December 2025 Summaries

Supabase and Stripe have collaborated to create the Stripe Sync Engine, an open-source tool designed to integrate Stripe data directly into Postgres databases, allowing applications to efficiently manage and analyze billing information. This integration enables users to perform complex queries and data analysis without the limitations of API rate limits, using mechanisms like webhooks for real-time updates and scheduled backfills for historical data. The sync engine is particularly useful for applications that require a first-class integration of billing data, offering faster and more reliable data queries compared to the Stripe Foreign Data Wrapper, which translates SQL queries into API calls. With features like incremental sync, flexible JSONB storage, and a new CLI, the engine is optimized for maintaining up-to-date Stripe data within local databases, making it easier to identify user behaviors such as account conversion, churn risk, and revenue breakdown by subscription plans. Additionally, Supabase Queues are used to manage and distribute the data syncing process efficiently, and ongoing collaboration aims to enhance the engine with more features like a dashboard UI and pre-built SQL views for common metrics.

ChatGPT apps are extensions that allow ChatGPT to display interactive interfaces and connect to external services, consisting of an MCP server and a web component, with ChatGPT serving as the intermediary that determines which tool to call. The mcp-use is an open-source TypeScript SDK used for building MCP servers, enabling compatibility with edge environments by using Hono instead of Express. This allows MCP servers to operate on serverless platforms like Supabase Edge Functions and Cloudflare Workers. The guide outlines building a ChatGPT app for exploring Supabase databases, featuring tools like List tables and Execute SQL, which provide both data and interactive React widgets. The deployment process involves Supabase Edge Functions, known for their speed and scalability, and the app integrates with ChatGPT by enabling developer mode and connecting through a new ChatGPT App. The document also discusses organizing UI widgets, registering tools, and suggests extending the app by adding more tools, custom widgets, and implementing write operations with proper authorization checks.

Supabase's Metrics API enhances observability by integrating approximately 200 Postgres performance and health metrics into existing monitoring systems using the Prometheus exposition format, providing a unified view of application, infrastructure, and database metrics. This integration allows for better correlation of events, such as API latency spikes and query performance, across the entire stack. Supabase supports open-source and open standards, ensuring metrics remain portable and compatible with various tools like Grafana, Datadog, and AWS Managed Prometheus. This approach facilitates effective monitoring and alerting for issues such as connection saturation, CPU pressure, and replication lag, while also promoting flexibility in tool choice without vendor lock-in. The Metrics API is available on all hosted Supabase projects and can be set up with Grafana Cloud, self-hosted Prometheus, and Datadog Agent, with detailed integration guides available in the updated documentation.

iceberg-js is a lightweight and developer-friendly library created to fill a gap in the JavaScript ecosystem for managing Iceberg catalogs, which were previously dominated by tools for JVM languages like Java and Scala. This open-source library is designed to be generic, minimal, and type-safe, supporting any Iceberg REST Catalog implementation without being tied to specific vendors or engines. It focuses solely on catalog management without including engine-specific integrations or advanced features like branching or time travel queries. Compatible across various JavaScript environments, including Node.js, modern browsers, and major bundlers, iceberg-js supports multiple authentication methods and is available under the MIT license. The creators emphasize maintaining its focus on catalog management and recommend pairing it with a query engine for data operations. Contributions aligning with its goals are welcomed on GitHub.

Broadcast Replay for Realtime is a new feature designed to prevent users from missing critical messages in real-time applications, even in the event of connection drops, mid-session joins, or page reloads. This feature addresses the challenge of users losing important context due to network interruptions by allowing private channels to access earlier messages, thereby improving user experience without the need for complex state management solutions. Broadcast Replay can be applied in various real-world scenarios such as chat applications, live sports updates, and collaborative editing by displaying recent messages to users as they re-enter or reconnect, enriched with metadata for distinguishing replayed messages. Currently available in Public Alpha for Supabase JavaScript client version 2.74.0 or later, the feature requires appropriate RLS policies and invites user feedback for further refinement.

Supabase for Platforms is a white-label backend solution designed to allow platforms to provision and manage fully managed backends for their users, leveraging the complete Supabase stack, including Database (Postgres), Auth, Edge Functions, Storage, and Realtime. It caters to AI builders and platforms like Lovable and Bolt.new, which have experienced rapid growth by offering seamless backend solutions that handle infrastructure complexities, enabling users to focus on feature development. The offering includes scalable compute options, a lightweight embedded Supabase Dashboard, and the ability for users to claim their projects for full ownership while the platform maintains API access for ongoing support. This model is particularly beneficial for platforms where users are either unfamiliar with infrastructure management or prefer a more hands-off approach, while also accommodating developers who want more control through OAuth integrations and a robust Management API.

Supabase has enhanced its authentication capabilities by integrating OAuth 2.1 and OpenID Connect into Supabase Auth, allowing projects to function as full-fledged identity providers. This development opens up various opportunities, such as AI agents authenticating through the Model Context Protocol (MCP), facilitating secure third-party developer integrations, enabling enterprise single sign-on, and offering granular access control. The implementation seamlessly integrates with existing Supabase Auth features, leveraging methods like password authentication, magic links, and social providers, while complying with modern security standards, including PKCE and JWT-based access tokens. The integration supports Row Level Security (RLS) policies and extends to OAuth clients, enabling secure and customizable access to user data. Supabase Auth provides an OIDC discovery endpoint, facilitating easy integration with enterprise systems, while ongoing improvements include refining granular permissions and enhancing MCP server development directly within Supabase, fostering an open development environment.

The text discusses the development and implementation of asynchronous streaming capabilities for Foreign Data Wrappers (FDWs) in PostgreSQL, particularly in the context of integrating with analytical systems like ClickHouse. Traditional FDWs operate in a synchronous, batch-oriented manner, which can lead to high memory usage, poor interactivity, and inefficient resource utilization due to blocking during data fetches. To address this, a new approach using Rust's async runtime has been introduced, allowing FDWs to stream data incrementally through a bounded channel, which maintains predictable memory usage and improves query responsiveness by delivering results as soon as data becomes available. This innovation transforms FDWs into components suitable for modern data pipelines, enabling PostgreSQL to efficiently query remote data sources with low latency and scalability, and positions Wrappers as a universal data gateway within the SQL ecosystem.

Kiro powers, when integrated with Supabase, enhance the capabilities of the AI assistant by providing immediate access to a user's Supabase database schema, best practices for edge functions, security guidelines for Row Level Security (RLS) policies, and SQL syntax optimized for Postgres. These powers facilitate the management of database schemas by suggesting migrations and structuring tables according to Postgres best practices, and they allow Kiro to review edge functions for performance and best practices, including guidance on using the Deno runtime and environment variables. Additionally, they automate security and performance checks, identifying potential issues like missing RLS policies and inefficient queries. Kiro powers are built on the Model Context Protocol (MCP), which intelligently loads and unloads context to avoid overwhelming the user, ensuring a streamlined development experience. By using these powers, developers can quickly access best practices, security checks, and platform knowledge, allowing them to build full-stack applications more efficiently within Amazon's Kiro IDE.

Supabase has released seven new security-focused email notification templates for its Auth service, designed to alert users to sensitive account actions such as password and email changes, identity linking or unlinking, and modifications to multi-factor authentication (MFA) methods. These notifications provide context about the actions taken, allowing users to verify their legitimacy. The updated Dashboard allows for individual enablement and customization of each notification to align with brand identity. Users can programmatically manage these templates using the Supabase Management API or the Supabase CLI, and configure the Auth email send hook for additional customization and internationalization. Future plans include expanding notifications to cover events like new device logins or suspicious activity, and Supabase invites user feedback to enhance these features.

Supabase has introduced Analytics Buckets, a feature designed to manage and store large data sets in Supabase Storage, providing a solution for analytical workloads that differ from the capabilities of Postgres, which excels in handling transactional data. Analytics Buckets offer cost-effective storage, schema evolution, time travel, and full audit history, and they work in tandem with Postgres by storing historical data and running heavy analytical queries. Data is stored in Parquet files on S3, with Apache Iceberg handling metadata management, allowing for scalable and efficient querying using Iceberg-compatible tools like Apache Spark and Amazon Athena. The system supports automatic data replication and streaming via Supabase ETL, facilitating near real-time updates without manual intervention. This structure enables the separation and independent scaling of compute and storage, optimizing performance and cost by maintaining recent data in Postgres and archiving older data to Analytics Buckets, making it ideal for large-scale data analysis.

Supabase ETL offers a streamlined process for setting up Extract, Transform, Load (ETL) operations directly through the Supabase Dashboard without requiring any coding. It differentiates between OLTP and OLAP by recognizing Postgres's limitations with large-scale analytics, providing a solution that captures changes in the Postgres database and delivers them to analytics destinations like Analytics Buckets and BigQuery in near real-time. The setup involves creating a publication to define tables to replicate, connecting these to a destination, and monitoring the pipeline's status. ETL facilitates faster queries on large datasets and reduces storage costs by transferring data to systems designed for analytics, while maintaining a complete change history with an append-only format. Although in private alpha, Supabase ETL is open source and is designed to work alongside read replicas, catering to both application read scaling and analytics needs.

Vector Buckets offer a scalable solution for storing and querying large sets of high-dimensional vectors, enabling features such as semantic search, recommendations, and media similarity detection, by integrating with Supabase and Postgres. Designed to handle workloads up to tens of millions of vectors, Vector Buckets store embeddings in S3-backed object storage, allowing for efficient similarity search without overloading traditional database schemas. They support various applications, including AI documentation search, product recommendations, and media de-duplication, by using vector indexes that can be queried through familiar Supabase SDKs or directly via Postgres. The service, currently in Public Alpha, is free to use under a fair use policy, with plans to evolve based on user feedback and real-world usage.