Change Log

Weekly Changelog: New Geolocation API, Subscription Controls, and Usage Previews

Discover our latest platform updates! This week introduces a new Geolocation API for better onboarding, subscription status checks, and a new dry-run usage report.

-

✨ New Features

  • Geolocation Hints API: A new API endpoint is now available to help you create smoother onboarding experiences. It suggests the appropriate currency and locale based on a user's location, allowing you to pre-fill settings for new users.
  • Check Your Subscription Status: You can now programmatically check if your current organization has an active subscription or is in a trial period with a new, simple API endpoint.
  • Preview Usage Reports Safely: Our command-line tool now includes a --list option for the site usage command. You can now generate and review a report of site counts and usage across all your organizations without applying any changes.

🚀 Enhancements

  • More Powerful Subscription Management: We've rolled out several improvements to our subscription APIs. You can now add custom metadata to subscriptions, and we've enhanced the invoice preview endpoint to better support different billing intervals like monthly and yearly.
  • Enhanced Subscription Protection: For added security, subscriptions can now be marked as "read-only." This prevents any modifications to the subscription, offering greater protection for managed or sensitive accounts.
  • Improved Security Resource Stability: We've made our security services more resilient. The system will now automatically attempt to recover certain resources if they get stuck in a temporary error state, and we've clarified the error messages to be more helpful.
  • Paving the Way for Better AI Analytics: We've implemented a foundational system for tracking usage of our AI features. This behind-the-scenes work is the first step towards providing you with more detailed usage analytics and insights in the future.

🛠️ Bug Fixes

  • More Reliable AI Conversations: We've improved the stability of our AI-powered features. By adding more robust error handling, we've reduced the chance of conversations failing due to unexpected API behavior from the underlying language models.