Automated OpenAI Model Deprecation Alerts
Every night, WebRun signs in to the OpenAI Platform, reads the published deprecation notices and changelog, checks which of those models your projects still call in their usage, posts Microsoft Teams a list of the projects exposed with the days remaining, and messages you on Telegram when a retirement date gets close.
How do I know when a model my code depends on is being retired?
WebRun signs in to the OpenAI Platform every night, reads the published deprecation notices, and checks them against the models your projects actually call. It posts Microsoft Teams the exposed projects with days remaining and the recommended replacement, and messages you on Telegram as a retirement date gets close.
- The migration is planned rather than triggered by a failing call
- Every exposed project is named with days remaining and its replacement model
- Nothing your projects call slips past a retirement date unnoticed
Built for engineering leads · AI product teams · platform engineers · technical founders
What does WebRun do on every run?
The exact actions WebRun takes, in order - in plain language, so you can adjust anything.
-
WebRun signs in and gets to work
Opens
platform.openai.comin a real browser with your saved login - no setup, no API keys. -
1
OpenAI Platform - match deprecations to live usage
WebRun opens OpenAI Platform to match deprecations to live usage. - Sign in to the OpenAI Platform and read the published deprecation notices and changelog entries
- List the models with a retirement date and note the recommended replacement for each
- Open your projects and read their recent API usage by model
- Match the two lists: every project still calling a model with a retirement date, and the days remaining
Done when Every project calling a retiring model is identified with the date and the replacement.
-
2
Microsoft Teams - post the exposed projects
WebRun opens Microsoft Teams to post the exposed projects. - Post the engineering channel a list of projects exposed to a retirement, soonest date first
- Show the model, the retirement date, days remaining, and how much traffic the project sends it
- Name the recommended replacement model beside each row
- Say plainly when nothing your projects call has a retirement date
Done when Engineering has tonight's deprecation exposure in Teams.
-
3
Telegram - message you as a date gets close
WebRun opens Telegram to message you as a date gets close. - Message yourself when a retirement date falls inside your warning window
- Name the project, the model, and the days left
- Repeat the message as the date gets closer if the project is still calling that model
- Stay quiet on nights when nothing is inside the window
Done when You have been warned in time to plan the migration.
How is each run configured?
Secure by default
Connect once, stays signed in
WebRun signs in once and keeps each session in a persistent environment, so every run picks up right where it left off.
Every action is checked against this policy before it runs.
Questions, answered
Does it change my code or my API settings?
No. WebRun reads the deprecation notices and your projects' usage, then reports the overlap. It edits no code, changes no project setting, rotates no key, and switches no model. Migrating stays with your engineers.
How does it know which models my projects use?
It reads the recent API usage recorded against each project on the platform and matches the models found there against the models with a published retirement date.
How much warning does it give?
It reports every match the night it appears, and the Telegram message starts once a date falls inside the warning window you set, so the migration is planned rather than triggered by a failing call.
Put this on autopilot.
Turn it on in minutes - or have our team set it up for you.