
The **Zira plugin** for Claude and Codex learned a lot this week. It can now keep an eye on every
camera, set up your weekly reports for you, and help you build better Vision models from your own
production images.

> **Beta.** The Zira plugin is in beta. To try it with your team, request access at
> [sales@zira.us](mailto:sales@zira.us?subject=Zira%20plugin%20beta%20access).

## Device status for every camera

> *"Are my cameras OK?"* or *"Which cameras have a low battery?"*

Zira checks every camera at your sites and shows its status on one page:

- **Online or offline**, with how long a camera has been quiet. Cameras that stopped recently come
  first.
- **Battery and charging**, with a warning when a battery is low and not charging.
- **Temperature, power and device errors.**
- **Camera on or off, and camera moved**, so you know when a camera needs to be pointed again.
- **Wi-Fi, cellular and network cable.**

You do not need to set up production counts or targets first. Ask *"Check device status every
weekday at 6:30"* and Zira checks each morning, telling you only when a camera goes offline, comes
back, or reports something new.

![Device status for a pallet plant: one camera that went quiet first, then cameras to check, such as a moved camera, a low battery and a camera that is off.](/changelogs/zira-plugin-device-status-and-scheduled-reports/01-device-status.png)

## Weekly reports in a few questions

> *"Send me a weekly report for my saws."*

Pick the machines, and Zira recognizes each one: trimsaws, bandsaws, Pallet QA cameras and forms
each get a ready-made report with the numbers that matter for them. Choose a day, a week or a
month, and every report compares it with the one before:

- **Output, yield, fail rate and running time**, each marked better or worse.
- **Days worth a look**, such as a high fail rate, or long stops with no reason entered.
- **Machines that went quiet**, and since when.
- **Charts you can explore**, with **CSV** and **Open in Sheets**.

Zira runs it on the schedule you choose. In Claude, it can also keep a private page updated with
the latest report, so you open the same link every week.

![A weekly report for four saws: each number compared with the week before, the status of each machine, and units per day.](/changelogs/zira-plugin-device-status-and-scheduled-reports/02-scheduled-readings-report.png)

## Health checks that stay quiet

> *"Check my Ohio sites' devices every morning."*

Set up a daily health check for one site or all of them in a few questions. Each run updates the
health dashboard. When nothing changed, Zira tells you so in one line.

## Find more rare defects

> *"Find more raised nails for Cam 1 Quality."*

Rare defects are hard to train for because there are so few examples. Zira searches your camera's
recent production images for likely examples and opens a review page. You accept or reject each
one, and Zira turns the accepted images into a dataset, ready for you to add to the camera in Zira.
You can also check an existing dataset for labels that look wrong. This works with Pallet Vision
cameras.

## Test models on fresh production images

> *"How does the newest model do on yesterday's production?"*

Score a Vision model on recent images from a Pallet Vision camera's production line, images it has
never trained on, cut up the same way Zira prepares its training datasets.

## See an AI report before you schedule it

- **A sample first.** Before a scheduled AI report is saved, Zira writes a sample in the chat, so
  you see what your team will receive.
- **Fine-tune it in Zira.** Open the draft in Zira's agent editor, change it there or in the chat,
  and schedule it when it reads right.
- **End-of-day reports.** Ask for *"every day at 18:00 with today's data"*, and the report covers
  the day that is ending, not yesterday.

## Getting started

Request beta access at [sales@zira.us](mailto:sales@zira.us?subject=Zira%20plugin%20beta%20access), then read the guide:

- [Zira in Claude and Codex](/docs/ai-assistance/zira-plugin)


Battery, temperature, camera on or moved, network and power for every camera, now or every morning.

Device status for every camera

Pick your saws, Pallet QA cameras or forms. Each report compares with the week before and points out bad days.

Weekly reports in a few questions

Check every site's devices each morning and hear from Zira only when something changes.

Health checks that stay quiet

Find more raised nails in recent production images, review them, and build a dataset from them.

Find more rare defects

Score a model on recent production images it has never seen.

Test models on fresh images

Read a sample in the chat, fine-tune the draft in Zira, and get end-of-day reports that cover today.

See an AI report before you schedule it

The Zira plugin for Claude and Codex can now check the battery, temperature, camera and connection of every camera, set up weekly reports for your saws and Pallet QA cameras in a few questions, find more rare defects in your production images, and show an AI report before you schedule it.

Zira Plugin: Device Status, Scheduled Reports and Rare Defects (Beta)


Your AI chat is now a control room for Zira. Install the **Zira plugin** in **Claude** or
**Codex**, and ask about your plant in plain words: your production data, device health, tasks,
dashboards or device setup. When you want something done, just say so.

Zira signs in as you, sees only what you can see, and asks before it creates or changes anything.

> **Beta.** The Zira plugin is in beta. To try it with your team, request access at
> [sales@zira.us](mailto:sales@zira.us?subject=Zira%20plugin%20beta%20access).

## Install from the marketplace

Find **Zira** in the plugin marketplace of Claude or Codex and click **Install**. Sign in with your
Zira account the first time you use it, and pick the site you work with most.

Then type `/zira` in the chat box to see every Zira action in one list.

## Ask about your production

> *"Show me the performance of Trimsaw 1 over the past week."*

Zira answers in the chat and opens an interactive report beside it: a short written summary, the
key numbers, and charts you can explore by date, interval and machine.

![Zira's answer about Trimsaw 1 in the chat, with the interactive report beside it.](/changelogs/zira-plugin-for-claude-and-codex/04-trimsaw-answer.png)

## Save, schedule and share reports

- *"Save this report as Trimsaw weekly"*, then rerun it with fresh data any time.
- *"Run Trimsaw weekly every Monday at 7:00"*, and a new report is ready each week.
- *"Share this report with my team"*, by sending the file or a private link.
- Take any chart into a spreadsheet with **CSV** or **Open in Sheets**.

## How accurate is your Vision model?

> *"How accurate is the newest model on Bandsaw 1?"*

Zira picks up the camera's dataset and model, runs the model on your computer, and opens an
evaluation report: precision and recall for every class, and every image marked up with what the
model got right, missed or added. Change a class's confidence threshold and the scores update
live. Nothing is uploaded or changed in Zira.

![A model evaluation: overall scores, a score for each class, and images with correct, missed and extra finds marked.](/changelogs/zira-plugin-for-claude-and-codex/05-model-evaluation.png)

## Your device fleet at a glance

> *"Show me the health of my device fleet."*

A health dashboard lists the devices that need attention first, such as a camera that stopped
sending data or a device running hot. Ask Zira to look into a finding, prepare follow-up tasks,
share the dashboard, or repeat the check every working day. In Claude, a live fleet status sits
above the chat box, and `/zira-fleet` opens the full panel.

![A health dashboard with the devices that are down listed first.](/changelogs/zira-plugin-for-claude-and-codex/07-health-report.png)

## Tasks without leaving the chat

Ask *"Show my open tasks"* to see your work across sites. Or describe a new task, review the
preview, and say yes. The assignee is notified and you get the link.

## Find anything at your site

Ask for your companies, sites, machines and dashboards, and get links straight into Zira. Ask
*"How did the plant do yesterday?"* and Zira reads your dashboards and sums it up in a few lines.

## Dashboards, forms, data sources and AI reports

- **Dashboards**: name the data sources, and Zira starts the dashboard and opens Zira's AI builder
  to add the charts.
- **Forms**: Zira drafts the questions with you and opens the form editor for you to save.
- **Data sources**: Zira creates a Vision or Pallet QA data source from the right template.
- **AI reports**: Zira schedules a written summary of your dashboards, emailed to the people you
  choose.

## Vision, step by step

Setting up a Zira Vision or Pallet Vision device? Zira guides you from pairing and mounting
through recording, datasets, training and validation. It can also open a dataset for annotation,
or build a smaller, balanced dataset.

## And a few more things

- **Every answer links into Zira.** Machines, dashboards, tasks and datasets open on the right
  page with one click.
- **Remembers your site.** Pick a default company and site once; say *"change my default Zira
  site"* to switch.
- **A preview before every change**, and a warning when a similar task is already open.
- **Load more dates from the report.** Pick earlier dates in a report and click **Load**; weeks
  already downloaded are reused, so it stays fast.
- **Times in your site's time zone**, so "yesterday's second shift" means what you expect.
- **CSV and Open in Sheets** on every chart and table.
- **Quiet daily checks.** A scheduled health check can tell you only when something changes.
- **Today's problems first.** Devices that have been quiet for a long time are listed separately.
- **Offline and back-online notices** in Claude while you work, for your saved site.
- **Reports work offline** and stay on your computer until you choose to share them.
- **Private links** for sharing a report or health dashboard, with you choosing who can open it.
- **Same actions in Claude and Codex**, so your team can use whichever assistant they prefer.

## Getting started

Request beta access at [sales@zira.us](mailto:sales@zira.us?subject=Zira%20plugin%20beta%20access), then read the guide:

- [Zira in Claude and Codex](/docs/ai-assistance/zira-plugin)


Install Zira from the Claude or Codex marketplace and type /zira to see everything it can do. Email sales@zira.us for access.

Zira in your AI chat (beta)

"Show me the performance of Trimsaw 1 over the past week." Get the answer and an interactive report.

Ask about your production

Rerun a report with fresh data, have it ready every Monday, or share it with your team.

Save, schedule and share reports

Score a model on a dataset class by class, with every missed and extra find marked on the images.

How accurate is your Vision model?

A health dashboard puts the devices that need attention first, with live fleet status in Claude.

Your device fleet at a glance

See your open tasks, or describe a task, dashboard, form or AI report and confirm the preview.

Tasks, dashboards and forms

Set up a device, open a dataset for annotation, or build a smaller, balanced dataset.

Vision, step by step

Install the Zira plugin in Claude or Codex and control Zira from your AI chat. Ask about your production data, check your device fleet, schedule reports, manage tasks, build dashboards and forms, and set up Vision devices in plain words.

Run Zira from Claude and Codex (Beta)


Until now, a camera that spotted a problem could tell you about it. In this release it can also do
something about it — a routing rule can switch an andon light or trip a PLC contact the moment it
matches. Alongside that: a trim saw calibrates itself while you watch, models push to a named slot
and tell you they got there, and teaching a camera something new no longer asks you to fill in a
form full of machine-learning vocabulary.

## Ask Claude about your plant

**In beta.** You can now connect a site to **Claude** and ask about your operation the way you would
ask a supervisor:

- *"Why did the trim saw stall last night? Check its readings around 22:00."*
- *"What was yesterday's output on line 3?"*
- *"Show me the open tasks for the Rehovot plant."*

Claude finds the right data source by name and looks the readings up in your live Zira data. No more
copying readings into a chat window or pasting answers back out.

Open your **profile picture** and choose **Connect to Claude**, pick the site, choose what Claude may
do, and paste one line into Claude. That is the whole setup.

Choose **Look only** and Claude can answer questions but change nothing. Choose **Look and act** and
it can open and update tasks, record readings and post to a channel — asking you before each one.

We are still refining this one, and we would like to hear how it goes.

![Connecting a site to Claude: pick the site, choose what Claude may do, press Connect.](/changelogs/vision-hub-training-and-datasets/connect-to-claude/01-connect-panel.png)

Read more: [Connect Zira to Claude](/docs/ai-assistance/connect-to-claude) ·
[the full release note](/whats-new/connect-zira-to-claude)

## The AI Assistant, from any page

Until now, the AI Assistant came with a trade: to use it, it had to become your home page. Plenty of
people would rather start the day in the feed or on a dashboard — so the assistant now has a door of
its own.

- Look for the **AI Assistant** icon — the sparkles — in the left menu, just above **Chats**. It
  opens the same assistant, with the same answers, previews and builders, and your Work mode stays
  exactly as it is.
- **Your saved chats come with you.** A conversation started from the icon is under **Saved chats**
  wherever you open the assistant.
- **On a phone,** open the menu and tap the same icon.
- **Heading back is one click:** open the menu and click the Zira logo to return to your home page.

If the AI Assistant is already your Work mode, nothing changes — it is still the first thing you see,
so the icon stays out of your way.

Read more: [Home AI Assistant](/docs/ai-assistance/home-ai-assistant) ·
[Choose Your Work Mode](/docs/getting-started/profile-preferences/home-page)

## Let a rule switch something on the floor

Routing decides which lane a pallet goes down. Now a rule can also **do something** the moment it
matches — so a decision on the screen becomes an action on the floor.

**Actuators** drive the camera's relay outputs. A relay is a dry contact, so anything that takes a
contact closure can be driven from a rule:

- **Andon lights** — put the red lamp on when a severe defect is routed, and clear it when the line
  recovers.
- **A machine PLC** — stop or start the line, or fire an eject, by closing the contact the PLC is
  already watching.
- Horns, diverters, gates — anything on a relay.

Pick the actuator and relay, then choose **on**, **off**, or **pulse**. A pulse takes a duration,
which is what an eject or a momentary stop signal wants. Actuators that publish named functions of
their own offer those too, with whatever fields the function asks for.

Only actuators the camera currently reports as **online** can be chosen. If a rule points at one
that has gone away, it stays on the rule as a warning rather than being quietly reassigned — so you
find out before it drives the wrong thing.

**Webhooks** are the other half: send an HTTP request when a rule matches, to tell an MES, a
scheduler, or anything else that should know. A rule can call several.

Rules in the **Default** stage can now be set to always match, so pallets nothing else caught can
still trigger an action — not just fall through to a lane.

### And you can tell whether the camera took it

Saving routing sends it to the camera, but a save is not proof it was applied. The toolbar now shows
**Awaiting sync**, **Syncing**, or **Up to date** per camera, with its name when more than one is
paired. Cameras that were offline pick the change up when they reconnect.

## Calibrate a trim saw from your desk

Measuring what a trim saw actually cuts — the length of a board going in and the length coming out —
used to mean commissioning the camera at the machine. It is now a run you start from the console and
watch on the live view.

Open **Trim saw stops** beside the live view, press **Collect**, and let the saw run. The camera
watches boards go past and captures a line for each stop it finds. Stop when you have enough, tell
it which cut size each line is, and apply. That is the calibration.

- **You only declare the sizes.** The positions are the camera's own measurement, so there is
  nothing to tape-measure and nothing to type in from the floor. Declare the sizes up front if you
  know them, or label each line after collecting if you would rather see what turns up.
- **A camera set up against a measured reference needs only one cut size** instead of two. When that
  applies, the window says so and shows the reference it was given.
- **Add to a calibration, or start over — and it asks which.** A saw that has drifted can be
  re-measured while keeping its existing sizes. Clearing them is now a single confirmation naming
  how many stops will go, rather than a question buried under settings for the run afterwards.
- **A run that nobody finished shows up on the data source**, so a collection left open does not
  quietly block the next one.
- Stops are set up by **the side of the board they sit on** — left, right, top or bottom — rather
  than by which end of the travel they are nearer.

This needs the current camera app on the device.

## Send a model to a camera, and know it landed

Cameras can now hold **several models at once**, so pushing one asks which **slot** it goes into.
The slot list comes from the camera itself, so a camera with four of them needs nothing configured
here.

Knowing what a camera is *actually* running used to be guesswork. The Models tab now lists each
model the camera reports, matched against what is stored in Zira, and says how confident that match
is — an exact fingerprint match, or only a likely one based on name and size. Where it is unsure,
you can ask it to **verify**, and it checks the file byte for byte.

Adding a model is easier too: `.pt` files upload directly, and a new dialog collects the type,
classes and where it came from at the point you add it.

## Training that explains itself

There is a **new [Training guide](/docs/ai-vision/training)** in the help centre, covering the whole
path from a labelled dataset to a model running on a camera.

The training form is now called **Train a model**, and it opens by telling you what a run does: it
learns from a dataset, it runs in the cloud, you can close the page, and it takes minutes to hours.

- Settings are grouped into **what you are making** and **how it trains**, so it is obvious which
  handful of fields matter.
- Every field has a plain-language explanation beneath it — *"How many times it reads every image.
  100 is a good start."*
- The expert controls have not gone anywhere; they are collected under **Advanced**, clearly marked
  as things a normal run never needs.
- Only one field is required: the name.

### Fewer ways to waste a run

Cameras need a converted copy of a trained model, and some of the newer conversion formats cannot
run on the camera app yet. If you pick one, the form now says so clearly before you start — instead
of leaving you to discover it when the file will not go to the device.

## Understand how a run went

- Finished runs score **each class separately** on how often the model was right when it fired and
  how much of the real thing it found — so instead of one overall number you can see exactly which
  defect still needs more examples.
- You no longer have to wait for the end. The run's charts now plot **every class as it trains**, so
  you can watch one defect catch up — or fall behind — while there is still time to do something
  about it. Switch between the two scoring styles to tell "found it at all" from "placed it
  accurately".
- The run also lists the **image variations** it generated from your pictures, so you can see what
  the model was trained against. If you need different ones, you can now set them yourself under
  **Advanced**.
- If the run trained on a merged dataset, the job shows **which datasets went into it** and how they
  were split.
- The Training tab says what it is for, and when it is empty it tells you how to start your first run
  rather than showing a bare table.

## Group your datasets

Once a camera has been running for a while, its dataset list gets long. Now you can put datasets into
**groups** — by product line, by camera, by whatever suits you.

- Add a group from a dataset's menu, under **Groups**. A dataset can be in several.
- **The list opens in sections.** A data source with groups is organised, and opening on the
  unorganised version hid that. The plain list is still one click away on the group icon.
- **Rename a group from its heading** and every dataset in it follows. Rename it onto a group that
  already exists and the two merge — you are told before you confirm.
- Datasets that are not in any group collect under **Ungrouped**, after the real groups.
- Anything you put in **archived** drops to the bottom and stays closed, so finished work is out of
  the way without being deleted.

Dataset **tags** have been retired. They did the same job as groups without any of the behaviour, so
a dataset could end up labelled twice and sorted once — groups now do it in one place.

The Datasets table was tidied up while we were there: the toolbar is one consistent set of controls
instead of five different-looking ones, notes show their text in the list rather than an icon, and
the versions count is a link you can click.

## Merging datasets, without the wizard

Combining datasets used to be a three-step wizard ending on a page of counts, duplicate reports,
per-dataset split cards and half a dozen checkboxes. It is now a single short form that asks the four
things a merge actually varies: which datasets, what to call the result, how to split it, and which
classes to keep. Everything else is simply handled.

The detail has not gone — it is in a **Merge analysis** section you can open when a merge looks
wrong, showing per-dataset and per-class counts, repeated images and box quality.

Choosing which classes to keep also works properly now: the list reorders as you press the arrows,
and the number beside each class is the class number the model will learn.

## Follow work that outlives the page

Merges, training runs and camera-file builds take minutes and keep running after you close the
window that started them — or close Zira altogether. **Background work**, the history icon above the
Datasets table, is where you follow them.

It carries a badge with the count, turns into a spinner while something is still going, and opens a
list of what each piece of work is doing and how it ended. The snackbars that start these jobs now
point you straight at it.

## Models that say what they are

- The tab now has **two clearly named lists** — what the selected camera reports it is *Running on
  this camera*, and everything *Stored for this data source*. It used to be one word above both.
- Every model can carry a **type** — what it looks for — which the Models tab groups and filters by.
  You can set it while training, or correct it afterwards from the model's menu.
- **Version** and **Format** are their own columns instead of chips crowded into the name, so long
  model names stop wrapping onto two lines.
- **Opening a training run keeps your place.** It now opens in a panel over the list rather than
  navigating the whole page away from the models you were reading.
- The row menu says what each action is for in plain terms: **Build a file for the camera**,
  **Fine-tune from this model**, **Change what it looks for**.
- Models built in a format the camera app cannot run yet are marked, so you find out in the list
  rather than at the moment you try to send one.

## Quality you can watch live

- **QA snapshot widgets update the moment an image arrives** instead of polling on a timer, so a
  dashboard on the wall keeps pace with the line.
- **A marks report** gathers what reviewers marked across a site, so you can see what is being found
  without opening each session.
- **Auto-defects** flag known defect patterns as they come in.
- A **Validation hub** brings session review together in one tab, and the site list now shows at a
  glance which data sources are stable and current.
- Snapshot widgets take dashboard date filters, so one dashboard can be pointed at any shift.

### Judge a snapshot without leaving the dashboard

Opening a snapshot from a QA Snapshots widget used to load the whole review page inside the
dashboard. It now opens a reviewer built for the job:

- **A magnifier shows the frame at the camera's own resolution** — which is what makes a nail head
  or a hairline crack readable at all. On a pallet camera the reviewer can also draw the board and
  stringer crossings and let you pick the ones a defect is in, and the mark records exactly which
  crops you meant, so the next person looks at the same ones.
- **Your filter comes with you.** The arrows step through the snapshots the widget is already
  showing, so a defect filter carries into the reviewer and keeps going past the end of a capture
  session.
- **Marks already on the image are shown**, so a frame does not get judged twice.
- **Back returns to the dashboard** instead of leaving it, with every other widget still as you
  left it.
- **Or mark without opening anything.** Put the review categories on the card with **Show mark
  buttons**, and on a single-image widget you can draw the crossings there too — tap the ones a
  defect is in, press a category, done. Five steps become two. Opening the snapshot is still the
  better call when the judgement is close, because there the image is full resolution.
- **On a tablet, pinch to zoom** — the magnifier starts off there, since a panel that size costs
  more screen than it gives back. Tap a crossing to select it, and if you do turn the magnifier on,
  press and hold to move it.

## Dashboards and agents

- **Correction fields for Data Research widgets.** Apply a calculation to every reading *before* it
  is grouped, then group, measure or filter by the result — the one place a correction can change
  what a chart's categories actually say.
- **Alerts that get out of the way.** A popup condition alert can now fade itself out after a set
  delay instead of covering the widget until the next reading arrives.
- **AI agents can work across several dashboards** at once, take a time range, and be told what
  answer style you want.
- Charts default to sensible sizes and font scales, gauges to a sensible colour, and a data table
  can be printed.

## More ways to get images in

- **Take a photo straight from your phone** when filling a form or adding images, with the camera
  opening reliably on mobile.
- **Copy into a dataset keeps your choices** — split and annotation settings persist between copies.
- **Shuffle & split** gained real options and a preview of the result before you commit, and both
  copies and merges now honour a **target resolution**.
- Dataset images show **tiny box** and **edge box** chips so weak annotations are easy to spot, and
  label visibility, class colours and thumbnails stay in step.

## On the cameras themselves

Plenty changed on the cameras too. There are no new screens to learn — it is mostly things being
more accurate, and less being lost.

**Sharper pictures, steadier counting.** Exposure, ISO and zoom can be set by hand, the preview can
be rotated and stays that way, and a camera can now sample at high resolution without slowing down
the detection it is running at the same time. A frame-rate default that had been quietly holding
cameras at 15 frames a second is fixed. Occupancy areas can be any shape rather than a rectangle,
counting triggers can be limited to a single label, and pallet cameras are better at nails, lead
boards and keeping detections inside the pallet — with calibration that stays correct when the
camera is mounted at an angle. Stations running several cameras together coordinate more reliably.

**Less gets lost.** Shift counts now survive restarts, recoveries and shift changes instead of being
dropped, and a reading the server rejects is held rather than discarded. Recordings keep their raw
and validation clips together, pick up stalled uploads again, and recover from encoding errors
rather than ending up half sent. Cameras also show their CPU and battery temperature, so a unit
working too hard is visible before it starts dropping frames.

## Around the rest of the app

- **Image review** — a new **Pallet mark** filter narrows a session to one product or batch. It shows
  up only when the session actually contains marked pallets.
- **Pie charts** can now show a **legend**, for when the slice labels are too tight to read.
- **Indicator widgets** read **True** or **False** on a yes/no field instead of a raw value.
- **Chart tooltips** hold still, so the **view entries** button stops sliding away as you reach for
  it — and sorted Data Research charts now draw their axis correctly, embedded ones included.
- **The tasks board** scrolls to the bottom, loads more reliably, and opens the card you clicked.
- **Schedules** can be added at site level, and straight from an AI widget.
- The app **remembers more between visits**, so returning to a page you were just on is quicker.

## Read more

- [Connect Zira to Claude](/docs/ai-assistance/connect-to-claude) — new guide
- [Home AI Assistant](/docs/ai-assistance/home-ai-assistant) — opening it from the left menu
- [Routing](/docs/ai-vision/routing) — actuators, webhooks and device sync
- [Trim Saw Stops](/docs/ai-vision/trim-saw-stops) — calibrating a trim saw
- [Models](/docs/ai-vision/models) — rewritten
- [Training](/docs/ai-vision/training) — new guide
- [Datasets](/docs/ai-vision/datasets) — groups, merging and background work
- [QA Snapshots](/docs/dashboards/qa-snapshots-widget) — live updates, marks and reviewing from the dashboard
- [Image Review](/docs/data-sources/image-review) — pallet mark filter


Connect a site to Claude and ask about your operation in plain language. Claude reads your live data, and acts only when you let it.

Connect Zira to Claude

Keep the feed or a dashboard as your home page, and open the assistant any time from the sparkles icon in the left menu.

The AI Assistant, from any page

Routing rules can now drive a relay — an andon light, or a PLC contact to stop, start or eject — and call a webhook.

Let a rule switch something on the floor

Measure cut lengths by watching boards go past on the live view. No tape measure, no standing at the machine.

Calibrate a trim saw from your desk

Push to a named slot on cameras that hold several, and see exactly which model each one is really running.

Send a model to a camera, and know it landed

Every setting says what it does in plain words, and each class is scored as the run goes — so you know which defect needs more images.

Training that explains itself

QA snapshots update the moment an image arrives, and a marks report gathers what reviewers found across a site.

Quality you can watch live

Routing rules can now switch an andon light or drive a machine PLC, a trim saw calibrates itself while you watch from your desk, models push to a named slot and prove they landed, and training explains every setting in plain language. You can also connect a site to Claude and ask about your operation out loud, and open the AI Assistant from any page without making it your home page.

Act on What the Camera Sees


You can now connect Zira to Claude and simply ask about your operation. Claude reads your live Zira
data to answer, and — when you allow it — does the work in Zira for you.

No more copying readings into a chat window or pasting answers back out.

## Ask in plain language

Once connected, ask the kind of question you would ask a supervisor:

- *"Why did the trim saw stall last night? Check its readings around 22:00."*
- *"What was yesterday's output on line 3?"*
- *"Show me the open tasks for the Rehovot plant."*

Claude finds the right data source by name, looks up the readings, and answers. You do not need to
know which metric or device to point it at.

## Let Claude do the work

Give a connection permission to act and Claude can open and update tasks, record readings, and post
an update to a channel — always asking you before it does.

Prefer to start carefully? Choose **Look only** when you connect, and Claude can answer questions
but cannot change anything.

## Connecting takes two clicks

Open your profile picture and choose **Connect to Claude**. Pick the site you want Claude to work
with, choose what it may do, and press **Connect**. Copy the single line Zira gives you, paste it
into Claude, and you are done.

Each connection covers one site, so connect each site you want Claude to reach.

![Connecting a site to Claude: pick the site, choose what Claude may do, press Connect.](/changelogs/connect-zira-to-claude/connect/01-connect-panel.png)

## Claude can read the guides too

There is a second, separate connection that gives Claude the Zira help center — the product guides,
the API reference, and these release notes. It is public and read-only, so it needs no setup beyond
the address. Teams that use both get answers grounded in how Zira actually works, not guesswork.

## Getting started

- [Connect Zira to Claude](/developers/how-to-guides/connect-zira-to-claude)
- [Connect the documentation](/developers/how-to-guides/connect-the-documentation-mcp-server)


Pick a site, press Connect, paste one line into Claude. No keys to manage by hand.

Connect in two clicks

"Why did the trim saw stall last night?" — Claude looks it up in your live data.

Ask about your plant

Open a task, post readings, or share an update to a channel — Claude asks you first, every time.

Let Claude do the work

Choose a read-only connection and Claude can answer questions but change nothing.

Look-only when you prefer

A second connection lets Claude search Zira's guides and API reference while it helps.

Claude knows the product too

Connect Zira to Claude and ask about your operation in plain language — Claude reads your live production data, and can open tasks and post updates for you.


This release opens Zira's documentation to the world. Guides, the developer reference, and release
news now live on their own public site — no sign-in, readable by anyone you send a link to, and
indexed by search engines. Alongside it, the API reference is finally complete: every endpoint shows
what to send, what comes back, and what to do when something fails. Inside the product, device
settings now come from the device itself, each data source gets its own Training tab, and routing
rules can call a webhook.

## 📚 A public help center at help.zira.us

Everything that used to sit behind a login now has a public home.

- **Guides, developer docs, and What's New** are all served from [help.zira.us](/docs), and every
  page works without an account. Send a colleague — or a prospective customer — a direct link and it
  simply opens.
- **Search engines and AI assistants can read it.** The old in-app help was invisible to both, so
  people looking for an answer never found ours. Now they will.
- **Existing links keep working.** Old `/docs`, `/developers` and `/whats-new` addresses forward to
  the new site automatically, so nothing you have shared or bookmarked breaks.
- **Support stays private.** Opening a support ticket still happens inside Zira, where we know who
  you are.

## 🔌 A complete API reference

If your team connects Zira to another system, the reference is now something you can build against
rather than guess at.

- **Every endpoint documents its response.** Each one shows the real body you get back, with a field
  table explaining what each value means and which values can be empty. Previously most endpoints
  described only what to send.
- **Request and response examples are copy-ready**, with a one-click copy button.
- **A familiar layout.** Each endpoint is laid out in tabs — description, parameters, authorization,
  headers, body, response — so it reads the way an API client does.
- **Errors are documented once, properly.** Every endpoint lists the statuses it can return and what
  each one actually means, so you can handle failures instead of discovering them.
- **Corrections throughout.** A number of endpoints listed fields that no longer exist, marked
  optional fields as required, or omitted parameters you cannot work without. Those are fixed.

## 🤖 Ask about the endpoint you are reading

An assistant now sits beside the reference behind a button that stays with you as you scroll. Ask how
to authenticate, what a parameter does, or for a snippet in your language, and it answers from the
documentation for the endpoint you are on — so the field names and addresses match what the system
really accepts.

## ⚙️ Device settings come from the device

Camera and vision device settings used to be described in two places: once on the device and once in
Zira. When a device gained a new option, Zira did not know about it.

Now the device publishes its own settings, and the dialog renders exactly that. You see the options
your device actually supports, read-only values are clearly shown as read-only, and a device can gain
a setting without waiting for a Zira release.

## 🎓 Training, per data source and across the account

Each data source now has its own **Training** tab showing the training runs it has been part of,
along with their results and metrics. Administrators also get a single page listing every training
job in the account, so a run that needs attention is easy to find.

## 🪝 Routing rules can call a webhook

A routing rule can now send its decision to another system as it happens, so downstream tools react
immediately instead of waiting for the next poll.

## 🗂️ Smaller things that add up

- **Sharper image review.** Filter a session by the classes actually present in it, and jump between
  sessions without losing your place.
- **Smarter agents.** Agents now run on the current generation of AI models only, with faster,
  leaner requests. What used to be labelled "GPT" is now simply "AI".
- **QA Snapshots.** Mark an image straight from its card, with clearer labels on the widget.
- **Dashboards.** More control over table columns and bar colours, and pie chart colours follow the
  palette you expect.

## 🐛 Fixes

- Hover states across the app no longer come out too dark in dark mode.
- Editing an agent's configuration no longer shows a stale copy of what you just saved.
- Deleting all of a label file's labels now removes the empty file instead of leaving it behind.


Guides, the API reference, and release news all live at help.zira.us — no sign-in, and shareable with anyone.

Help and developer docs, now public

Every endpoint now shows its parameters, headers, request body and a real response example you can copy.

A complete API reference

An assistant sits beside the reference and answers questions about the endpoint you are reading, with working code.

Ask about any endpoint

Camera and device settings dialogs now show exactly the options that device supports, so nothing is stale or missing.

Device settings from the device itself

A Training tab on each data source shows its training runs and results, with a single place to watch every job.

Training, per data source

Send a routing decision straight to another system as it happens, instead of polling for it.

Routing rules can call a webhook

This release opens Zira's documentation to the world. Guides, the developer reference, and release news now live on their own public site — no sign in, readable by anyone you send a link to, and indexed by search engines

A Public Help Center, and a Complete API Reference


This release is about getting answers and getting through work faster. The home AI assistant can now help you connect your own systems to Zira — ask it how to pull readings or create a task programmatically and it replies with the exact request and working code you can copy straight into your project. Image review gains sharper filtering, a much deeper zoom, and random sampling for quick spot-checks. QA Snapshots can display annotated images, a merged dataset can go straight into training, and small touches across dashboards and the Vision Hub smooth out everyday work.

## 🧑‍💻 Ask the assistant how to connect your systems

If your team pulls data out of Zira or pushes data in from another system, you can now just ask the assistant about it.

- Ask something like **"how do I get readings from a data source using the API?"** and you get the exact request to make, along with a working example you can copy and run.
- Ask for the language you actually work in — **cURL, Python, or JavaScript** — and you get a complete, ready-to-run snippet, not a sketch. Every example is formatted as a proper code block with a one-click copy button.
- One-tap buttons under each answer let you switch languages, get the parameters explained, or find out how to create an API key — no retyping.
- Answers link straight to the matching page in the [developer reference](/developers/api-reference/get-readings) so you can dig deeper.

Everything the assistant tells you comes from Zira's published integration reference, so the field names, addresses, and options match what the system actually accepts.

## 💬 Follow-ups that remember what you were asking about

The assistant now holds on to the subject of a conversation. Ask about reading data, then simply say **"can you give me a Python snippet?"** — it knows you still mean reading data, and answers directly instead of asking you to explain again. Move on to a different topic and it follows you there.

## 🔍 A sharper image review

- **Combine filters freely.** Pick several options within a filter and several filters at once to narrow a session down to exactly the images you want to look at.
- **Filter by catalog code** alongside the existing defect and disposition filters.
- **Zoom up to 20x** to inspect fine detail, and the mark you are working on now stays clearly highlighted as you look around.
- **Select a random sample** of images from across the whole filtered session — a quick way to spot-check quality without paging through everything.
- More images load per page by default, so there is less clicking to get through a session.

## 🖼️ Annotated images in QA Snapshots

The QA Snapshots widget can now draw detection overlays directly on its images. Turn on **Show annotations** when you set the widget up, and optionally limit the overlays to defects only so a wall of results stays readable at a glance.

## 🚀 From merged dataset straight to training

When you merge datasets into a new training set, you can now start training in the same step. The training settings are pre-filled from the merge, so a version that used to need a second round of setup is ready to go on its own.

## 🗂️ Smaller things that add up

- **Open in a new tab.** Validation sessions and datasets now open in a new tab from their menu, or with a middle click — handy when you want to compare two side by side.
- **Faster validation lists.** The validation list appears immediately and refreshes quietly in the background.
- **Hidden widgets are marked.** While editing a dashboard layout, any widget that is hidden on a particular screen size now shows a small badge saying so, so nothing disappears unexpectedly.
- **Cleaner widget defaults.** New KPI, Data Streaming, and Indicator widgets start with a centered, comfortably sized label, and gauges sit properly centered.
- **What's new in the app bar.** Release news now shows as a small chip next to your profile picture, and lets you know when a new version is ready to load.

## 🐛 Fixes

- Video validation now keeps the correct raw or validated view when you move between videos.
- Deleting labels from the label configuration works reliably again.
- Chart widgets use their intended default line colors, and gauge options behave correctly.


Ask how to read or send data programmatically and get the exact request plus a ready-to-run snippet in cURL, Python, or JavaScript.

Ask the assistant for integration code

Say "now in Python" or "what about the parameters" and the assistant stays on the same topic instead of starting over.

Follow-ups that keep their place

Combine several filters at once, filter by catalog code, and zoom to 20x to inspect the fine detail.

Sharper image review filtering

Select a random set of images from across a whole filtered session — a fast way to spot-check quality.

Pull a random sample

Show detection overlays right on the widget's images, with an option to display only the defects.

Annotated QA Snapshots

Start training the moment a merged dataset is ready, without going back to set it up by hand.

Merge a dataset and train in one step

Release news now appears as a small chip next to your profile picture, so it is easy to find and easy to ignore.

What's new, right in the app bar

The home AI assistant now answers questions about connecting your own systems to Zira with real, copy-ready code — and remembers what you were asking about, so a follow-up like "give me a Python snippet" just works. Image review gets sharper filtering, a much deeper zoom, and a way to pull a random sample from a session. QA Snapshots can now show annotated images, merged datasets can go straight into training, and validation sessions and datasets open in a new tab.

Ask the Assistant for Code, and Review Images Faster


This release makes Zira quicker to drive from the first click. The home AI assistant now helps you before you've even chosen a site, adds a "/" command menu for getting anywhere fast, and lets you change sites mid-conversation. On the QA side, the Vision Hub pulls validation sessions from every site into a single review surface with always-on filters, shows what's recording as it happens, lets you record or test devices in bulk, turns the activity feed into a scannable table, and adds new ways to merge and generate training data — and image-by-image review is smoother — keeping your view settings as you page through a session, and now showing every captured side of a pallet in one place. It also puts hands-on device control in the hub: open a device's settings in one dialog and drag its masks, zones, and rulers right over the live view — plus new stringer controls when you build pallet training data.

## ⌨️ A new "/" command menu in the assistant

- Type **/** in the assistant composer — or tap the **/** button — to open a command menu.
- **Jump straight to a recent dashboard or data source**, or start a common action: build a dashboard, create a task, form, or data source, or browse your dashboards, sites, tasks, and companies.
- Ask **what the assistant can do**, and **switch your AI provider** (OpenAI or Claude) right from the menu — the provider choice now lives here instead of the toolbar.

![The "/" command menu open in the assistant composer, showing Create & build actions and recent dashboards to jump to](/changelogs/slash-commands-menu/slash-menu-overview.png)

Keep typing after the **/** to filter instantly — type part of a name and the menu narrows to the matching dashboards and data sources, so you can open a specific item in a couple of keystrokes.

![Typing "/band" filters the menu down to matching dashboards and data sources](/changelogs/slash-commands-menu/slash-menu-filter-band.png)

## 🧭 Help before you pick a site — and switch sites from chat

- The assistant is now useful **before you select a site**: it's aware of your **companies** and your **open tasks across every site**, so you can ask basic questions right away.
- When something needs a specific site, the assistant offers a one-tap **"Select a site"** — and once you choose, it **automatically continues** the request you started.
- Just say **"switch to \<site\>"** to change your working site without leaving the conversation.
- Questions like **"how many tasks do I have?"** now return **accurate totals** across your sites, with a gentle nudge to focus on a single site when that helps.

## 🗂️ Review validation sessions across every site

- A new **Validation** tab in the Vision Hub gathers the **latest validation session for each device**, grouped by site, so you no longer have to open each site on its own.
- Every row shows **status, val/raw/marks counts, and the session date** — click the date to open the full review in a new tab.
- **Always-on filters** sit in a toolbar right below the tabs — narrow by session window and by whether a device has a recent session, has **marks found**, is **not yet reviewed**, or has **no activity** — so you can jump straight to what needs attention.
- **Select or clear all** in-view rows with one checkbox, and **expand or collapse every site** in a single tap.
- Expand any site to load its most recent sessions on demand, so the page stays fast even with many sites.

## 🔴 See what's recording — and record or test from the hub

- Devices that are **recording right now** show a clear live indicator on their row, no matter who started the recording.
- Need footage from a device that's missing a recent session? **Start a recording — for one device or several at once — without leaving the page.**
- **Run a test on selected devices in bulk** and watch progress right there, so you can confirm coverage across many devices at once.
- This makes it quick to spot gaps and fill them in, so every device stays covered.

## 🎛️ Tune a device — and place its overlays — right from the hub

- Open a **Device settings** dialog for a device and adjust its configuration in one place, organized into clear tabs instead of a long wall of options.
- Settings are grouped into tabs — **Camera, Sampling, Detection, Measurement, Validation, Overlays** and more — and stay **synced with the device**, so you always see what's actually running.
- A new **Overlays** tab lets you **drag the pieces that matter directly over the picture** — the detection **mask**, **ruler lines**, and the **zero point** — and line them up against a **live view, a recorded video, or a still image**.
- Changes are sent to the device for you, so setup that used to mean typing raw pixel values becomes a hands-on, visual adjustment.
- Works for both vision and pallet devices, each with the settings that apply to it.

![The Device settings dialog, with Camera settings open and tabs for Sampling, Detection, Measurement, Overlays and more down the side](/docs/data-sources/remote-control-05.png)

![The Overlays tab, dragging a detection mask, ruler line, and zero point directly over a recorded video frame of a pallet](/docs/data-sources/remote-control-06.png)

## 📋 A cleaner Vision Hub Feed

- The **Feed** tab now renders as a **compact table** instead of social-style cards, so a wall of device alerts becomes easy to scan.
- Each row lines up **site, device, message, and time**, with actions right where you expect them.
- When you handle many sites at once, you can tell rows apart at a glance.

## 🧩 New dataset merge, generation, and pallet-building tools

- **Merge datasets** into a single set — combine sources into one dataset (or a new version), with options to skip empty labels and clean up duplicate boxes as you go.
- **Generate focused training datasets** from a labeled set, with more control over cropping, image sizing, and output quality.
- **Build pallet training data with new stringer controls:** when you create a pallet from an image, you can now set its **stringer configuration**, so the generated boxes match how the pallet is actually built.
- Together, these tools make it faster to shape your raw images into training-ready data.

## 🖼️ Smoother image-by-image review

- **Step through a session faster:** prev and next are now **floating buttons right over the image**, so you can page through snapshots without moving your eyes off the picture.
- **Your view stays put:** zoom, pan, the RAW/VAL toggle, and the measurement overlay now **carry over from one image to the next**, so you don't have to re-set them on every snapshot.
- The view controls stay together on **a single, tidy row**, keeping the review surface clean while you focus on the details.

## 🪵 Review every side of a pallet

- Some lines capture a pallet from **more than one side at the same time** — top, left, and right. Image Review now brings **all of the captured sides together**, instead of showing only the top.
- The details panel is now a clean set of collapsible sections — a **Pallet** summary plus **one section per side** — each showing its result right in the header so you can scan at a glance and open just the side you care about.
- Open a side to see **its own snapshot** alongside its inspection results. If a side wasn't captured, the section simply says so instead of showing a broken image.
- **AI inspection** and your saved **Marks** sit in the same panel, so everything about the current pallet is in one scrollable place.

![Image Review showing a pallet with collapsible Pallet, Top, and Left sections — each with its own result — plus the snapshot and inspection details for the selected side](/docs/data-sources/image-validation/01-image-validation-overview.png)

## ✨ Why this matters

- The assistant gets you moving from the first screen — answers before a site is chosen, and everything a keystroke away with "/".
- QA review moves from site-by-site to a **single cross-site surface** with filters that surface exactly what needs attention, so nothing slips through the cracks.
- Live recording status, one-click recording, and bulk testing keep your coverage complete without hunting for the right device.
- Tuning a device and placing its overlays visually, right over the live view, turns setup into a quick, hands-on adjustment.
- Image review flows image to image with your settings intact, so long sessions go faster.
- Every captured side of a pallet is in one place, so a defect on any face is easy to confirm.
- Better dataset tooling — now including stringer-aware pallet building — shortens the path from raw images to a model you can trust.

## 📚 Related guides

- ⌨️ [Home AI Assistant](/docs/ai-assistance/home-ai-assistant)
- 🗂️ [Validation](/docs/ai-vision/validation)
- 🎛️ [Remote Control](/docs/ai-vision/remote-control)
- 🖼️ [Image Review](/docs/data-sources/image-review)
- 📋 [Feed](/docs/feed/channel-feed)
- 🧩 [Datasets](/docs/ai-vision/datasets) and [Dataset page](/docs/ai-vision/dataset-page)
- 🧰 [Vision overview](/docs/ai-vision/overview)


Type / in the assistant to jump to a dashboard or data source, browse, create, or switch AI provider — all from the composer.

A new "/" command menu

The assistant now knows your companies and open tasks across sites, and can select or switch your site right in chat.

Help before you pick a site

The new Validation tab gathers the latest validation session for each device, grouped by site, with always-on filters to zero in on what needs attention.

Review sessions across every site

See which devices are recording live, and start recording or run tests on one device or many at once, right from the page.

Record and test from the hub

Open a device's settings in one dialog and drag its masks, counting zones, and rulers right over the live view.

Tune a device from the hub

Merge datasets into one, generate focused training sets, and build pallet training data with new stringer controls.

Better dataset tools

See each captured side — top, left, right — with its own snapshot and results, and step through images with your zoom, pan, and overlays kept.

Review every side of a pallet

The home AI assistant now helps before you even pick a site, adds a "/" command menu, and lets you switch sites right from chat — while the Vision Hub brings validation sessions from every site into one place with powerful filters, live recording, and bulk testing, a cleaner Feed table, and smoother image-by-image review that now shows every captured side of a pallet. This release also adds hands-on device control — tune a device's settings and drag its overlays right over the live view, plus richer dataset tools for merging, generating, and building pallet training data.

A Smarter Home Assistant, Cross-Site QA Review, and Hands-On Device Control


This release puts you in charge of how Zira opens and makes the AI Assistant a first-class way to work — beside your dashboards, in more natural conversations, and on the AI provider you prefer. It also brings the biggest set of Vision improvements yet, so the whole path from data to results keeps getting faster and clearer.

## 🧭 Set your Work mode

- The **Home Page** preference is now **Work mode** — a clearer way to choose how Zira opens when you sign in.
- Pick from three focused cards in **Preferences → Work mode**:
  - **AI Assistant** — start every day in an assistant that can chat, navigate, and build from one place.
  - **Dashboard** — open straight into the numbers you watch.
  - **Feed** — stay on top of activity and updates.
- The redesigned selector shows a **Current** badge on your active mode and a clear check on the one you're picking, so the choice is easy to see at a glance.

![Work mode in Preferences](/changelogs/vision-workflow-upgrades/work-mode/Screenshot%202026-07-02%20at%2017.12.08.png)

![Choosing between AI Assistant, Dashboard, and Feed](/changelogs/vision-workflow-upgrades/work-mode/Screenshot%202026-07-02%20at%2017.12.20.png)

## 🤝 An assistant beside your dashboards

- Your dashboard now has a **companion panel** powered by the same Home AI Assistant, right next to the live view.
- Ask questions about **the dashboard you're looking at** — the assistant reads its loaded data to answer.
- Make changes in place — the assistant's edits flow into the dashboard you're already working on, keeping you focused on one screen.

![The AI Assistant panel docked beside a live dashboard](/changelogs/vision-workflow-upgrades/dashboard-companion/Screenshot%202026-07-02%20at%2017.11.39.png)

## 💬 More natural conversations

- Ask follow-up questions like counts or details and get a **direct answer from what's already on screen**, instead of the assistant re-listing everything.
- **Change direction mid-chat** — say "create a form" partway through and the assistant switches to the new task smoothly.
- Answers now share **one clean, compact layout**, so lists, choices, and suggestions all look consistent and are quicker to scan.
- If you step away from a multi-step flow, it collapses into a tidy **"paused / Resume"** summary you can pick back up anytime.

## 🔀 Choose your AI provider

- Pick the AI provider that powers your assistant — **OpenAI or Claude** — right from the composer.
- Gentle brand icons make the current choice clear without cluttering the chat.

## 👁️ Big upgrades across the Vision workflow

From building datasets to annotating images, training models, detecting defects, and reviewing results, each step of the Vision workflow is now faster, clearer, and better connected.

## 🗂️ Stronger dataset tools

- Datasets now support **versions**, so you can keep a clear history as your data grows.
- Generate **derived datasets** from existing ones to spin up focused training sets quickly.
- **Shuffle images** between folders to balance and organize your data.
- Add **tags and notes** to keep datasets easy to find and understand later.
- **Export a dataset to a ZIP** with a progress indicator, so large exports never leave you guessing.
- Galleries are now **paginated** and easier to browse, even with large datasets.

## ✏️ Faster annotation

- **Bulk label editing** lets you add, replace, or remove labels across multiple images at once.
- New **keyboard shortcuts** make moving through images and marking them much faster.
- The new **Describe to Annotate** action runs auto-annotation from a plain-language description of what you want boxed — with a clearer, simpler setup — giving you a strong starting point to refine.
- **Duplicate-box tooling** speeds up repetitive marking when objects repeat across a frame.

![Annotation page with bulk labels and the Describe to Annotate action](/changelogs/vision-workflow-upgrades/annotation-bulk-labels-ask-zira.png)

## 🧠 Train and export models

- Model training now offers **more setup options**, so you can tune a run to your needs.
- Export a trained model, including **TFLite**, to run detection directly on a device.

![Model training dialog with export options](/changelogs/vision-workflow-upgrades/model-training-and-export.png)

## 🔍 Automatic defect detection

- Run **auto-detect on a session** to surface defects without marking everything by hand.
- Detect **structural defects** on boards and pallets as part of the validation flow.
- See results come to life with **live detection overlays and segmentation masks**, plus overhang dimension lines in review.

![Validation session with auto-detect overlays and segmentation masks](/changelogs/vision-workflow-upgrades/auto-detect-overlays-and-masks.png)

## 🖼️ Clearer image and video review

- Image review can now show the **full inspection result** in a collapsible view when you want the details.
- Switch between **raw and validation** views to compare exactly what the device saw.
- Sessions now follow the **actual recording time**, so the order matches what happened on the line.
- Video review keeps the current moment in view with **auto-scroll**, and you can **create a task** straight from a video mark.
- The review sidebar is now **cleaner and more consistent** — Marks, Inspection, and Auto Detect share one tidy card design.
- **Collapse any card** to focus on what matters, and the sidebar now scrolls as a single column instead of each card scrolling on its own.
- Common actions like **Run Detection** sit right where you need them at the bottom of the Auto Detect card.

## 🎥 Bulk video recording

- Capture **several validation videos in one sequential flow**, so collecting evidence across a session takes far fewer steps.
- Recording defaults were refined to get you to a usable clip faster.

## 🧭 Guided Vision in the Home Assistant

- The **Home Assistant** can now guide you through the full Vision workflow, from setting things up to reviewing results.
- Your assistant session is kept in the page address, so you can refresh or share a link and pick up right where you left off.

![Home Assistant guiding the Vision workflow](/changelogs/vision-workflow-upgrades/home-assistant-vision-workflow.png)

## ✨ Why this matters

- The whole Vision journey — data, annotation, training, detection, and review — now feels like one connected flow.
- Common tasks take fewer clicks, and AI can do more of the heavy lifting along the way.
- Teams can move from raw images to trustworthy QA results with far less manual effort.

## 📚 Related guides

- 🧭 [Choose your Work mode](/docs/getting-started/profile-preferences/home-page)
- 🤖 [Home AI Assistant](/docs/ai-assistance/home-ai-assistant) and [Dashboard AI Assistant](/docs/ai-assistance/dashboard-ai-assistant)
- 🗂️ [Datasets](/docs/ai-vision/datasets) and [Dataset page](/docs/ai-vision/dataset-page)
- ✏️ [Image annotation](/docs/ai-vision/image-annotation)
- 🧠 [Models](/docs/ai-vision/models)
- 🔍 [Validation](/docs/ai-vision/validation)
- 🖼️ [Image review](/docs/data-sources/image-review) and [Video review](/docs/data-sources/video-review)
- 🎥 [Remote control](/docs/ai-vision/remote-control)
- 🧰 [Vision overview](/docs/ai-vision/overview)


Choose how Zira opens when you sign in — an AI Assistant that can chat, navigate, and build, a dashboard, or your feed.

Set your Work mode

A companion panel reads your live dashboard and helps you ask questions and make edits in place.

An assistant beside your dashboards

Get answers from context, switch tasks mid-chat, and pick up paused flows right where you left off.

More natural conversations

Pick OpenAI or Claude right from the assistant composer.

Choose your AI provider

Stronger datasets, faster annotation, model training and export, and automatic defect detection.

Big Vision workflow upgrades

Inspect full results, switch raw and validation views, and record validation videos in bulk.

Clearer image & video review

Set up training with more control and export trained models, including TFLite, for on-device use.

Train and export models

Choose how Zira opens with the new Work mode — including an AI Assistant home that can chat, navigate, and build for you — now with an assistant beside your dashboards, more natural conversations, and your choice of AI provider, plus the biggest set of Vision workflow upgrades yet.

A New Work Mode, a Smarter Assistant, and Big Vision Upgrades


Dashboards can now include a dedicated **QA Snapshots** widget for recent Vision QA images. The goal is simple: help teams spot the latest review evidence quickly without leaving the dashboard too early.

## Recent QA images can live directly on the dashboard

- The new **QA Snapshots** widget shows the latest images from one selected Vision data source.
- Teams can use it to keep recent QA evidence visible beside monitoring, routing, or live data widgets.
- Each item shows the image time and lightweight review context so the widget stays easy to scan.

```text
Screenshot placeholder
Dashboard showing the QA Snapshots widget with recent images.
Expected file: /public/changelogs/qa-snapshots-widget/01-widget-overview.png
```

## Teams can choose the display style that fits the dashboard

- Use **List** when the dashboard needs a tighter layout.
- Use **Gallery** when visual scanning matters more.
- Image size and text size can also be adjusted so the widget fits different dashboard styles.

## Full review stays close to the dashboard

- Clicking a snapshot opens the full image review experience in a full-screen dashboard view.
- People can inspect the image, review marks, and move through the session.
- **Back to dashboard** makes it easy to return without losing the working context.

```text
Screenshot placeholder
QA Snapshot Review opened from the dashboard with the back action visible.
Expected file: /public/changelogs/qa-snapshots-widget/02-review-from-dashboard.png
```

## AI can help create it

If you use **AI Dashboard Builder**, you can ask for this widget directly by naming the data source and the display style you want.

Examples:

- create a QA snapshots widget using AirDroid
- add a QA snapshots widget in gallery mode
- show the latest 4 QA snapshots for Zira-R&D

Related guides:

- [QA Snapshots](/docs/dashboards/qa-snapshots-widget)
- [AI Dashboard Builder](/docs/dashboards/ai-dashboard-builder)
- [Image Review](/docs/data-sources/image-review)


Dashboards can now show recent QA snapshot images from a Vision data source, with quick review context and an easy path into full image review.

QA Snapshots Widget


This update brings a clearer review experience for both QA session images and validation videos. The goal is simple: make it easier to move through a session, understand what the device saw, and leave helpful review marks without losing context.

## Image Review is ready for QA sessions

- QA session images now open in a dedicated review page.
- The page keeps the current image in the center, the marks and inspection summary on the left, and the rest of the session on the right.
- Detection overlays, inspection details, zoom, download, and image-by-image navigation are all available in one place.

![Image Review page](/changelogs/image-and-video-review/01-image-review.png)

## Video Review stays focused on recorded evidence

- Validation videos continue to use a dedicated review page built for playback and mark timing.
- The session gallery stays visible while you review.
- Marks remain attached to the exact moment in the clip, which makes follow-up conversations much easier.

![Video Review page](/changelogs/image-and-video-review/02-video-review.png)

## Why this matters

- Teams can review one full session without jumping back and forth between pages.
- QA feedback is easier to capture in a way that others can understand later.
- Image and video review now feel more consistent, which makes training and support simpler.


QA review pages now make it easier to inspect session images and validation videos, follow the full session, and leave clear review marks.

Image Review and Video Review


When something feels unfamiliar, you can now ask the AI Assistant for help directly from the Help experience.

## What you can ask

- Where did a feature move?
- How do I create a form, data source, or dashboard?
- What changed in the latest release?
- Where can I find docs or support?

## How answers work

The assistant uses Zira product docs and update notes to give guidance, then shows related sources when there is more to read.

It is meant to be the fastest first step when you are not sure where to look.



AI Assistant can answer questions about Zira and point you toward the right docs, support options, or product updates.

Help and Support with AI Assistant


Creating new work in Zira can now start with a conversation instead of a blank setup screen.

## What you can create

- **Forms** can start from a plain-language description of the information you want to collect.
- **Data sources** can start with AI-drafted titles, descriptions, and fields.
- **Dashboards** can start from the question you want the dashboard to answer.

## How it helps

AI Chat gives you a stronger first draft, then keeps you in control before anything important is applied.

- You describe what you need in everyday language.
- Zira drafts the structure for you.
- You review the result and adjust it before saving or applying it.

## Why this matters

This release makes setup work faster for new users and less repetitive for experienced teams. Instead of building every field, widget, or schema from scratch, you can start from a draft and spend your time shaping the result.



AI Chat can now help you create forms, data sources, and dashboards from a plain-language request, giving you a faster first draft to review and refine.

Create Forms, Data Sources, and Dashboards with AI Chat


Home can now open directly to a dashboard instead of always starting from the feed.

## What changed

You can choose a dashboard as your custom home page. This is helpful when your team starts the day from a live operations dashboard, KPI view, shift overview, or any other dashboard that matters more than the default feed.

## Why it helps

- Teams can start from the same operational view.
- Managers can open directly to the dashboard they check most often.
- Individual users can make Zira feel closer to the way they work.

## Where to find it

Open account settings or preferences, then choose the dashboard you want as your home page.



You can now choose a dashboard as your Zira home page, so the first screen opens directly where your work begins.

Set a Dashboard as Your Home Page


Zira can now help turn messy real-world input into a structured form draft.

## What changed

- Upload an image, such as a paper form, note, or work order.
- Paste free text when the information already exists in a message, report, or field note.
- Let Zira draft the form values from that input.
- Review the draft before submitting.

## Why it helps

This is useful when the information already exists somewhere else and you do not want to copy it field by field.

The review step is important: AI helps with the first pass, but you still decide what should be submitted.



Forms can now be filled from an uploaded image or pasted text, with a review step before the AI draft is applied.

Fill Forms from an Image or Free Text


The Help Center is now the starting point when you want to understand Zira, find support, or catch up on product changes.

## What is inside

- **Docs** for learning how Zira works.
- **What's New** for product updates and release notes.
- **Support** for getting help from the Zira team.
- **AI Help Assistant** for asking questions in plain language.

## Why it matters

This release changes several familiar UI areas, so Help is designed to be close by. If navigation, forms, dashboards, chat, or AI flows feel different, the Help Center gives you a place to orient yourself without leaving the product.



The new Help Center brings docs, product updates, release information, support paths, and AI help into one place.

New Help Center


Navigation has been refreshed so the main workspace feels cleaner and common destinations are easier to scan.

## What looks different

- The sidebar now starts from a cleaner, more compact layout.
- Search lives inline in the sidebar.
- Some icons and entry points moved.
- Notifications and Help are easier to reach from the app frame.
- The top app bar is calmer, with fewer competing actions.

## What to expect

Things may look a little different the first time you open Zira after the release. The main areas are still there, but the path to them is cleaner and more consistent.

If something feels unfamiliar, open Help from the app frame or use the Help Center to search the docs.



The Zira sidebar is cleaner, search is now inline, and several icons moved as part of the refreshed navigation experience.

Refreshed Navigation


The chat experience has been refreshed across the drawer, conversation view, and new-chat flow.

## What changed

- Creating private and group conversations now feels more guided.
- The chat drawer uses a cleaner search and navigation style.
- Mini chat opens as a centered preview when you need a quick conversation surface.
- Sending, editing, and seeing recent messages should feel more immediate.

## Why it helps

Starting a conversation should feel lightweight, and returning to active conversations should feel predictable. This refresh makes chat easier to scan and smoother to use alongside the rest of Zira.



Chat creation and messaging have been refreshed so conversations are easier to start, preview, and manage.

Updated Chat Experience


This release brings together the biggest Zira Web improvements in one place: AI can help you create faster, navigation is cleaner, Help is easier to use, and Vision workflows now connect more clearly from setup through validation.

## Create faster with AI

AI is now a more practical starting point when you need to build something instead of starting from a blank screen.

- **AI Dashboard Builder** can draft widgets, layout, and structure from a plain-language request.
- **AI Form Builder** can generate fields, options, validation, and calculated structure for forms and custom data sources.
- **AI Help Assistant** can answer product questions and point you to the right guides without making you search first.

![AI Dashboard Builder preview showing the proposed change before apply.](/docs/dashboards/ai-dashboard-builder/05-ai-draft-preview.png)

![AI Form Builder showing generated fields and preview.](/docs/ai/ai-form-builder/03-ai-builder-result.png)

Related guides:

- [AI Dashboard Builder](/docs/dashboards/ai-dashboard-builder)
- [Build Forms with AI](/docs/forms/build-forms-with-ai)
- [AI Help Assistant](/docs/ai-assistance/help-assistant)
- [Create a Data Source](/docs/data-sources/create-a-data-source)

## Navigation is calmer and easier to scan

The main app frame has been refreshed so common actions are easier to find without feeling crowded.

- The left sidebar uses a cleaner layout.
- Search now sits inline inside the sidebar list where it is needed.
- The top app bar is simpler, with page search and account actions easier to spot.
- Help and other common destinations are easier to reach when something looks unfamiliar.

![Updated app navigation with sidebar and top bar visible.](/docs/navigation/app-bar-and-sidebar/03-vision-sidebar-and-top-bar.png)

Related guides:

- [App Bar and Sidebar](/docs/navigation/app-bar-and-sidebar)
- [Change Your Preferences](/docs/getting-started/profile-preferences/preferences)

## Home and Help now work better as starting points

Zira is easier to personalize and easier to learn when you first arrive.

- You can set a dashboard as your personal home page so Zira opens directly into the workspace that matters most to you.
- The Help Center gives you one place for docs, product updates, support paths, and the AI Help Assistant.

![Home Page picker with dashboard selection.](/docs/getting-started/profile-preferences/home-page/02-home-page-picker.png)

![AI Help Assistant showing an answer with related docs.](/docs/ai/ai-help-assistant/03-answer-with-docs.png)

Related guides:

- [Choose Your Home Page](/docs/getting-started/profile-preferences/home-page)
- [AI Help Assistant](/docs/ai-assistance/help-assistant)

## Vision workflows are more end to end

The Vision experience is easier to follow from setup through validation instead of feeling like separate disconnected steps.

You can now move more clearly through this flow:

1. Create and pair an **AI-Vision** data source.
2. Use **Remote Control** to collect sample images or validation videos.
3. Organize images into **Datasets** and manage dataset versions.
4. Start training from a dataset version and review results through **Models**.
5. Use **Videos** and **Tasks** to work through validation and follow-up activity.

![AI-Vision data source editor with built-in fields and preview.](/docs/data-sources/ai-vision-data-source/03-fields-and-preview.png)

Related guides:

- [AI-Vision Data Source](/docs/data-sources/ai-vision-data-source)
- [AI Vision Overview](/docs/ai-vision/overview)
- [Remote Control](/docs/ai-vision/remote-control)
- [Datasets](/docs/ai-vision/datasets)
- [Models](/docs/ai-vision/models)
- [Videos](/docs/ai-vision/videos)
- [Tasks](/docs/ai-vision/tasks)

## Everyday work should feel smoother

This release also improves the smaller moments that shape daily work:

- chat creation and messaging feel lighter and more direct
- dashboard and data-source setup flows are more guided
- form filling from image or text helps you start from a draft instead of entering everything manually

If you are returning to Zira after a short break, the fastest path is to open **Help**, ask the **AI Help Assistant** what changed, and follow the linked guides from there.


Use AI Chat to create forms, data sources, and dashboards faster from a plain-language request.

Build with AI Assistance

Upload an image or paste notes, then review the AI draft before applying it.

Fill forms from image or text

Move from image sampling to annotation, model training, and validation in one clearer workflow.

End-to-end Vision workflow

The sidebar is cleaner, search is inline, and key actions are easier to reach.

Refreshed navigation

Set a dashboard as your Zira home page so the first screen matches your daily work.

Custom home page dashboards

Find docs, updates, and guided answers in one place when something feels unfamiliar.

Help & Support with AI Assistant

Conversation creation and messaging now feel lighter, faster, and easier to manage.

Updated Chat experience

Zira Web now feels easier to learn and faster to use, with stronger AI creation flows, cleaner navigation, a clearer Vision workflow, a better Help experience, and smoother day-to-day setup across dashboards, forms, and data sources.

Zira AI Assistant and UI Refresh


This update simplifies everyday navigation and moves dashboard actions into a calmer, more predictable toolbar.

## What changed

- The left drawer now starts collapsed by default.
- If you open the drawer, it stays open as you move between routes.
- The open drawer is wider, giving sidebar lists more room.
- Notifications moved from the top app bar into the left sidebar rail.
- Notifications now open in a large dialog with a dark backdrop.
- The top app bar is cleaner, with search and the profile avatar as the primary actions.
- Dashboard toolbar actions now live in **Dashboard options** next to the time range.
- Full screen moved into **Dashboard options** under **Actions**, below **Share**.
- Dashboard toolbar styling now matches the page background.

```text
Screenshot placeholder
Collapsed left rail with Notifications and Help near the bottom.
Expected file: /public/changelogs/navigation-dashboard-toolbar-refresh/01-left-rail.png
```

```text
Screenshot placeholder
Notifications dialog opened from the left rail.
Expected file: /public/changelogs/navigation-dashboard-toolbar-refresh/02-notifications-dialog.png
```

```text
Screenshot placeholder
Dashboard toolbar with Dashboard options next to the time range.
Expected file: /public/changelogs/navigation-dashboard-toolbar-refresh/03-dashboard-options.png
```


A simpler left sidebar, a cleaner dashboard options menu, and a larger notifications dialog make common actions easier to reach.

Navigation and Dashboard Toolbar Refresh


This update replaces the small profile menu with a larger account settings dialog, making everyday account tasks easier to find and manage.

## What changed

- Clicking your avatar now opens a full account settings dialog with vertical tabs.
- Profile, Preferences, Companies, and Change password now live in one place.
- The Profile tab has a clearer layout for profile image, personal details, and signature updates.
- Preferences now uses dropdown controls for notification tone and date format, with a simpler home page selector.
- Companies now opens directly inside account settings, with search, company images, and a simplified company-name list.
- Sign out remains available from the account settings sidebar.

```text
Screenshot placeholder
Account settings dialog with Profile, Preferences, Companies, and Change password tabs.
Expected file: /public/changelogs/account-settings-refresh/01-account-settings.png
```


A larger account settings dialog now brings profile, preferences, companies, and password updates into one cleaner place.

Account Settings Refresh


Working with forms now feels much more direct, whether you are building a new schema or trying to complete one quickly in the field.

## Create forms and task forms with AI

- The **Add Form** flow now gives you clearer starting points, including **Form With AI** and **Task With AI**.
- The same builder-style AI experience is also available when you edit an existing form or task form.
- AI can draft the title, description, fields, options, and validation so you can start from a strong first version instead of a blank editor.

## Open forms in a cleaner full-page flow

- Opening a form from the **Forms** tab now takes you into a full-page form experience instead of a modal.
- The page is easier to read, easier to complete, and easier to return from when you are done.
- After submit, you return to the Forms tab more smoothly.

## Fill forms with AI from text or images

- Full-page forms now include **Use AI**.
- You can paste free text or upload a photo of a paper form or work order.
- Zira first shows a draft so you can review what AI understood before anything is applied to the form.
- The same flow is useful for follow-up edits, so you can revise the draft before applying it.
- Full-page forms also include a faster **Scan Image** path for the most common tablet workflow: take a photo and let Zira fill the form behind the scenes.

## Why this matters

- Form creation is faster because AI can draft the first structure for you.
- Field teams can move faster from paper or notes into a digital form.
- Teams can choose between a fast scan path and a review-first draft path depending on the job.
- The review step keeps the AI flow useful without feeling risky.
- The full-page form layout feels more like a focused workspace and less like a temporary pop-up.


You can now create forms and task forms with AI, open forms in a cleaner full-page experience, and use AI Fill in both a fast Scan Image flow and a reviewable draft flow.

Forms Are Faster to Build and Fill


Building a dashboard no longer has to start from a blank canvas.

## Describe the dashboard you want

The new **AI build dashboard…** action helps you start from intent instead of manual setup.

You can ask for:

- a specific dashboard structure
- a focused KPI and chart mix
- drilldown when it makes sense
- a more open request like “make me a dashboard based on the data”

## Get a real draft before you apply it

The assistant now returns a reviewable dashboard plan with:

- a suggested dashboard title and description
- a list of planned widgets
- responsive layout previews for desktop, tablet, and mobile

That makes it easier to review the dashboard story before anything is added to the page.

## Start with a stronger first version

The first version of AI Dashboard Builder focuses on:

- **Header**
- **Data Research** widgets

That means the generated dashboard stays focused on the most useful summary and comparison views while giving you a much faster starting point.

## Why this matters

Dashboard creation is now faster, more guided, and easier to shape around a real question.

Instead of deciding every widget one at a time, you can start with a draft that already has:

- a clear structure
- a better mobile-to-desktop layout
- a stronger dashboard story from the start


You can now describe the dashboard you want, let AI choose a strong starting set of widgets, and apply responsive layouts in one flow.

Meet the New AI Dashboard Builder


Creating a new data source now feels much more guided from the very first click. This update brings a cleaner create flow, better AI template browsing, and a brand-new AI Builder assistant for drafting custom schemas inside the editor.

## A better first step

- **Add Data Source** now opens a more guided create dialog directly.
- AI-focused options are highlighted first, while collector-based site options remain available below.
- **Add Existing Data Source** is now part of the same create surface instead of a separate first step.

## AI templates are easier to choose

- **AI-Vision** and **AI Pallet QA and Routing** now stay inside the same dialog when you move into template selection.
- Templates are shown as cards with descriptions, so it is easier to understand what each setup is for before creating the data source.
- The template flow still finishes by creating and attaching the data source to the current channel.

## Custom schemas now have AI Assist

- The custom data-source editor now includes **AI Assist**.
- You can describe the structure you want, and the assistant can draft a title, description, and first-pass field list directly into the editor.
- Suggested fields are shown with more detail before you apply them, including field type, metric matching, validation, options, and expressions when relevant.

## Why this feels different

- You can get from idea to first draft much faster.
- AI-based data-source setups are easier to understand before you commit.
- Custom schemas take less manual setup work.
- You get a clearer preview of what AI is about to apply before it changes the editor.


Data-source setup is now faster and more intuitive, with a redesigned create flow, richer AI template selection, and a new AI Builder assistant inside the custom editor.

Smarter Data Source Creation Has Arrived


We refreshed the AI experience across dashboard insights and AI Agents to make it easier to ask better questions, review the right widget context, and shape answers before saving.

## A clearer AI setup flow

- The shared AI setup surface now has a cleaner layout with clearer sections for prompts, answer style, time range, and widget context.
- The prompt area is easier to scan and the main actions are more obvious.
- Widget context is presented more clearly so it is easier to confirm what the AI is using before you generate an answer.

## Better answer previews

- AI answer previews now feel more polished and easier to read.
- Generated insights use richer markdown styling, so headings, lists, and structured responses are easier to scan.
- The saved AI insight widget on dashboards now matches the newer AI presentation style more closely.

## A smoother path from exploration to reusable agents

- It is now easier to move from a one-time AI insight into a reusable agent workflow.
- The AI Assistance and AI Agents experiences now feel more connected, so testing a prompt and refining a reusable setup is more natural.
- Embedded AI insight flows now present the same style more consistently.

## Why this matters

- Faster prompt testing
- Cleaner answer review
- Better confidence in widget context
- A more guided AI workflow for reusable reporting



A cleaner AI experience across dashboard insights and AI Agent setup, with better answer previews, clearer structure, and a more guided workflow.

AI Assistance and AI Agents Refresh


This update focused on the chat drawer, the create-chat flow, the conversation experience, and the loading behavior around messages.

## Chats and new chat flow

- The Chats drawer now matches the same search and navigation language used in Vision.
- Opening a full conversation is more direct, while the quick preview flow remains available when needed.
- Creating a new private or group conversation now feels lighter and more step-by-step.
- Selected people in the create-chat flow are shown in a more compact and cleaner way.

## Conversation experience

- The mini chat preview was redesigned as a centered modal, giving it a clearer role as a quick chat surface.
- Full chat and mini chat now share a more consistent message composer style.
- Conversation spacing, date separators, reply previews, and footer layout were refined for a cleaner reading experience.
- Dark mode contrast was improved in a few key places, especially around reply previews and message surfaces.

## Performance and responsiveness

- Common chat-list updates now happen more quickly, especially when new messages arrive.
- Latest-message edits update the chat preview faster.
- Message sending and message editing feel more immediate through lighter update handling and optimistic UI in the main conversation view.
- Several loading states were softened by replacing attention-heavy linear progress bars with gentler inline loaders.
- Message and socket handling were tightened to reduce stale UI states and unnecessary rerendering.

## Follow-up

We are continuing to refine group conversation previews in future updates.


Chats now feel cleaner, more natural to use, and more responsive across creation, preview, and conversation flows.

Chat Experience and Performance Improvements


This update focused on the left drawer and overall navigation flow, with an emphasis on consistency and quicker movement between sections.

## What changed

- The left drawer now feels more consistent across Chats, Channels, Sites, and other navigation sections.
- Search now lives inline inside the drawer, making it faster to browse and switch without opening extra dialogs.
- Drawer layout and section ordering were tightened so moving between sections feels more natural.
- On conversation pages, the Chats drawer now opens automatically to preserve navigation context.
- On mobile, drawer behavior was cleaned up so the screen remains fully interactive after the drawer closes.



A cleaner left drawer with inline search, more consistent section behavior, and smoother navigation across desktop and mobile.

Sidebar and Navigation Refresh


This update focused on making the Vision devices flow easier to scan and easier to stay inside while navigating between devices.

## What changed

- Vision devices now use an inline drawer experience that is faster to browse and easier to scan.
- Users can stay in the drawer, search devices, review statuses, and open a selected device while keeping the Vision context visible.
- Visual hierarchy inside the Vision drawer was refined so the list is easier to read.



The Vision devices drawer is now easier to browse, faster to search, and better aligned with the rest of the app.

See what is new in Zira.

Zira Plugin: Device Status, Scheduled Reports and Rare Defects (Beta)

The Zira plugin for Claude and Codex can now check the battery, temperature, camera and connection of every camera, set up weekly reports for your saws and Pallet QA cameras in a few questions, find more rare defects in your production images, and show an AI report before you schedule it.

Browse all updates