> For the complete documentation index, see [llms.txt](https://docs.myrspoven.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.myrspoven.com/release-notes/2026/august-2026.md).

# August 2026

**Version 2026.8**

Covers changes since the June 2026 release.

***

### Highlights

The AI can now **recover on its own when a control signal goes bad**. Where a faulty signal used to leave a building stuck, the optimizer requests a retrain, the next model drops the bad signal, and optimization carries on without anyone intervening. **CO2 distribution** also reaches buildings without a ventilation flow signal, and **EcoGuard** discovery finds CO2 and humidity sensors alongside temperature, so more sites get air quality visibility without extra configuration.

In BackOffice, **component edits now reconcile in place** instead of deleting and recreating, so identity, history and relations survive an update. Data delivery now uses an updated streaming connection for greater reliability.

Our **new customer interface** is in closed beta with a selected user group. It puts a whole property portfolio in one view, with per-building health checks and signal configuration.

***

### myPortal — Customers

* **CO2 distribution for more buildings** — buildings without a ventilation flow signal now get CO2 distribution published, extending air quality visibility to sites that previously showed nothing.

<figure><img src="https://2074044488-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FQ0IsWgRagqUQj5pl27zJ%2Fuploads%2Fgit-blob-00fde4717a20cf4a2d513a265f020318ebda0ec8%2Fimage%20(6).png?alt=media" alt="myPortal view showing CO₂ distribution data."><figcaption></figcaption></figure>

### BackOffice & Integration - Partners

#### **Integrations**

* **Yggio** — data delivery now handles connection shutdowns more reliably.
* **Building-data integration** — observations now flow through the updated streaming connection.
* **Mixed-schema buildings** — meter schema is set per device rather than per building, so a single site can run several meter types.
* **EcoGuard** — discovery now finds CO2 and humidity sensors, not only temperature, and maps registers by utility code so a shared measuring point can no longer feed a ppm value into a temperature signal. Discovered signals also carry readable descriptions instead of requiring a look inside the raw data.
* **Siemens Building X** — writes now update only the intended fields, and the watchdog is updated only by signals that were actually written successfully.
* **Building-data integration** — binary signals are read as raw historical values, so true/false trends are no longer lost to averaging.

#### **BackOffice**

* **Component edits keep their identity** — updating a component reconciles in place instead of deleting and recreating it. Renames are detected, relations are merged rather than rebuilt, and a component is only removed when an upload leaves it with no signals and the deletion is confirmed.
* **Partial signal uploads** — an upload can cover a subset of a building's signals; anything outside the file is left untouched.
* **Signal configuration editor** — backend support for the new editor, with per-field and building-level validation and field-level read and edit permissions.
* **Searchable signal dropdowns** — the dropdowns on Connect Sensors are now searchable, which matters on buildings with long signal lists.

**Fixes**

* **Dependency updates** — updated dependencies address known issues.

### Smart Control

* **Self-healing on faulty control signals** — when the optimizer finds a controllable signal it can no longer trust, it now requests a retrain instead of staying stuck. The next model drops the bad signal and optimization recovers on its own, with a cooldown so a building can't flood the training queue.
* **Fewer false frozen-signal alarms** — circuits with a stopped pump get a longer frozen threshold, and humidity readings are excluded from frozen checks entirely.
* **Clearer error reporting** — the optimizer now distinguishes a missing signal from a missing limit, so the cause is visible from the message.
* **Data quality made visible** — training now warns when an unusually large share of its input is dropped, surfacing signal problems that previously passed unnoticed.
* **Less cooling-season noise** — hot rail breaches on buildings without cooling are reported as warnings rather than failures during cooling season.
* **Faster, steadier optimizer** — occupancy handling and slow-run thresholds were retuned, cutting run time and variance.

**Fixes**

* Valid values are no longer discarded when a series contains gaps.

### Data Platform

* **Clean meter data** — Cleaned meter data now lands in the shared meter tables, tagged by environment, making it available to reporting and analysis.
* **Per-meter metrics** — a new pipeline computing period metrics per meter, groundwork for per-meter energy reporting.
* **Building metadata and system alerts** — building metadata is now available in the data platform, and system alerts are generated directly from the underlying tables.
* **Monthly KPI Summary expanded** — new KPIs plus a Data Issues section, so data quality problems appear alongside performance.

These platform changes aren't directly visible in myPortal, but they are what make richer Insights and reporting possible.


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation by asking a question.

Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters:

```
GET https://docs.myrspoven.com/release-notes/2026/august-2026.md?ask=<question>&goal=<user_goal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.
