MEATER Cloud API, with the account of the MEATER app. Each probe becomes a Gladys device, with its temperatures and its cook. Three scene triggers mark the moments of a cook (almost ready, remove from heat, resting done), and a dashboard widget shows one cook at a glance.
Everything is read only: Gladys never starts, changes or stops a cook. The cook is set up in the MEATER app, as usual.
Requirements
- Gladys 5.1.0 or later, with Internet access.
- A MEATER probe: MEATER, MEATER+, MEATER 2 Plus, MEATER Pro, MEATER Pro XL, or a MEATER Block.
- A MEATER Cloud account, signed in in the MEATER app.
- During the cook, the MEATER app (phone or tablet) or the Block must stay connected to the probe (Bluetooth) and to the Internet: MEATER Cloud only knows what the app or the Block relays.
Configuration
In Gladys, open Integrations, then MEATER, tab Configuration:
| Field | What to enter |
|---|---|
| E-mail address | The e-mail address of your MEATER Cloud account. |
| Password | Its password. |
| Language | Language of the feature names, of the cook states and of the cut names sent by MEATER Cloud (French by default). |
| Refresh during a cook | Every 15 s, 30 s (recommended) or 60 s while a cook is in progress. Every minute otherwise. |
Save, then click Test the connection: the message says how many probes MEATER Cloud sees right now.
Then add the probes:
- Start a cook in the MEATER app (a probe only shows in MEATER Cloud during a cook).
- In the Discovery tab, click the search button, then Add to Gladys on each probe.
A probe seen once stays in the Discovery tab, even between cooks.
Devices
One device per probe, named "MEATER probe" followed by the 8 first characters of its identifier (rename it in Gladys). Every feature is read only.
| Feature | Content |
|---|---|
| Internal temperature | Temperature at the tip of the probe, in the meat (°C). |
| Ambient temperature | Temperature around the meat, in the oven or on the grill (°C). |
| Target temperature | The target set in the app for the cook (°C). |
| Peak temperature | The highest internal temperature reached during the cook (°C). |
| Remaining time | The time left estimated by MEATER (minutes). Kept at its last value while MEATER is still estimating. |
| Elapsed time | Time since the start of the cook (minutes). |
| Cook state | No cook, Not started, Set up, Cooking, Remove from heat, Resting, Done, Slightly underdone, Slightly overdone, Overcooked!. |
| Cook | The name of the cut or the name given to the cook in the app. |
| Cook in progress | 1 from the start of the cook to the end of the rest, 0 otherwise. |
| Probe connected | 1 while MEATER Cloud sees the probe, 0 otherwise. |
When a probe is no longer seen (cook over, probe back in its charger, phone out of range), "Probe connected" goes to 0 at once and the last readings stay. After 30 minutes without news, the cook is considered over: "Cook in progress" goes back to 0 and "Cook state" to "No cook".
Dashboard widget
MEATER cook shows one probe: the name of the cook, the internal, target and ambient temperatures (live), the remaining time, the state of the cook, the elapsed time and whether the probe is connected.
In its settings, pick a probe, or leave it empty: the widget then shows the probe that is cooking.
Scene triggers
In a scene, Add trigger, then pick one of the MEATER triggers. Each one fires once per cook, even if Gladys restarts in the middle of it.
| Trigger | When | Filters | Information passed on to the scene |
|---|---|---|---|
| Cook almost ready | The remaining time estimated by MEATER goes under 5, 10, 15 or 30 minutes, meat on the heat | Probe (optional), threshold (mandatory) | Cook, remaining time, threshold, internal and target temperatures |
| Remove from heat | MEATER says to take the meat off the heat and let it rest | Probe (optional) | Cook, internal and target temperatures |
| Resting done | The rest is over and MEATER gives its verdict | Probe, result (both optional) | Cook, result (Done, Slightly underdone…), internal and target temperatures |
The texts passed on (cook state, result) follow the language of the configuration.
Action
Test the connection (Configuration tab) signs in to MEATER Cloud and tells how many probes it sees right now, or what is wrong (refused password, service unreachable, too many requests). After a refused password, it is also the way to try again without saving the configuration.
Scene examples
The words in brackets are information of the trigger, inserted in the text with the variable picker of the scene editor.
- Set the table: trigger "Cook almost ready" with the threshold "10 minutes", action "Send Message": "[Cook] is ready in 10 minutes, time to set the table."
- Off the heat: trigger "Remove from heat", actions "Make the Lights blink" in the kitchen and "Talk on a speaker": "Take [Cook] off the heat, it is at [Internal temperature] °C."
- Dinner is served: trigger "Resting done" with the result "Done", action "Send Message" to the whole family: "Dinner is served!"
- Grill too hot: trigger "Device state change" on the "Ambient temperature" of the probe, above 300 °C: a message to check the grill.
Refresh and quotas
MEATER Cloud allows 60 requests per minute at most, and recommends 2. The integration reads the account:
- at the pace chosen in the configuration while a cook is set up or in progress (every 30 s by default);
- every minute otherwise, as the API documentation advises;
- never twice within 2 seconds;
- after a failure, after 1, 2, 4… minutes, up to 10 minutes (5 minutes at least when MEATER Cloud asks to slow down).
One request answers for every probe of the account. The integration only sends Gladys the values that changed (and every value at least every 6 hours, so none looks stale), and it does its own refresh: it does not depend on the Gladys scheduler.
What is stored in the integration's data volume: the MEATER Cloud token (it does not expire) with a fingerprint of the account, and the list of the probes seen with the memory of their current cook. Neither the e-mail address nor the password is written there; the password stays in the Gladys configuration.
Troubleshooting
- "MEATER Cloud refused the e-mail address or the password": use the address and the password of the MEATER app, then save the configuration again. The integration does not retry by itself, so as not to lock the account.
- No probe in the Discovery tab: a probe only shows during a cook, while the MEATER app or the Block is connected to it and to the Internet. Start a cook, click Test the connection, then search again.
- The values no longer move: the phone running the MEATER app is out of Bluetooth range, asleep, or offline. "Probe connected" tells whether MEATER Cloud still sees the probe.
- "MEATER Cloud asks to slow down": another program reads the same account too often. Choose a slower refresh during a cook.
- More details: the integration logs (Supervision tab). Set the
environment variable
LOG_LEVEL=debugto see every refresh.
Limitations
- Cloud only: Gladys does not talk to the probe over Bluetooth. The data arrives with the delay of the MEATER app, of MEATER Cloud and of the refresh pace (30 s by default).
- The MEATER Cloud API is in beta: MEATER may change it.
- The API gives one name per cook (the cut, or the name you typed): there is no separate recipe name.
- The remaining time is MEATER's estimate; it is unknown at the start of the cook.
- The temperatures come in degrees Celsius.
- Read only: no cook setup, no alarm from Gladys.
- Developed without the hardware nor an account: the behaviour was written from the official API documentation and from the Home Assistant integration. Feedback welcome.
Configuration settings
These are the settings MEATER asks for in its configuration screen in Gladys.
| Setting | Type | Required | Description |
|---|---|---|---|
| MEATER Cloud account | section | No | The probes are read through the official MEATER Cloud API, with the account of the MEATER app. In the app, sign in to MEATER Cloud first. A probe only shows while the MEATER app or Block is connected to it and to MEATER Cloud, during a cook. Read only: Gladys never changes a cook. |
| E-mail address | string | Yes | The e-mail address of your MEATER Cloud account. |
| Password | secret | Yes | |
| Language | select | No | Language of the feature names, of the cook states and of the cut names sent by MEATER Cloud. |
| Refresh during a cook | select | No | How often MEATER Cloud is read while a cook is in progress. Every minute otherwise, as the MEATER documentation advises. |
How to install MEATER in Gladys
- In Gladys, open Integrations: MEATER appears in the catalog, next to the native integrations, with a community badge.
- Click Install. Gladys pulls the Docker image (
ghcr.io/guim31/gladys-meater:1.0.1), starts it in a sandbox isolated from the core, and generates the integration's interface (devices, discovery and configuration). - Open the Configuration screen of the integration, fill in the settings, and save.
- You can also install it directly from its repository URL: https://github.com/guim31/gladys-meater.
MEATER requires Gladys >=5.1.0. The catalog inside Gladys refreshes every hour, so a new version becomes available at most one hour after its release.
Not running Gladys yet? It is free and open source: follow the installation guide to get started.
About external integrations
MEATER is an external integration: a community integration packaged as a Docker container and published on GitHub, that Gladys installs in one click and runs in a sandbox isolated from its core. It is published and maintained by guim31, not by the Gladys core team.
- Browse all integrations, native and community
- Build and publish your own external integration
- Source code on GitHub — source of this documentation