Every release, newest first. Each opens with what changed and what may need you to act; the detail is one click further. Security fixes follow our disclosure policy.
Nothing is charged during the beta, and business organizations get Starter's limits.
Saves change only what you changed, pages show only what your role can do, and unused settings are gone.
Web app & notifications
Folders show their nightly copy where an unused backup choice sat. New databases pick Single, Standby, Cluster or Fleet; a typed count above five used to create one instance. Also gone: storage types, app labels and a job's registry pick.
Apps & projects
A save used to rewrite the command, its arguments and exec health checks, which could stop an app from starting. The restart notice shows only when one follows.
A rollout keeps the old version serving until the new one accepts connections. With klops, a worker that listens on nothing needs --port 0.
Values set with klops or an AI tool were stored and never read. The next run reads them.
A job that calls a URL can be saved now and shows its address. An automatically sized job stays automatic, and Resources has a way back to Automatic.
The domain editor says the app is private and offers Make public, in place of a switch that snapped back.
Used by now names the file, so a key a file reads no longer shows as unused before you delete it.
Builds & deploys
The app, its health check and its PORT setting all move to the port found in the code.
Databases & backups
Deleting such a disk on its own is refused now, so it can't vanish beyond Recently deleted.
The buttons were there and the server refused. Viewers no longer see them in Studio.
Account & organization
Members now offers Owner, so the only owner of a team can hand over and leave.
Web app & notifications
Deleting a project mentions its 7 days in Recently deleted, firewall pages stop claiming denies they don't add, and your account shows your real role.
Viewers read settings without buttons that fail on click. Members no longer see token, repository and pull credential controls. Admins no longer see Delete organization or controls on an owner's row.
One health picture everywhere, warnings that open the broken thing, and channels you can tune, mute and quiet.
Web app & notifications
Under Notifications, nothing about it reaches channels or incident mail for up to 7 days. The top bar still shows what is wrong, and an incident that outlasts the mute is mailed afterwards.
Inside them only "something is down" comes through. Everything else arrives when they end, in Swiss time.
Apps & projects
It names each rule in plain words and saves at once. The CPU, memory and volume thresholds keep their defaults; klops still changes them.
A crash opens the app's logs, a failed database backup its backups, a missing certificate its domain. Alerts name the app or database in plain words, and a certificate that did not renew warns two weeks before it expires.
Web app & notifications
Something is down, needs you soon, activity. A new channel starts with the first two on. Existing channels hear what they heard before.
klickops now rejects certificates from authorities that are no longer trusted.
Apps & projects
klickops now rejects certificates issued through intermediate authorities that are no longer trusted.
Projects with the same name stay apart between organizations, share links stay safe, and pages say when loading failed.
Apps & projects
Deploying from your own code into an existing project needs the admin role when it sets the project's shared secrets or settings, as changing them directly does.
Every access check now looks at the same project the action reaches, also when another organization uses the same project name.
Opening a share link takes the visitor to the app it was made for, or to your own domains. Any other address is refused.
Making a domain public and private again no longer brings back the share links that went off with it. Links also work for projects whose name another organization uses.
Apps & projects
An alert for an app could stay silent when another organization had a project and an app with the same names and the same alert firing.
An app deleted while asleep comes back awake and sleeps again when idle. An app that shared a domain with another app gets its paths on that domain back.
Bringing back a withdrawn invite or going back after "Make it live" used to fail without a word. Both now show what went wrong.
After "Show more", the lists of sources, pages and countries no longer drop back to five a minute later. An app's page also links back to Your apps whenever you have more than one app.
An app whose image names a tag and a digest reported the registry as unreachable on every check. One pinned by digest alone could be moved to the newest tag by automatic updates. It now stays as you set it.
A project moved into an organization that already had one of that name left two projects with one name there, and one of them could not be opened. The move now says the name is taken.
The policy list emptied whenever one refresh failed.
The dialog offered to connect a domain when reading the app's own domain failed, even for an app that has one. It now says the read failed, with Try again.
A project list could show zero apps and nothing healthy for a few seconds when another organization had just loaded its own.
A restart or a step back to an earlier version that met another change failed with an unexpected error. It now tries again, and a database change that meets one says so instead of blaming your input.
Making a link turns a public app private first. When the link then failed, the app page still said anyone could open it. Going back to an older version also says so again when the version list is slow.
A failed request showed no versions, no rules, or that no app used a disk.
Builds & deploys
Registries showed none configured after a failed request. It now says so, with Retry.
Databases & backups
It used to open the database's overview, one more click away from the copy you wanted.
A cache database named like an app took over the app's address, and deleting it took the app offline. Creating either one now says the name is taken.
A database created on a major version, like 16, could be moved to an older 16.0 release by automatic updates. It now updates to the newest release of its version only.
With 16.9, 16.10 and 16.11 available, a database on 16.8 now updates to 16.11 in one restart instead of stopping at 16.9 first.
A copy of such an app's volume never started and raised a "not ready" alert.
After a larger disk, every later change to the store, version updates included, failed with a message about shrinking it. Changes work again.
Command line & AI tools
Checking an app, reading its picture again and getting a download address for its code are no longer marked read-only. Pages and sources your visitors send now reach the model marked as data.
Adding a verified domain now points at klops orgs verified-domains check, and the database import and export examples use klops databases, not the old names that are going away.
klops databases resources says it is going away. It has answered "not found" for every database since July. The command and its AI tool now say so and point at klops databases get; both go after 13 November.
Restoring a night of a saved folder is now marked as destructive, so your AI client asks you before it replaces the folder.
Account & organization
Invitations, Help, API tokens, verified domains and connected apps showed "none yet" after a failed request. Who may join could be saved over without loading first.
It no longer promises how people join, which lives under Members.
One failed refresh could show "Organization not found" or say nobody had access.
The page checked your role in the last project you opened instead of in the organization, so it could show the build registry read-only to an admin.
It used to say nothing was planned yet. Try again loads it once more.
Before, signing in from a link sent you to your home page instead of the page the link named.
Signing the CLI out and removing someone's project role could report success while nothing changed. Both now say to try again.
Web app & notifications
Filters, searches, ports, scaling numbers and the reply box used to be read out as "edit text".
They used to read "Tokens" and "Repos".
When we take a decision on your request back, the Roadmap page no longer shows the reason we gave before.
Monthly bills settle exactly, corrections go back to your wallet, and what's new arrives twice a month.
Web app & notifications
Each half month opens with what changed and what may need you to act. Every release it holds is one click further.
klickops now gathers its releases twice a month under one title. The monthly summary lists those titles and still says what may need you to act.
Releases still ship every day. What's new gathers them by half month and only lights up then, or at once when a change needs you. The Update ready notice is gone: the next click loads the new version.
Billing & plans
When usage was taken off a closed month, the statement changed but the charge stayed. Now the amount goes back to your prepaid balance, and your billing timeline says why.
The usage and plan price up to the delete were left off every statement before. Now that month closes like any other.
Their short runs were billed as your app's compute and disk, while a backup only bills the space it takes in the backup bucket. Network file storage on every storage class now bills at the network file price.
It showed usage beyond the credit that Free writes off at the month's end. The credit mails also name your plan's included credit again.
Their size was counted from a listing that stopped there, so the rest cost nothing. The size now comes from the storage itself, for every bucket.
When the credit ran out mid-month, Free was charged the rest of the month at once and could stay paused into the next. Now the month's end settles it like any other Free month.
An app and its release command share a name, and their usage used to appear as one line. The total stays the same.
Account & organization
It could only fail: each account starts one organization, and you join others when someone invites you.
Deletes wait 7 days, sleeping apps read asleep, new databases start reliably, and statements come in whole Rappen.
Web app & notifications
The separate Organizations page is gone: the menu at the top of the sidebar lists your organizations, switches between them and starts a new one.
Apps & projects
One crash or out-of-memory stop used to mark an app as needing attention until it was redeployed. Now it counts for an hour after the app runs again.
Your projects, app lists and sidebar say Asleep, previews that sleep keep their Open button, and a check of a sleeping app says to wake it first.
A new image for an app or a scheduled job, and a database update, now reach your organization's channels, once per new version instead of every few minutes.
Databases & backups
One whose backups were ready at once could wait forever to start. Backups now begin a minute after it is up, and a database already stuck this way starts again on its own.
Billing & plans
A month closes once all its usage is recorded, and a paid statement no longer leaves a fraction of a Rappen that kept apps paused.
Backups show seven nights and alert when one fails, light and dark mode read better, and updates finish running requests.
Web app & notifications
klickops runs a few database imports and file uploads at a time and reads only the end of large files for your AI tool.
Databases & backups
Owners and admins get one email a day, the bell lists it until the next copy works, and notification channels get backup.failed. A nearly full backup bucket is reported before it fills.
Web app & notifications
Cards now stand off the page in both themes, and dark mode turns from black to a deep blue. Links, form fields and status words are easier to read.
Apps & projects
A burst of creates, from a script or Terraform, waits its turn instead of holding up every other request.
Web app & notifications
A request sent just before an update finishes first, and live views reconnect on their own.
Private apps let in only the people you allowed, and an org without a plan pays its overage.
Apps & projects
klickops keeps an app's picture only when its image is a sensible size, and reads one at a time.
klickops checks every visit against the app it actually reaches.
Web app & notifications
Every release builds the dashboard with the same pinned tools as our checks.
Databases & backups
Back up now shows waiting to start, then copying now, then the result, without a reload. A database or folder without a good copy for a day reads behind.
Old pieces of a saved folder's copies are removed only after a safe wait, so a copy running at the same time keeps everything it uses.
Billing & plans
When the usage source did not answer, that hour used to be recorded as zero for good. klickops now tries it again for two days and bills what really ran.
An organization without a plan now pays its usage above the included credit at month end, like every paid plan.
App pictures stay on screen, checking an app works for more apps, and a restart pulls a new image.
Apps & projects
After you type a new value for a secret in an app's configuration, Show lets you read what you typed; saved values keep their own Reveal.
Apps & projects
It no longer asks for an own domain's status when the app has none.
A picture shown once stays on screen while you move between your apps, instead of flashing back to the name.
klickops asks your app the way a visitor's browser does, so an app that reads the visitor's address from a header answers, and a preview image on your own domain shows.
Push a new image under the same tag, restart, and the app runs it. A scheduled job does the same on every run once it is saved again.
Each card's icon now comes from the picture klickops keeps, not from a request to your app.
App cards show each app's own preview image or icon, read straight from its page.
Apps & projects
klickops reads the preview image your front page shares, or its icon, straight from your app, so private apps get one too and it appears right away.
Command line & AI tools
A command that sends many requests at once waits the time the platform names and carries on, instead of stopping with "too many requests".
Cockpit's Your apps cards show each app as it looks, with Visit and Copy link on the card.
Apps & projects
A card holds the live page in a small browser frame, with Visit and Copy link right on it. Several projects get a section each.
Memory is billed for the minutes a workload ran, Free write-offs stay, and Cockpit says what to do next.
Apps & projects
The window says what is true right now, then shows both choices with the page a visitor gets: your app, or the sign-in page.
Beside the picture, every fact has its button: make it private, fill the empty keys, go back a version. Visitors and cost show as charts, and your AI tool's questions sit one click away.
Account & organization
Pick your tool, do its one step, and the window shows Connected the moment you allow it. The top bar names the tool you connected.
Apps & projects
We now read the keys your SvelteKit code imports from its env modules, so the keys list names what the app uses, marks a key with a fallback as optional and a public one as public.
Billing & plans
It no longer says backups bill your volumes' used space, or that you pay the greater of credit or usage: you pay the plan's price plus usage above the credit.
Usage beyond Free's credit that we wrote off at month end is no longer taken from a top-up or promo credit that arrives later in the month.
A scheduled job, a build or an app that sleeps and wakes used to bill a full hour of memory for every hour it touched; now each minute counts once.
An app and its release step, or an app and a database with the same name, each used to replace the other's usage for the hour; both now count.
When your code needs a database, klickops offers PostgreSQL; Cockpit shows keys in one place and the month's usage.
Apps & projects
The Keys window lists the empty keys first, each with the button that fills it, then the set ones with a line on what each is for.
Billing & plans
One line shows what your apps used so far against what your plan includes. Below, each app's share, its database included, and a hint when a smaller plan would do.
Apps & projects
When your code needs a database address to start and names no other database, we offer PostgreSQL and fill DATABASE_URL, instead of saying no database is needed.
Free apps sleep after half an hour without visits and wake on the next one; each version says what changed.
Apps & projects
Put one to sleep from the CLI or your AI tool: it uses no compute while it sleeps, and the next visit wakes it within a few seconds.
Apps & projects
A switch in the app's scaling settings turns it on or off, and an app that sleeps says so on its page instead of "Stopped".
They use no compute while asleep, so the credit lasts longer, and the next visit wakes them within a few seconds. Paid plans stay always on.
When a chat changes your app's code, it writes one line about the change. The app's page and its list of versions show that line.
Account & organization
It says what klickops does, points newcomers to the waitlist and follows your system's light or dark.
A sleeping app wakes on its next visit again, and other apps' domain changes no longer wait for it.
Apps & projects
An app you put to sleep answered with an error instead of waking, and while it slept, changes to other apps' domains could be held back. Now a visit shows "Waking up" and the app is back within seconds.
Apps say when a rebuild installs a Node security fix, and going back after a rebuild runs the old version.
1 change may need you to act
Builds & deploys
klickops records the Node version each build installs. An app that misses a fix shows "Rebuild recommended" with a Rebuild button, which keeps your code and installs the newest patch of the same version.
Builds & deploys
Each start printed two warnings that looked like errors although the app ran fine. New apps start quietly now, and existing ones do after their next build.
Rebuilding the code that already ran replaced the version you would go back to, so going back started the rebuild again. Each rebuild now keeps its own image.
Saved folder nights show their real size, the backup bucket can't be deleted by mistake, and false backup alerts stop.
Apps & projects
An alert that was still firing arrived again each time klickops was updated. Now each alert is sent once, and again only when it fires anew.
Databases & backups
A database whose backup failed once stayed listed for a day, even after later backups succeeded. It now drops off with the next good backup.
A project whose only backups were saved folders could delete its backup bucket, and every night in it. The bucket stays while a folder or database backs up into it, and its page offers no upload or delete.
Each night of a saved folder showed 0 files and 0 bytes, although the copy held the folder. It shows how many files and how much it holds now.
Nightly copies of saved folders are written and kept again, and backup alerts only fire for real failures.
Databases & backups
A database whose last backup failed stayed in alert for a day, even when every backup after it succeeded. The alert now ends with the next good backup.
The last update still could not write the nightly copies of saved folders. They are written and kept now, and you can put a night back or download it.
Nightly copies of saved folders store through a client the storage accepts, and the Cockpit redesign continues.
Builds & deploys
Checking a repository reads the line from the app's own words: its env settings in the code, else the comment above the key in its example file.
Apps & projects
The version is stored at once and builds in the background, with Undo once it is live. In reviews long lines wrap, the page scrolls instead of each file, and a first version shows its files plain.
Databases & backups
The nightly copies of saved folders failed when they wrote to the backup bucket. They write to it now, and the next night copies as planned.
Nightly copies of saved folders reach their backup bucket again, and the Backups page counts only copies that worked.
Databases & backups
The Backups page counted a folder or database as backed up while its last copy had failed. It now counts only the ones whose last copy worked and says how many need you.
Since the last update, copying a saved folder each night failed because the copy could not reach your backup bucket. It reaches it now, and the next night copies as planned.
Every project now backs up its data, Hobby starts at once, and Free includes more room.
1 change may need you to act
Apps & projects
Previewing or saving an app file or a connection setting now ends a template that loops for too long or writes more than 256 KiB, and shows the reason. Normal templates render as before.
Linking a certificate to an app and deleting a certificate now act on the project's own certificates. Credentials the platform keeps in a project stay out of reach of both.
An extra address that forwards to a domain answers with the forward only. Private apps stay behind their sign-in on every address, and forwards you already have are updated when the platform starts.
An upload may hold 10,000 files and folders together. The platform handles four uploads at once, and a fifth is asked to try again in a minute.
Builds & deploys
An organization's registry address must be a public name, and uploaded code is stored there over a connection that never reaches inside the platform.
Databases & backups
An app can no longer take your backup bucket's keys as a variable or a mounted file. A project copy made with its secrets leaves them behind. Only klickops writes to the backup bucket.
A connection's preview never shows a stored password or key, and a value that carries one is always kept as a secret on the app.
Copying, putting back and downloading a saved folder now run in the same isolated environment as the app, and keep no extra privileges.
Web app & notifications
The activity feed is matched to the organization as well as the project name. Actions the platform records for no organization appear in no project's feed.
Apps & projects
On an app built from uploaded or AI-written code, Edit opens its files. Review the changes, then save them as a new version that builds and goes online. A version someone saved in between is shown before you save.
On an app built from uploaded or AI-written code, the Source tab lists every version and shows its changed files as a review, with colours. Compare any two, and go back from there.
Databases & backups
Databases and saved folders go to the project's backup bucket, including those that had no backups yet. Earlier recovery points stay until they expire; ask support to restore one.
Billing & plans
Hobby applies the moment you choose it and comes on your next statement. A paused account can ask for a top-up, and Free starts again on the 1st.
Billing & plans
They bill as bucket storage, by the bytes kept, instead of counting your volumes a second time. Earlier backups no longer bill.
Enough for a small app and its database. Your Free projects get the extra disk on their own.
Apps without a Dockerfile build again, the Vault generates secrets, and backups live on one page.
Apps & projects
Alert notifications in your channels now come from the platform's monitoring alone.
Builds & deploys
The address for sending your code reads as random characters. Addresses handed out before this release stop working; ask for a new one.
The platform's registry credential stays inside the platform, and images from the platform's registry must come from your own organization's builds.
Command line & AI tools
The first terminal to collect an approved sign-in gets the session. Any later attempt reads the sign-in as expired.
Apps & projects
Write a file once with placeholders like {{ .vault.DB_PASSWORD }}; klickops fills it in and keeps it current when a value changes. The preview hides secrets.
A new app can begin from a small server or a static site, the AI reads the rules an app follows here, and a bucket connects under the names every S3 library knows.
Deploy now takes an uploaded folder with no repository, and shows how to connect your AI tool so it can write the app and put it online.
When a push fails to build or its database update fails, the owners and admins get one email saying why, and that the previous version is still online.
Deploying from a repository from the terminal or your AI tool can keep one folder, such as uploads or a SQLite file. Each update then restarts the app. The folder is not backed up yet.
A secret every visitor could read, a Stripe test key, a publishable key in the secret slot or a Supabase key of another project gets a warning when you save it. The key is stored anyway.
klickops lists the records to set, checks them for you and adds the domain with HTTPS once they are right, so the certificate works the first time.
An app built from uploaded code can run a version as a preview with its own address and an empty database, then make it live without building again.
Give an address and a schedule; an error answer fails the run, and a header can read its value from your secrets without it ever being shown.
The app receives the person's email, account id and role, or the share link a guest used, and a visitor cannot fake them. Apps uploaded from a chat start private.
An app you uploaded or your AI tool wrote now lists every version with who made it, its files and what changed, its preview with Promote and Discard, and a source download.
Self-signed, signed by your project's own authority, or publicly trusted. They renew on their own, and Regenerate issues one again with a new key.
Put a secret, a setting or a certificate at a path in an app, read-only, or give it a different variable name. AI tools can do the same.
Pick a length and format, or a key pair type, and klickops makes the value without showing it. Regenerate makes a new one the same way; AI tools can too.
Databases & backups
Every change is saved into the project's backup bucket, with a full copy each night. You can go back to any moment in the last 7 days, on every plan.
Restore now offers "Replace this database": it keeps its name and its apps restart on the data you picked. For 7 days, Return brings back what it had before.
Each app's folders go to the project's backup bucket, 7 nights kept, from a snapshot so open files stay consistent. Back up now copies one at once.
A database created together with an app takes PostgreSQL's security releases on its own, daily between 02:00 and 04:00 UTC. It may not answer for up to a minute while it restarts.
Its app stops for the moment it takes and starts again. Download saves a night as a .tar.gz without changing anything.
Each row says how far back it goes and has its own Restore. A saved folder's nights can be put back or downloaded, in Studio and on an app's Saved files row.
Command line & AI tools
klops recommendations app lists the protections that hold for it now, such as no admin rights, its own network and https, each read from the running app.
klops apps repo-deploy --generate AUTH_SECRET stores a strong random value once and keeps it on every re-run; --address-key NEXTAUTH_URL keeps a key at the app's address.
Apps & projects
A change made from Claude or another AI client, or with klops, shows its name beside yours.
Each version that goes live is checked again. When it starts saving files in a folder that updates throw away, or moves to PostgreSQL without updating its tables first, your app says so.
After a rollback the app named the newer commit; it now names the one you went back to, and klops apps revisions list says which versions carried a problem to fix before sharing.
A domain whose certificate fails names the cause in one sentence, such as a CAA record or a proxy in the way, and Try again starts it over at once.
It says the app is private and offers sign-in, and an expired or turned-off share link says so instead of showing a sign-in form.
A file opens with line numbers and its type. Edits and uploads show what changes before saving, and you can download a file; a secret one counts as a reveal.
Builds & deploys
Keys in the code, fallback secrets and debug mode are found on each build and each config save, shown on the app, and never stop a deploy.
A cancelled sign-in, an install waiting for your GitHub organization's owner and a repository with no code yet each get their own message, and Check again reads the code you just pushed.
The apps that build from a repository show their folder, project, status and when they last built. klops repos apps and your AI tool list them too.
Databases & backups
At 70% of its size limit the database is listed there, so you see it before backups stop, without opening the database.
Command line & AI tools
klops config list, set and delete are klops configmaps, the team names are set-org, use-org and apply-org-defaults, and adding a member is klops orgs invites create.
Billing & plans
The usage report splits every project into its apps, databases, buckets, scheduled jobs and builds, so you see which one uses your plan's included amount.
Web app & notifications
Each job is one line with its schedule, next run and state, like apps, databases and buckets.
Apps & projects
A failed build carries its last lines, a check asks for the app's own address as a visitor does, and a picture starts at once instead of timing out.
Your apps, the app's page, klops open and the end of a deploy show the app's domain, and each database and bucket says which apps use it.
When another change, or a secret save's restart, lands while you save, your save is refused with "changed by another update" instead of undoing it.
A project that already has an app, database, bucket or scheduled job of that name now refuses the deploy before anything is created. The Deploy page says so on the app and lets you rename it.
The second create is refused with "already exists", and the project the first one made stays.
Builds & deploys
An app in a pnpm, yarn or npm workspace builds from the repository root with its own scripts. Submodules and Git LFS files are fetched instead of left empty.
A branch or repository that isn't there now answers "not found", credentials the Git server refuses answer as rejected, and an unreachable server answers as unreachable.
The archive a Mac makes was refused, and its hidden metadata files ended up in the version. Now it is taken as is, the metadata files are left out, and a refused upload can be sent again.
The steps list the fetch of your code instead of a clone that never ran, and a failure the platform cannot classify shows the build's own error line.
A new app without a Dockerfile, from a repository or from uploaded code, failed at its first build. It builds again, and a build that fails this way now says so.
Databases & backups
A project with a volume on shared file storage no longer shows its nightly backup as failed or mails a backup incident. Volumes you back up as files are still included.
An update now applies only inside the window you set for the database, such as daily 02:00 to 04:00 UTC. A database without a window updates whenever one is ready, as before.
A database made in the wizard shares its memory with its backups instead of taking more.
Command line & AI tools
Buckets share the storage gateway and have no pods of their own. klops buckets restart reported a restart that never happened.
Billing & plans
Projects, apps, scheduled jobs and databases created at the same moment are counted one after another, so the plan's limits hold.
Account & organization
A double submit of the first-run form creates one organization; the second answers that you already belong to one.
When two owners step down or leave at the same moment, one of them is refused, so the organization never ends up without anyone who can manage it.
Web app & notifications
A link between two handbook pages used to open the German page.
Put an app online without a repository, make apps private with their own share links, and seven security fixes.
1 change may need you to act
klops up in a folder, or let your AI assistant send its code: klickops checks it, keeps each version, builds it and deploys it. New apps start behind sign-in. Apps & projects
The checks behind apps that require klickops sign-in now answer only the sign-in gate itself, so a sign-in or guest link works only through it.
Making an app public turns its share links off for good, and the service that checks sign-in for your apps holds no key that could sign anyone in.
You: Share links made before this release stop working. Make new ones where you still need them.
Billing & plans
When klickops can't check your organization's standing for a moment, it asks you to try again instead of letting the change through.
An organization deleted while a project is created in it keeps the project and says so, or the project is not created.
When klickops can't read your plan for a moment, it asks you to try again instead of creating the project. Projects that started without their limits get them back.
Account & organization
Webhook addresses and signing keys of your channels are stored apart from the rest of the channel. Nothing changes for you.
The web app tries again on its own, and klops auth logout warns you when the session is still valid.
Apps & projects
After an upload klickops checks the app's home page, and on a failure it returns the reason and the app's last lines of output. Private apps included.
It gets a small picture of the live page, private apps included. klops apps screenshot saves one too.
Private asks people in your workspace to sign in on every address of the app, including ones you add later; everyone else needs a share link. Making it public turns its share links off.
Builds & deploys
Your AI assistant or klops apps source put sends only the changed files. Each upload is kept as a version: list them, download one, or go back to it.
Apps & projects
A domain you brought that asks for team sign-in now signs your team in and brings them back. Signing in lasts up to 12 hours instead of 15 minutes, and signing out ends it.
While klickops or klickops.io updates, a page lists what is paused and what keeps running.
Web app & notifications
While the dashboard or the website updates, a board lists what is paused and what keeps running, and the page returns on its own when the update is done.
Private persons can now use the Free plan, and help is one click away when you reach a limit.
Account & organization
If you live in Switzerland, you can create an organization for your own projects. It stays on Free; paid plans remain for businesses.
Account & organization
It stays visible on every page, and tapping it says what beta means.
When you need a service klickops does not offer, an app that starts containers or your own cluster, the assistant says so and who can help.
It asks you to accept the terms and links to them, instead of summing them up.
Account & organization
The record of failed and refused requests keeps the counts and drops who made them.
At a limit, klickops says who can host more for you, and stuck scheduled jobs say why.
Account & organization
A third box says things can still change, and that you keep your own backups of anything important.
When you reach a plan cap, a project's capacity or a size limit, the web app, klops and AI assistants say who can host more for you.
Account & organization
The API, klops and AI assistants get the limit's name with the refusal, so a script or an assistant can tell a plan cap from a typo.
It lists what we keep about how you use the web app, the command line and AI assistants, for how long, and how to object.
Apps & projects
A scheduled job using an image klickops built for you now pulls it with your project's credentials, like the app it came from.
A job whose run cannot start shows Stuck with the reason on its card and on the project overview, instead of Running.
A run stuck on its image or crashing on start now shows as Stuck with the reason, instead of Running for days. "Every minute" schedules read correctly too.
Assistants can write to a bucket, and changing a connection's variables no longer leaves old ones behind.
Databases & backups
They could only list a bucket before. A file of up to 3 MB goes in one call.
Builds & deploys
The project's Builds page is gone; an app's Source tab shows its commits, builds, previews and where its images go. Old links open the apps list. klops previews list takes an app to show only its previews.
Databases & backups
The endpoint, bucket, region and keys, without the internal address repeated beside them.
Apps & projects
One log answer could hold thousands of lines and fill the assistant's memory within a few questions. Each answer now keeps the newest lines and says when older ones were left out.
The number above an app's configuration counted every line except the ones read from a database or bucket.
Databases & backups
Renaming or dropping a variable of a database or bucket connection left the old one on the app. Now it goes.
The platform firewall blocks PostgreSQL's usual port 5432. The import dialog, the command line and the large-dump error now say so and point to uploading a dump instead.
New apps built from code build again, previews sit on the Source tab, and each organization gets a monthly summary.
Account & organization
Once a month you get your organization's month: what it cost, what runs, what needs a look, and what is new in klickops. It replaces the weekly release notes email.
Apps & projects
Every line shows its name, where the value comes from and the value, with databases and buckets beside the vault's shared values. One Add menu covers every kind.
Builds & deploys
An app's Source tab shows each open pull request's preview and turns previews on or off. A newly connected repository starts with them off, since a preview uses your live database. Repositories you already connected keep theirs.
Builds & deploys
Builds with the default engine stopped before they started. Apps on the older engine were not affected.
See every commit with its build, switch branches from a list and rescan the code.
Builds & deploys
An app built from code has a Source tab. It lists the branch's latest commits with their builds, switches the branch from a list, and rescans the code for a new port, key or database.
Databases & backups
Each one lists its connection values and the apps that read them. Give an app a connection from the vault or from the app's Configuration tab.
Builds finish about 15 seconds sooner.
Builds & deploys
Every build used to scan its image for a parts list it then threw away. That step is gone, so builds finish about 15 seconds sooner.
Builds keep a cache, use Railpack for new apps and pull public base images through a mirror.
Builds & deploys
A build of the branch your app goes live from saves its build steps, and the next build skips installing dependencies again when they stay the same.
Images from Docker Hub, GitHub and other public registries download faster. Private base images still come straight from your own registry.
They plan their build without downloading a package manager first. Apps you already run keep their engine; you can switch one under Advanced in its build settings.
Every assistant tool now says whether it reads, changes or deletes, and a connection lasts up to 90 days.
Command line & AI tools
Every tool now says whether it only reads, changes or deletes, and has a readable name. Your assistant can let reads through and ask you before anything is deleted.
Account & organization
A connection now lasts until you leave it unused for 30 days, and 90 days at most, instead of asking you to sign in again every week.
Changing one domain setting keeps the others, a starting app's logs say so, metrics hover shows only instances running then, and AI assistants can check your plan before creating things.
Billing & plans
An assistant connected to klickops reads what your plan includes and what is already used, so it plans within the limits instead of finding them through a refused create.
Apps & projects
AI assistants and klops domains update change only what you name, so turning off HTTPS no longer clears the domain's sign-in, redirect or controller settings.
Asking for an app's logs before its first instance runs now says what it is waiting for instead of "Something went wrong". A crashing app shows the output of its last run.
Hovering a chart on an app's Metrics tab lists the values at that exact time. An instance from an earlier version no longer appears with a value from minutes before.
Files and other multi-line values in the Vault and an app's Configuration tab open in a multi-line editor and save exactly as written.
Apps & projects
Changing a value that holds a whole file, like a config file, in the Vault or on an app's Configuration tab opens an editor that keeps every line break and space. Before, saving could flatten it to one line.
Each project gets a Vault where editors can reveal a saved value, a repository deploys with its database in one call, klickops checks code before the first build, and built images are private.
Builds & deploys
Apps pull their built images with a pull-only credential. Each running app restarts once to pick it up.
Apps & projects
When an app keeps its entries in a SQLite file or saves uploads next to its code, the check hands you a prompt for your AI tool that moves them to a database or file storage.
Checking a repository lists what will keep the app from starting, such as a start script that runs a development server, and hands you a prompt to paste into the AI tool that wrote the code.
Checking a repository finds login secrets with a fallback value, a session secret written in the code, and debug mode switched on, each with a prompt for your AI tool. The values are never shown.
When the lockfile pins a Next.js, React, Vite, React Router or SvelteKit version with a known serious flaw, the check names the version to upgrade to and hands you a prompt for your AI tool.
A table in your migrations without access rules, a policy open to everyone, or Firebase rules that allow anyone or are still in test mode is named, with a prompt for your AI tool that closes it.
Press Reveal next to a hidden value on an app's Configuration tab or the project's shared secrets, or run klops vault reveal. Each reveal shows in the project's activity; API tokens cannot reveal.
A provider key written into a file, or a committed .env file, is named by file and line with a prompt for your AI tool. The value is never shown. Make a new key: git keeps the old one.
It shows which apps read each value and under which name. Add a value, generate a secret or import a .env file, and pick the apps that use it. Old links to the Secrets and Plain config pages land there.
Databases & backups
Name a database with the deploy and klickops creates a small one, connects the app to it as DATABASE_URL and runs the detected migrate command before each version. It counts against your plan and is not backed up yet.
Command line & AI tools
Ask for it with --public in the CLI or public over the API and the app gets a secure https address on the platform domain in the same call. It opens once its certificate is issued, usually within a minute.
Apps & projects
Values from the project's Vault, from a database and those only the app reads sit in their own groups, with one save for all of them. Move an app's own value into the Vault in one click.
A domain's HTTPS section offers automatic certificates, one from the Vault, or uploading your own. Certificates are listed in the Vault, where you replace an uploaded one when it is renewed.
After Deploy the page shows each app building, starting and getting its secure address, then the addresses to visit. An app's page shows the same steps until its first version is live, never "Stopped".
Checking a repository now finds a SQLite file, an upload folder or a declared volume, says which folder the app can keep between updates, and names the file and line it found it in.
The app carries one cause (out of memory, crashing, a key it asked for, not answering, image not found) and a failed build its reason (no build script, a compile error with file and line, install failed, repository unreachable).
Command line & AI tools
Keys go in as secrets, the release command and whether every push goes live are set, and a database of the project is connected as DATABASE_URL. A re-run finishes what a failed run left out.
Web app & notifications
Every page in German and English uses shorter sentences and everyday words. The website has a new look, a marketplace page and a price calculator.
Apps & projects
A new app on Free is sized so the database, the app and a release command run or a rolling update fit the plan together; before, an update beside a database stalled and a release command never ran.
Builds & deploys
A repository built with TanStack Start or React Router in framework mode used to be taken for a static site. It is now built and started as the server it is, with its own start script.
Databases & backups
The app and its release command wait for the database instead of failing to save or starting without DATABASE_URL; before, you had to wait for the database and try again.
Reviewed apps install in one click, klops signs in like the MCP tools and updates itself, the Deploy page takes your own domain and team sign-in, and a tab left open across a release no longer hangs.
Apps & projects
Double-click the title on the app's page, or set it under Advanced, klops apps update --display-name or the MCP tool. The app's name stays what addresses and commands use.
Under an app's address, choose "Use my own domain" and type the hostname. It has to be under a domain your organization verified; the page shows the DNS record to set.
On the Deploy page and when adding a domain, pick Anyone, Team sign-in or Private. With Team sign-in the app has an address, and only members of your organization can open it.
Command line & AI tools
klops update fetches the newest klops. It downloads the version your klickops runs, checks it, and swaps the binary in. Once a day klops tells you when a newer one is out; KLICKOPS_NO_UPDATE_CHECK=1 keeps it quiet.
Apps & projects
The marketplace, the Deploy page and the public blueprint page name the pinned version, who maintains the blueprint and when it last changed, and link to the source and to where you suggest a blueprint.
Command line & AI tools
klops auth login opens klickops, you pick the organization and allow, and the sign-in shows under Connected apps, where you can end it. Without a server it uses the folder's link, then app.klickops.io.
Web app & notifications
Connect your AI client with the address under Settings, Tools instead; it signs in through your browser. The program was marked deprecated in 2026.10.6.
Command line & AI tools
The Windows download under Settings, Tools exists now, and the page gives a PowerShell command that installs it and checks the download.
Web app & notifications
A tab left open across a release reloads itself on the next click instead of showing a spinner until you hard-refresh.
Builds that finished since 2026.10.7 could not be uploaded; they can now.
Builds & deploys
Since 2026.10.7 every build failed at its last step, so no new version could deploy. Running apps were not affected.
Nothing changes in the app; this release carries the command-line downloads that 2026.10.6 missed.
Command line & AI tools
klops downloads are back. The 2026.10.6 release shipped without its command-line downloads; this release carries them.
One Deploy page for your code, the marketplace and images, faster and safer builds, a klops that waits for your rollout, your AI assistant signing in through the browser, and one list of what needs you.
3 changes may need you to act
Apps & projects
Builds & deploys
Fetching your code, building it and pushing the image run one after another.
Apps built from a repository restart once during this update.
Databases & backups
Account & organization
A channel shows only the host of its address, and a webhook's signing secret stays hidden, also to AI assistants.
You: If an AI assistant has used your organization's notification settings, create a new webhook in Slack or Teams and add the channel again.
Web app & notifications
The language runtime and networking library behind the web app and API are updated to the releases that close these weaknesses.
Builds & deploys
A workspace, a compose file with several builds, or apps in their own folders now list each app with its own settings and keys, and shared packages are named as libraries.
Command line & AI tools
klops plans previews and installs a plan from your own code: apps, databases, secrets and addresses, created in order, with the same progress and messages as the marketplace.
Run klops link once; then klops shows the project's status, and deploy, restart and rollback wait until the app is up and print its address. New: doctor, open and logs --previous.
Web app & notifications
The health chip, klops attention and your AI assistant show the same list per organization: failing apps, databases and buckets, stuck certificates, failed backups and firing alerts, each with its cause and next step.
Apps & projects
Messages about apps, volumes, domains, backups and sign-in read as plain sentences and point at what you can change.
The export no longer asks you to pick between two bundle types, and project settings drop the GitOps mode switch, which the hosted platform never offered.
Builds & deploys
A Source section on the Overview names the repository, branch and how the app builds, and lets you choose between going live on every push or waiting until you press Build now.
A build that runs past 40 minutes now says it was stopped for time, and a build whose machine was taken away starts again by itself.
klops builds get shows how long a build waited for a builder, fetched your code, built and uploaded the image.
A build there runs with up to 4 CPU and 8 GB instead of 2 CPU and 4 GB, and is still billed only for what it uses.
PostgreSQL from Prisma, Drizzle or a driver is offered with its migrations; a hosted one such as Supabase or Neon stays where it is.
Each key the code reads is marked required, optional, set for you or development only, as secret or visible to visitors, with the file and line it came from. klops repos detect counts them.
Set its path, such as docker/Dockerfile or Dockerfile.prod, in the build settings; the build engine you pick when deploying is now kept too.
Sign-in, payments, email, AI and analytics services each list the keys they need and where to get one; klops repos detect shows them.
A new VITE_ or NEXT_PUBLIC_ value on an app built from a repository starts a build of the running version, so visitors see it.
It shows each app with its folder, project and status, names the apps nobody deploys yet, and offers Deploy from this repository.
Command line & AI tools
klops projects update --externally-managed and klops projects export --keep-labels print a deprecation warning and will be removed in a later release. A plain export stays the way out.
The SSO group flags of klops orgs create and klops orgs update, klops projects members and klops quota org warn that they are going away. Set a member's project role with klops projects member-roles.
Every command lists, gets, creates, updates or deletes, uses the console's names, and takes -p and --org the same way; klops databases replaces klops db. CLI-REFERENCE lists every command.
It sees 27 core tools and finds the others when it needs them. Prompts like triage and diagnose come built in, and deleting data needs the name typed again.
klops repos detect --plan prints the apps, addresses and generated sign-in secrets it suggests, ready for klops plans install.
Commands and flags say what they do in the words the web app uses: domains, scheduled jobs, volumes and pull credentials.
They name scheduled jobs the way the web app does and drop text it never needed.
Account & organization
Instead of "You don't have permission", you read which role the action needs and which one you have, so you know what to ask an admin for.
Web app & notifications
Page descriptions, empty pages, dialogs and messages say what you need to know in fewer words.
Apps & projects
Adding a second address to a domain gets a random one too. Addresses you have keep working. To choose a name yourself, connect your own domain.
Builds & deploys
A build of a pushed commit no longer fails with "not our ref" on servers that refuse a commit by its id.
A port in your example environment file counts, a dev server's port in a config no longer overrides the framework's, and a database's port in a compose file is skipped.
A build registry set in your organization's settings was ignored, so builds pushed to the platform's default one. They now push to yours.
VITE_, NEXT_PUBLIC_ and similar settings now reach the build, so the page no longer comes up blank. Keys named like a secret are not sent with your site; the build log says which.
Command line & AI tools
klops orgs members list shows who each member is. It printed empty ids for the members of your organization; it now lists each member's id, email, name and role.
Nothing changes for you; this release moves how klickops is built and shipped.
Builds & deploys
We changed how klickops is built and delivered. Nothing you use changes.
Your organization and billing data is now also kept in a separate backup location.
Account & organization
Changes to your organization, members, tokens and billing are now also kept in a separate backup location as they happen.
Builds keep tokens out of logs and images, check branch names, start Next.js apps, and let you clear build settings.
1 change may need you to act
Builds & deploys
A name git itself would not accept as a branch is refused with a clear message.
klickops refuses a new repository address with a token in it, and never shows or copies one saved before.
You: If a repository address of yours contains an access token, move the token to the connection's credentials and rotate it.
Builds & deploys
Emptying the branch, folder, build or start command in the build settings brings back the default instead of quietly keeping the old value.
The build starts them with Next.js itself instead of a missing script, and Go apps get build and start suggestions that work.
Repositories that keep the app in a folder build without extra setup, and builds keep connection files out of your image.
Builds & deploys
When the repository root holds no app, klickops finds the one folder that does and builds from there. Builds also leave connection files out of your image.
Repositories with a deploy key work again and explain why when they cannot connect, bucket files need editor access, and backup schedules run when they should.
2 changes may need you to act
Databases & backups
Browsing and downloading a bucket's files needs editor access, like its access keys.
You: If a viewer should not have seen a database backed up to a bucket, change the passwords stored in that database.
Apps & projects
When something in a project calls one of its apps by the public domain and is blocked, the Network view shows the app's in-project address to use instead of allowing the detour.
Databases & backups
A full backup bucket stops backups and, later, the database. Data & backups now warns once a bucket with a size limit is 70% full, and backup destinations set up from now on store compressed snapshots.
Web app & notifications
It guarded an address nothing serves any more. Platform administration stays in the operator console, behind a superuser check.
Apps & projects
A minimum of 0, a maximum below the minimum or a CPU target outside 1-100 now fails with a message naming the field instead of leaving a half-saved setting.
It used to be dropped without a warning. It now runs before every later version change, and before the first build of an app deployed from a repository.
A secret or config file mounted when the app is deployed no longer leaves it waiting for storage that never arrives, and storage marked for file backups is backed up from the first deploy.
Builds & deploys
A repository whose build file is named Containerfile or dockerfile was offered as buildable, then failed with "no Dockerfile". It now builds.
An app named like a preview keeps its branch, an uploaded certificate named like one is kept, and deploying from a repository never reuses a connection to a different URL.
A preview skips the app's release command, so an unreviewed pull request never changes your live data. The command runs once the change is merged and deployed.
An app set to build from a branch other than the repository's default now builds and deploys on every push to that branch, not only after "Build now".
Loading branches and commits failed for every repository using a deploy key. When a repository cannot be reached, the message now says why: a rejected key or token, a wrong URL or branch, or an unreachable server.
The deploy screen shows the detected build and start commands as suggestions and only uses the ones you type, so a Vite or React site is served from its built files instead of a guessed command.
Databases & backups
A daily database backup set from klops or an AI assistant ran every hour instead. Existing schedules are corrected automatically, and a custom schedule in the web app now takes the usual five fields.
It archives to the same bucket as the database it came from, counts toward your plan's database limit, and a restore to a taken name or a future time is refused with a clear message.
Account & organization
Permission, "already exists" and "not found" messages for organizations and their channels now use the same word as the rest of the product.
Web app & notifications
The project overview and the Apps list left the address blank for any app whose domain was added under Domains. They now show its first domain, and "+N" when it has more.
The deploy screen used to offer the https link as soon as the app was healthy, while its certificate was still being issued. It now says it is securing the address and shows the link when the certificate is ready.
CPU alerts now fire for apps without a CPU limit, bucket sizes keep to your plan everywhere, and AI tools can set every app setting the web app can.
1 change may need you to act
Account & organization
You: If you run klickops yourself, check each organization's default project quota.
Command line & AI tools
When an AI assistant deploys or changes an app through the klickops tools, it can now also set project secrets and configuration, env from domains, and image updates, as the web app can.
Apps & projects
Apps on paid plans have no CPU limit, so the CPU alert could never fire for them. It now compares usage with the app's CPU request when there is no limit.
Command line & AI tools
Replacing an uploaded certificate through the klickops tools removed its CA chain, because the tools could not send one. They can now.
Billing & plans
Creating a bucket already capped its size at your plan's limit; changing the size afterwards did not.
Account & organization
After deleting your last organization, the setup screen links to your account settings, so you can delete your account instead of creating a new organization.
Web app & notifications
On a plan that caps a bucket, it no longer offers "No limit" or sizes past the cap, and it says what the cap is.
A channel's event filter matches exact names, so a typo or a wildcard stored fine and silenced the channel. Such a name is now refused, with the list of valid ones.
Release notes open with a summary and what needs you, you hear back when something you asked for ships, you can delete your own account, and API tokens stay with your organization.
Account & organization
Settings, Account lists what goes (sign-in methods, tokens, memberships) and what your organizations keep, then signs you out everywhere. An organization's only owner is told which one to delete or hand over first.
Command line & AI tools
klops scheduled-jobs runs lists each run with its status and duration, logs shows the newest run unless you name one, and jobs can read project secrets and keep their image updated, from klops and your AI tools.
Account & organization
A token keeps working after the person who created it leaves or deletes their account. The token list shows who created each one, so you can review them when someone leaves.
Web app & notifications
Release notes list each closed hole under Security with its severity, what was at risk, what was checked for misuse and what, if anything, you need to do.
Each release shows what may need you to act and its highlights first, and folds every other change away by area: in the app, on the changelog page and in the weekly email.
Web app & notifications
When a run kept crashing, whether started by hand or on schedule, the project overview linked its alert to an app named after the run, which showed "App not found".
Builds, previews and restores stay inside their own project, a restore refuses what it cannot bring back, and the command line catches up with the web app.
3 changes may need you to act
--team or the AI tools klickops_patch_scaling, klickops_list_builds, klickops_get_build or klickops_trigger_build need the new names listed under Improved.klickops_create_restore, klops recovery object and klops apps networking set are gone; Improved and Fixed name what replaces them.Builds & deploys
If your build registry runs on plain HTTP or uses a self-signed certificate, turn on "Allow plain-HTTP / self-signed certificates" in its settings.
You: If your build registry uses plain HTTP or a self-signed certificate, turn on "Allow plain-HTTP / self-signed certificates" in its settings.
Build history and logs, previews, restore history and the concurrent build count stay with your project. Disconnecting a repository also unlinks the apps built from it; they keep running their last version.
An app deployed from a repository without a chosen branch builds only from its default branch.
Databases & backups
A restore leaves firewall rules, network isolation and resource limits as they are now, and brings the project in line with the platform's settings once it finishes. klops recovery object and its two AI tools are gone: they rolled back every configuration entry of a kind, not only the one named.
Account & organization
Sign-in methods, passkeys, email preferences, terms and project roles cannot be changed with an API token.
Changing the plan and requesting a top-up or a bigger plan work like managing members. A token for the whole organization can still read the bill.
Command line & AI tools
klops scheduled-jobs now creates, changes and deletes scheduled jobs, sets and removes their config values and secrets, and shows the logs of a single run. klops logs <app> --since 6h reads an app's earlier logs, including those of pods that no longer exist. The AI tools (MCP) gained the same log reads and a single call that gathers a project's recent events with an app's settings, pods and recommendations.
klops support list shows the requests you filed and klops support get prints one with its whole conversation; your AI assistant can read them through the MCP server too. Asking for help and replying still happen in the web interface, where your screenshots and the page you were on come along.
Web app & notifications
When a page fails to load, a Report this button opens a support request that already names the page and the error, so you only have to add what you were doing.
Command line & AI tools
klops no longer accepts --team; use --org. The AI tools klickops_patch_scaling, klickops_list_builds, klickops_get_build and klickops_trigger_build are removed; klickops_set_scaling, klickops_list_app_builds, klickops_get_app_build and klickops_trigger_app_build do the same. klops config set-team, klops use-team and klops quota apply-team-defaults still work but now warn that they will be removed: use the org names instead.
klops and the AI tools. klops apps networking set and klickops_set_app_networking could not change a domain and stored settings nothing used: change a domain under Domains or with klops domains apply. klickops_create_restore skipped everything the project still had, so a restore over a running project brought nothing back: klickops_restore_project_recovery restores the project the way the Backups page does, optionally only named apps and volumes. klops apps networking get now reports the class and certificate issuer of the app's actual domain.
klops adds and removes an app's sidecar and init containers, applies a recommendation's automatic fix, turns backups of a standalone volume on or off, follows or stops a database import, reads and replaces your organization's join allowlist, and shows the bill, past statements, metered usage and wallet. A scheduled job's run logs now open only under the job that ran them.
Billing & plans
Every meter on the Billing page has its own colour, tuned for light and dark mode and easier to tell apart with colour blindness. Build usage no longer shares the grey of backups and the month-end forecast.
When usage is taken off your bill or your included credit is changed for you, the Statements timeline now says what changed and by how much, instead of your figures moving with no explanation.
Web app & notifications
Icon and text buttons no longer draw a grey box around themselves, warnings and highlighted cards show their intended border colour, and loading spinners visibly turn.
Clicking a field's label now focuses the field, and screen readers announce what each field is for, in every form from deploying an app and creating a database to domains, scheduled jobs, certificates and organization settings. Choice buttons such as the builder, backup frequency or weekday pickers announce which option is selected.
Buttons that show only an icon, such as removing a secret, resetting the terminal or jumping to the newest log line, now say what they do and to what. Copy buttons that show a value, like a domain or a certificate fingerprint, now say they copy it.
Database cards, status bars, restart messages and firewall policies now talk about copies and projects, the way app cards already did, instead of switching to infrastructure terms from one screen to the next.
Each person can keep up to 20 log, live view or terminal sessions open at once, and a log view starts from at most the last 5,000 lines.
When your platform administrator replies to one of your organization's support requests, your Slack, Teams, webhook and email channels now say so, next to deploys and alerts; the answer itself stays in the conversation for the person who asked. Feature requests from the Roadmap page now reach the platform team the moment you send them.
Instead of a mail for every release, you get one summary of the previous week's releases.
A new release is announced by the What's new button at the top: it names what shipped for a few seconds, then shows how many releases you have not read until you open them. The notes now open on everything since your last visit, not only the newest release, with a timeline of every release beside them. A tab left open through an update says "Update ready", and one click reloads it straight into what changed.
Apps & projects
The address other apps use to reach an app, shown on its Overview and Networking tabs, pointed at a location that does not exist. Both tabs now show the address the app actually answers on.
An app setting bound to a domain's address kept the original project's address after a clone, and later changes to the clone's domain never reached it. It now names the clone's domain and follows it; a setting bound to a domain the clone does not copy is removed.
A cloned app lost its volumes, configuration keys, database connections, Git repository, update checks and icon, and its next edit could remove environment variables it still used. The copy now keeps all of it; only the results of past update checks start fresh. An app cloned into another organization leaves its Git repository behind, because the repository belongs to the original organization, and the clone says so.
Two saves of an app's configuration, a volume, a firewall rule or a project's capacity that arrived together could answer one of them with an error; both now succeed and the later one wins. Cloning a project now also lists what could not be copied apart from what it leaves out on purpose, and klops projects clone fails when the clone is incomplete.
Automatic sizing could cap a mostly idle app at a fraction of a core, so a long import or export crawled. On paid plans CPU is no longer capped; memory still is. An existing app picks this up the next time you restart it or save a change.
Where your platform administrator connected Grafana, the Metrics tab of an app, a database and a bucket had lost its button. It opens the matching dashboard again, for the same time range.
Builds & deploys
Build now, an app's first build and a watched repository's new commit now build just that app, where they also rebuilt every other app on the same repository and branch. Two projects with an app of the same name each get their own build, a build that finishes late never replaces a newer version that is already running, and an app set to only record new commits is no longer deployed by a push.
Build now answers with an error instead of reporting a build that never began, a push is retried by your Git host, and a watched repository tries the commit again on its next check. Your plan's limit on builds running at once now holds for every build, and a build is stopped after 45 minutes or about 20 GB of disk, so a stuck build no longer holds your organization's build slot.
Signing in with GitLab to pick projects always came back with an "invalid state" error, so no GitLab project could be connected that way.
The deploy wizard now says the repository is too large to inspect and suggests setting the subpath to the app's directory, which inspects only that directory. Before, one large repository could slow detection, branch and commit listing down for everyone. Apps that watch their repository for changes also no longer wait behind one unreachable repository.
When a repository's webhook could not be registered automatically (an SSH key, or a token without webhook permission), the secret the webhook needs was never shown, so every push was refused and the repository said no events had arrived. Connecting now shows the secret once, next to the webhook URL, with a copy button, and klops repos add prints it. For a repository connected earlier, open it and rotate the webhook secret to get a new one. The AI tools never see it.
When klickops set up the webhook on GitHub or GitLab itself, rotating its secret now updates that webhook too, so there is nothing to paste. Before, the provider kept the old secret and every push after the rotation was refused. If the provider does not accept the change, the new secret is shown once to paste by hand.
Pushing or deleting a tag, or deleting a branch, no longer starts a build. A push that arrived while the platform was restarting is now built once it is back, a preview whose pull request was closed in that window is removed within the hour, and a preview that could not be removed is retried when the delivery is sent again.
Databases & backups
Cloning a project with its volumes created volumes that never became ready, so the apps using them stayed stuck starting. A cloned volume is now created fresh and empty, as the clone promises, and its apps start.
Setting a version while editing a database from the CLI or an AI assistant now refuses an older version, and a new major PostgreSQL version waits for a completed backup from the last 24 hours, exactly like the Upgrade action. Asking the Upgrade action for an older version now says it was refused instead of reporting an upgrade that never happened.
Installing an extension through the API or the MCP server without naming a database answered with an error although the extension was installed, and applying a recommended fix that was already in place showed an error instead of the app's current recommendations. Error messages from connected repositories and bucket storage were also hardened.
The backup schedule kept running after a project was deleted, and its recovery points stayed in storage. Both now go with the project. Backup schedules set through the CLI or the API run at most once an hour, and recovery points are kept for at most 365 days.
Recovery points that appeared while the page was open were always shown as holding volume data, even when they held none. A restore now also stops the apps and scheduled jobs that are not running at that moment, so none of them starts using a volume while it is being restored.
Restoring an app, a volume or the project now first checks that the recovery point finished, belongs to this project, saved volume data and holds every volume it would replace, and says why when it does not. A volume created after the point stays untouched by a whole-project restore, an app restore brings back only that app's volumes, and a second restore of the same app or volume is refused while the first one still runs.
Command line & AI tools
Deleting an app, a database or a volume now spells out what goes with it, calls that replace a whole list or cannot be undone are flagged so your editor can ask you first, database queries are marked as read-only, and tools only your platform administrator may use are no longer offered. Changing a project's quota without that permission now tells you who can reset it, instead of pointing at a request form that no longer exists.
The tool (MCP) never passed the container's name on, so every call failed; it now takes the app and the container name separately, and a command and arguments too. The usage tool can also report a period other than the current month.
Through the AI tools (MCP), adding a pull credential always failed, and so did setting an organization's SSO groups, rotating a repository's webhook secret, marking a build registry as insecure, queueing a rebuild with new build settings, replacing a bucket's lifecycle rules, skipping database updates, editing a database's access rules, and scaling an app or cloning a project with options klickops never had. The tools now offer only what klickops accepts, in the shape it expects.
klops command line asks before it destroys something. Commands that delete, revoke, restore over or roll back something now ask for a yes on a terminal, and name what they are about to remove and in which project. Scripts and CI have no one to answer, so they need --yes (or -y) and otherwise stop with exit code 2 before anything changes. Applying a domain from a file also reports why an update was refused, instead of claiming the domain already exists.
klops config use-context, current-context, list-contexts, set-org and set-project could not be reached, because the commands for configuration values had the same name. Those are now klops configmaps list, set and delete; the old klops config list, set and delete keep working for two more releases and print a note naming the new command.
Billing & plans
Projects created on the free plan kept its limit of 1 vCPU and 1 GB per container after an upgrade or a promo code, so a larger app was refused. Upgrading now lifts that limit on existing projects too, and a promo code redeemed after your projects exist raises their capacity right away.
A database that could not start, or a disk that would not fit, told you to raise the project quota or request more room, which you could not do yourself. It now points at a bigger plan, or, on a self-hosted install, at your platform administrator. The API token dialog also no longer claims an admin token can manage members: no token can manage members or other tokens.
A project could run out of memory capacity at about half of what its plan includes. Your plan's CPU and memory now count only what your apps reserve.
If your plan already includes higher limits than your promo code, the page used to claim the code had raised them to its lower tier. It now only mentions the promo when it actually gives you more room.
A valid code on the website now takes you straight to sign-in, and error messages there follow the page language. When a code can't be used, you're told why: it expired, it's used up, it isn't valid, or it only covers another email domain. If the code is for a different address than the one your GitHub or Microsoft account shares, you're asked for that address and sent a sign-in code. After you create your organization, the credit is only reported as added once it has actually landed.
The dialog said your plan was at its limit and pointed at plans that do not exist there. It now says the project is out of room, leaves raising it to your platform administrator, and only closes.
Account & organization
A removed member could keep opening the organization's projects for up to half a minute. Their access now ends on their next click.
After naming your organization on first sign-up, the organization switcher showed a bare address with nothing to pick until you reloaded the page. It now shows your organization's name immediately.
On a self-hosted install, whoever created a project with no organization was refused as if they were not its creator when they tried to rename or delete it.
Web app & notifications
When the platform's infrastructure stopped answering, project pages failed while the status page still said everything was operational. It now shows the platform as degraded until the outage ends.
Building an image in the sandbox could fail on a COPY --chown step and ran every step slowly; both are fixed.
Sign in with a passkey or an emailed code, verify a domain once for your whole organization, and builds run in their own sandbox.
3 changes may need you to act
Apps & projects
Domains and certificates use the certificate authorities and routing options the platform offers. Wildcard certificates are no longer requested there: list each address, or upload a certificate you already have.
Existing rules get the change automatically, with nothing to re-save.
You: Nothing: existing rules were updated automatically.
A "Public internet" rule saved before the last update is brought in line automatically. A domain-name rule names a public website or service: use an address range for a public address and an app rule for apps in your project.
You: Nothing: existing rules were updated automatically.
Apps and builds keep reaching the internet as before. Existing projects get the change automatically.
You: Nothing: existing projects were updated automatically.
Cloning a project needs an admin of the organization it lands in, counts toward that organization's project limit, and starts with the same security, network and capacity defaults as a new project.
Apps inside a project keep reaching each other as before. Existing projects get the change automatically.
You: Nothing: existing projects were updated automatically.
A domain or certificate cannot use an address the platform answers on or one already in use. The three baseline switches on the Network page change only there. Where your platform administrator turned internet access off, ask them if a project needs an exception.
A protected app gets a sign-in that works only for that app, also on domains protected before this update.
You: Sign out everywhere once if you opened an app behind single sign-on that another organization runs.
Signing in to a protected app always returns you over a secure connection.
Apps keep the security settings you give them. Scheduled jobs, release commands, and init and sidecar containers run isolated too, and sidecar images get the same checks as your main image. Saving the Advanced tab sends only the settings you changed.
Apps, scheduled jobs and release commands reach object storage through its storage endpoint. Databases, buckets and backups work as before. Existing apps pick up the change the next time you save them.
You: Save each app once to pick up the change.
Builds & deploys
A Git push delivered twice starts one build, and a failed update check says why in one sentence. klickops no longer follows a Git host's redirect: if you moved a repository, update its address under Repositories.
A pull request from a fork gets a preview with your app's settings only when you allow forks.
Passwords and tokens reach Git exactly as you saved them.
Databases & backups
Access rules accept only password sign-in and name the line they refuse. An import from another server accepts the common connection settings only.
Connection details include the live password, so only editors see them. The query console runs one statement per request.
You: Change a database's password if a viewer should not have seen it.
Command line & AI tools
Pointed at a different server, the CLI asks for a token or a fresh sign-in there, and the AI tools do the same. AI tools tell your editor which of them only read and which delete or overwrite. Device sign-in links always point at your klickops address.
The short code that klops auth login --device prints can only be approved from a browser where you are signed in, not with an API token.
The install command downloads from, and points the CLI at, your klickops address.
Account & organization
The "Allowed to join" list accepts only addresses and domains on the email domain you sign in with, never a public provider such as Gmail. Invite anyone else by email.
You: Check your organization's members for anyone you do not know.
A GitOps-managed project accepts only export, database queries, backups and cancelling a running operation. An API token made for one organization works only there. A project deleted and recreated under the same name opens the new project right away.
An invitation goes to exactly one email address, and so does your organization's billing email; a pasted "Name <address>" is trimmed to the address. Each organization can send a limited number of invitations per day. Organization names and support request summaries fit on one line.
Signing out in the browser or with klops auth logout ends that session on the server too. After seven days you sign in again.
After you sign out, or an administrator signs you out everywhere, terminals and port-forwards ask you to sign in again.
After several wrong passwords for the same email, further attempts wait a while. Production installations require a strong administrator password and can switch password sign-in off for single sign-on.
You: Self-hosted installs: set a strong administrator password.
Roles and administrator access given to an email address apply once your identity provider has verified it. Signing in with GitHub uses your primary address when GitHub verified it, otherwise another verified address.
Web app & notifications
The monitoring endpoint answers only the platform's own monitoring.
You: Self-hosted installs: remove the chart option that publishes platform metrics before upgrading.
Screenshots on support requests must be PNG, JPEG, WebP or GIF, and other file types are refused with a clear message. Screenshots from the support form are converted automatically.
Account & organization
Settings now lists getting-started tips, release notes and incident mail with a switch each, and names the mails that always arrive because you have to act on them or should know: invitations, security notices, billing and support. Every optional mail also carries a one-click link to turn that kind off.
Adding a member now always sends an invitation, and the person joins only after they accept it under Invitations in the sidebar, or on the page they land on when they belong to no organization yet. Signing in never adds you anywhere by itself, and someone on an organization's "Allowed to join" list now gets an invitation too instead of being added. Inviting answers the same way for every address, so it no longer tells you who already has an account.
Members now have Leave organization under Your membership on the Members page. You lose access to its projects right away and need a new invitation to come back. An organization's only owner can't leave until someone else is an owner too.
Your account is now the email address you signed up with. Enter it on the sign-in page and klickops emails you a 6-digit code, no password needed. If you joined with a work address but your GitHub uses another one, sign in with GitHub, confirm your work address once, and GitHub signs you in to that account from then on. Connect or disconnect sign-in methods under Settings.
Web app & notifications
A welcome with the three steps to a running app, a reminder two days later if nothing is deployed yet, the next things most teams set up once your first app is live, and a note in the first week about deploying from Git, the CLI and your editor. Four mails at most, each sent once, and only while they still apply.
The owners and admins of an organization are emailed when an app keeps crashing or cannot start, when a domain still has no certificate an hour after you added it, or when a backup fails. At most one mail per incident per day, with a link to the app, domain or backup. Turn it off for the whole organization under Settings, or just for yourself.
It carries the same release notes the what's-new dialog shows, once per release, and never a digest of things you already saw.
Apps & projects
GitOps mode (externally managed) is no longer offered here, and deleting a project always removes everything in it. To take a project elsewhere, export its manifests from project settings. A project already in GitOps mode can still be switched back to normal editing.
Builds & deploys
Every build now runs in its own sandbox, walled off from other customers' builds and apps.
Command line & AI tools
After you type the code from your terminal, the approval page shows the IP address, the device and when the sign-in started, and lets you deny it. The code is no longer filled in from the link, and a denied sign-in stops the CLI right away with a clear message.
Account & organization
An organization that still has projects can't be deleted; delete or move them, then delete the organization. The message tells you how many are left.
Web app & notifications
Every mail addressed to a person, from an invitation to the monthly statement, now arrives with a plain subject, a short message and one button that takes you where the mail is about. The machine-readable prefix and the trailer of internal details are gone from these, and stay only on the mails your organization routes to a channel. Replies reach a person instead of bouncing.
Resource names, codes, logs and the terminal no longer fall back to a typewriter-style font on computers that don't have the klickops fonts installed.
Apps & projects
Your own domains are not copied and can be added to the clone again.
A deny rule for a domain could end up blocking every destination on its ports instead of just that domain. The firewall editor no longer offers a domain on a deny rule and saving one is refused. To block a domain, leave it out of your allow rules.
An app that had one switched on that is no longer offered has it switched off automatically, which restarts that app once.
Removing a project from its organization is no longer possible on the hosted platform; to change who owns a project, move it to another organization instead.
A domain protected by single sign-on used to keep its old protection settings until someone saved it again. It now gets the current ones when the platform is updated, and nothing else about the domain changes.
Billing & plans
Creating or copying a project into an organization that is on hold for billing now stops with the same message as every other change, until the hold is lifted.
Account & organization
Naming a new organization on first sign-in is now the owner's step and happens once; an invited admin or member goes straight to the dashboard. After that, the owner renames the organization in its settings, as before.
Web app & notifications
You can have up to ten open requests at once, send up to twenty messages a day on each, and attach up to 20 MB of screenshots a day. Over a limit, the message says which one and when it frees up. Messages without screenshots are never held back by the screenshot limit.
Buckets can be created again, and GitHub connections, pull credentials and protected apps only reach what belongs to you.
Apps & projects
An app uses only pull credentials of its own organization or project, or ones your platform administrator shared with everyone.
You: Rotate pull credentials added before 24 September unless you can rule out that they were read.
Signing in to an app behind single sign-on continues only to apps you can open. A sign-in you cannot use lands you back in klickops.
Builds & deploys
Connecting a GitHub account confirms with GitHub that you can reach the installation you add, so an organization sees only the repositories its members were given.
Databases & backups
The Add a bucket form refused every attempt with a technical message, so buckets, and database backups that need one, could not be set up.
On plans with a per-bucket size limit, Add a bucket opened with a size above that limit, so it showed a limit warning and a disabled Create button before you had changed anything. The size now starts at your plan's limit.
Account & organization
Opening a page by its address, reloading it, or following a bookmark could load details, such as plan limits, from the organization you had open before.
The Connect repository dialog refused every attempt with a technical message about an unknown field, so no repository could be added and nothing could be deployed from source. It was sending one field under a name the server stopped accepting.
Certificates renew themselves shortly before they expire, and on a protected site that renewal was being sent to the sign-in page instead of completing. Left alone, those sites would have started showing certificate warnings as each one came up for renewal.
Opening one gave a certificate warning and then a "not found" page, on every app and every language. The preview address was never published to the web server, which also blocked it from getting a certificate of its own. Open a pull request and its preview link works; existing previews recover on the next push.
Database imports finish, images that name their user start, and klops exec copies files into an app.
--from-url. Command line & AI tools
klops exec accepts piped input, so existing data can be moved into a volume when migrating an app onto klickops: tar cz -C ./seed . | klops exec myapp -- tar xz -C /data Previously there was no supported way in, which blocked migrating any app that carries data.
Command line & AI tools
klops apps get tells you when a release command is holding a deploy back, and how to retry or clear it. -o yaml matches one read with -o json. Apps & projects
Ending a Dockerfile with USER app is the recommended way to build a container, but klickops added a rule Kubernetes cannot check against a name, and every such image sat in Pending forever. Those images now run as the user they declare, and keep the rest of their hardening. klops apps deploy|update --run-as-user sets the user ID explicitly when you want to.
An app with a volume on a shared (RWX) storage class crashed the moment it started, with a kernel error nobody could act on: that storage cannot be combined with the extra process isolation klickops applies. klickops now notices the combination and drops the isolation instead of shipping an app that cannot start.
Setting a hostname under an app's networking settings reported success, kept the value, and created nothing. It now points you at Domains, and reading the setting back reports the app's real hostname.
Builds & deploys
The build list shows commits shortened, and pasting one back in queued a build that failed minutes later deep in the build log. It is now resolved for you, or refused immediately with a clear reason.
Command line & AI tools
It refreshed its session without saving the result, so it eventually failed with "unauthorized" while the command line kept working. It now saves what it refreshes, and says the session expired instead of looking like a permissions problem.
Apps that run fine no longer look broken, and a failed release command no longer blocks the screen.
Apps & projects
When a copy is moved off an overloaded machine, the old one can linger in the cluster for weeks. klickops counted every one of them as a live problem, so an app that was running fine showed "Needs attention", a row of red copy badges, and inflated restart counts. Those leftovers are now treated as history: they no longer affect status, restarts, or the copies list. The eviction itself stays visible under Events.
The "Deploy blocked" banner stayed up after you removed the release command, and after the version it complained about had gone live anyway. In that state Retry deploy did nothing. The banner now clears itself as soon as there is nothing left to retry, and Retry deploy starts a fresh run when there is.
The button used to switch tabs and stop there. It now jumps to the release command card and opens the output of the failed run without a second click.
Apps behind sign-in keep you signed in, and connected repositories and their credentials stay inside your organization.
Builds & deploys
An app builds only from a repository connected in its own organization, and project members never see repository credentials. A GitLab connection belongs to the organization that started it.
Web app & notifications
A library behind the web app and API is updated to the release its maintainers published with a security fix.
Apps & projects
Saving such a domain failed with a message about a setting that is not allowed, and would have removed its guest links. Saving works and the links stay.
The check behind protected apps could slowly run out of memory, and a burst of sign-ins could be mistaken for "not signed in".
the public address of the sign-in gateway now defaults to the platform address when it is not set separately, as documented.
Builds & deploys
, even when klickops had to adjust its name.
Billing & plans
A code typed on the landing page is applied once you own an organization, never to one you were merely invited to, and codes cannot be redeemed through API tokens. Campaign figures for your platform administrator no longer count expired credit as spent.
On installations where your platform administrator has not published one, it sat in the menu and then failed to open with an unreadable error. It now stays out of the way until there is something to install.
While klickops is restarting, an action that could not go through now says so plainly instead of failing with a technical error.
The Marketplace installs a reviewed app with its database, domain and configuration in a few clicks.
Apps & projects
Each one has a shareable page listing what it installs, so you can send a colleague the link before they have an account.
A project started from a blueprint shows that app's mark in the project switcher, its header and the project list.
Databases & backups
In a project holding a database and a bucket with the same name, the data section on the overview stayed empty.
Two services sharing a name in one project could show each other's instances and recent activity.
Naming a credential on an app whose image was not changing was accepted and then quietly ignored, so an app deployed before its credential existed could only be fixed by deleting it or by passing the image it already had. It now takes effect on its own.
klops registries create requires either --org <organization> or an explicit --global. Leaving the scope out used to share the credential with every organization on the cluster without saying so, and the confirmation it printed did not mention the scope at all, so the mistake was invisible. Scripts that relied on the old default need the flag added.
Mistyping a command used to print the help text and report success, so a typo in a script did nothing at all and the script carried on as if it had worked. Unknown commands now exit with an error. Anything in your automation that was quietly doing nothing will start failing, which is the point, but it is worth a look before you upgrade.
klops databases available lists the tuning options for each type, with an example for each, so asking for more storage than the size tier gives you no longer means guessing. The size tiers are deliberately conservative; --param storage=20Gi is how you go bigger.
They could be created and deleted but not listed, so a script had no way to check its own work.
Most databases worth moving sit on a private network, and the import refused them for good security reasons without saying what to do instead. It now shows you the one-line command that streams the data in.
A pull credential could be added and listed correctly while every app using it failed to start, because nothing connected the credential to the app. klickops now picks the credential that matches the image's registry automatically, and klops apps deploy --registry names one explicitly when you have several.
Names like orders_app were rejected, and so was every other separator, so a database user carried over from another system had to be renamed to make it in.
Creating a database, updating an app or running a query through the AI tools could fail with "request body is malformed" on input that was completely correct. Requests that are genuinely wrong now say which field is the problem.
Signing in after the tools were already running left them unauthorized for the rest of the session, with no hint that reconnecting would fix it. They now pick up a new sign-in on their own.
Leaving out something the tool needed produced an internal error that read like the tool itself was broken.
It used to print nothing at all, which was indistinguishable from having done nothing.
klops db query and klops db shell take --database, so a database created alongside a user is reachable without going around klickops.
It named one that was renamed some time ago, and the suggestion was garbled on the way out.
"Get help" opens a composer that stays out of your way: it tucks into the corner while you go and fetch the error text or take a screenshot, and your draft is still there when you come back. A request remembers what you were looking at, so you don't have to describe it. The same menu lists every request you have made and the replies on it, and points you at the handbook page for what you were doing.
Being invited into someone else's organization was enough to make every project creation fail outright. The organization you pick is now the one the project is created in.
If you were a member of someone else's organization as well as your own, parked credit could be redeemed into theirs. It now goes to an organization you own, and the organization shown by default stops reshuffling between page loads.
The plan advertises room for both, but a small PostgreSQL claimed the project's entire allowance on its own, so the database was accepted and then never started. A small database now takes half the allowance, leaving the other half for an app. Medium and large are unchanged.
When there was not enough room left in a project, a new managed service sat at "Provisioning" forever with no explanation anywhere. It now tells you the limit was reached and what to do about it.
Scanning the printed link still works, but the code is no longer the only way in: there is now a field for it above the sign-in options, and it shows what the code grants before you sign in. Checking a code never uses it up.
If a colleague had already set up access on a shared account, that was the only account offered. You can now install on your own account or a different organization from anywhere in the connect flow, and the "Manage on GitHub" link points at the right place for organization-owned setups.
A name like example.info silently left the Connect button disabled. Dots and other unsupported characters are now folded to hyphens as you type, and the rule is spelled out inline when the result still isn't usable.
Credit codes can now be minted in a batch and downloaded as a print-ready document — one coupon per page, each with the code and a scannable link that claims it. Codes are still redeemed exactly as before, on the sign-in screen.
The gondola replaces the old mark and runs through the whole product: the logo, the sign-in screen, the loading state and the browser tab. The interface moved to a lighter, paper-like surface at the same time.
Every metered hour is attributed to the app, database, bucket or scheduled job that caused it, so you can see what each thing costs instead of one number for the whole project. Anything that cannot be placed under a workload is still shown, rather than quietly folded into the total.
Usage has its own page, separate from Billing, where you pick a period and drill from your organization down to a project and then to a single workload, charting any level.
Upgrading to a paid plan or adding credit is now a request that a platform administrator confirms once the payment has arrived. You can see your pending requests and withdraw one you no longer want.
Plain values and sensitive values live on one page instead of two, with a clear separation between them, and the page no longer asks the same question twice.
An app row shows whether it is actually doing any work, a domain shows which of its three stages it has reached, a database offers its connection string, and a disk can be resized without leaving the list.
, so a schedule that exists but has not run yet no longer reads like a successful backup.
Creating, reviewing and revoking tokens is clearer, and a newly created token can be copied straight from the screen.
, with a tidier board.
One scope control replaces the two stacked switchers, and every destination now appears in exactly one place. Quotas moved out of Settings, and the three separate registry screens became a single Registries page.
rather than a wall of cards.
instead of one side at a time.
in plain language.
, and pages got their short introductions back.
Asking for a quota increase separately is gone: if you need more, the answer is a larger plan, which makes the limit you get predictable from the plan you chose.
While the per-workload breakdown was being rolled out, the hour in progress could be recorded in both the old and the new shape, so that hour's reserved capacity and cost were doubled. It was visible as an unattributed line carrying almost the same figures as the workloads beside it. Affected hours have been corrected, and an hour is now written once no matter how often it is measured. The same fix stops a project being billed for a workload that has since been deleted.
A code accepted while your organization was being created could end up not applied; it is now picked up on the next page load rather than needing support.
instead of just your statements, and spending details sometimes disappeared from the usage page.
, which is used for rate limiting and recorded against promo redemptions and terms acceptances. Self-hosted operators behind more than one proxy can set KLICKOPS_TRUSTED_PROXY_HOPS; the default suits the shipped setup.
with KLICKOPS_BUILDS_RUNTIME_CLASS, for operators who want builds further isolated from the rest of the cluster.
Platform operators used to open straight onto the console showing every customer; they now start where everyone else does and step into the console on purpose, from the user menu, the organization switcher, or the Organizations page.
The switchers, the command palette, the activity feed in the top bar, and the page you land on after signing in all build their lists from your own memberships.
Name your organization, tick the terms, done — the multi-step wizard and the separate terms popup are gone. A promo code from our website is applied automatically and shows as a credit badge, so you never have to type it twice. Inviting your team moved to where it belongs: the Members page, once there is something to show them.
This month's spend against your included credit, and your balance — with promo credit folded in and its expiry date right next to it. The charts and the per-meter, per-project, and per-workload breakdowns are one click away under "Spending details", and the Plan tab is now purely about choosing a plan.
The most urgent thing about your account shows as a single banner instead of a stack of warnings.
Every upgrade link in the product lands directly on the plan ladder instead of the cost overview.
Costs under one franc display with four decimals on the billing page, matching your PDF invoices instead of rounding to 0.00.
When your quota request is approved or denied, the email now includes the reviewer's note.
A month that was settled after you changed plans could be priced at the new plan instead of the one you were on. Statements now freeze with the plan that was actually in effect.
Entering the same code twice no longer shows an error after the credit was already granted, a code scoped to your company address now tells you to sign in with that address instead of a generic message, and an organization suspended for billing can redeem a code to get its workloads running again.
A sequence of suspend-and-restrict actions could leave apps stopped even after access was restored; they now come back up as expected.
Got a code? Enter it at sign-up or later under Billing to unlock free credit for a limited time. Your billing page shows the remaining promo balance and when it expires, and promo credit is always spent before your own. A code can also raise your plan limits for its duration, and during the closed beta a valid code gets you straight in.
If the klickops GitHub App was not installed on any of your accounts yet, connecting ended in "Connect failed, please try again". You are now taken to GitHub's install page instead, and land back in the repository picker after installing.
klickops now reads your container image and applies the right protection on its own: images that run as a normal user are fully locked down, and images that need root are isolated so root inside the container has no power on the platform (on clusters that support it). The deploy wizard no longer asks security questions — it simply shows what will apply, and overrides live on the app's Advanced tab.
New apps get a connection check on their port with zero setup, so a crashed app is restarted automatically. Switch to an HTTP path check in the same control if you have a health endpoint.
on the app page — it was previously tucked away at the bottom of Metrics.
Rules, live traffic, and creating new rules all live together; allowing a blocked connection drafts the rule right there, and old Firewall links redirect. Creating and editing a rule now use the same editor at the same size.
Buckets are "Buckets" everywhere (previously also "Object Storage"), running app instances are "copies" throughout, Secrets and plain config appear as one "Configuration" concept, and the Alerts page is titled Alerts.
Tabs for features your cluster doesn't offer (backups without a backup engine, firewall without a network layer) no longer render as dead ends, and crowded tab strips now show when there's more to scroll.
, freeing sidebar space. A dot on your avatar tells you when release notes are waiting.
Previously the value silently became 1 while the UI confirmed success. Stopping all copies now asks first, and the save button says what it will do.
, matching app and volume deletion — it destroys data and deserved the same guard.
Every preview now runs with KLICKOPS_ENVIRONMENT=preview and KLICKOPS_PREVIEW_PR set, so your app can skip schedulers, outgoing email, and other side effects when it's running as a pull-request preview instead of production.
The full REST API is now documented on its own reference page, generated from the same definitions the product runs on.
It now sits in the sidebar right next to What's new, in the command palette, and as a link from the release-notes dialog.
klickops now tells you what changed: a short summary pops up once after each platform update, and the full release history is always one click away in the sidebar.
The new Roadmap page shows what we're considering, planning, and building. Vote for what matters to you, request features you're missing, and watch shipped items close the loop.
Onboarding now asks for your organization's name instead of auto-creating one for you.