SEAL Extension Takes the Next Step: Advanced Search, GEO Features and AI Vector Integration

Last year, I wrote about how the TYPO3 Meets SEAL initiative laid the groundwork for a modern, modular search architecture in TYPO3. This resulted in the Index extension (index), the SEAL extension (seal), and a first proof-of-concept for AI-powered search with the SEAL AI extension (seal_ai). The response from the community was positive. Agencies experimented with the extensions, developers shared feedback, and it became clear: this was only the beginning.

At the start of 2026, I submitted a follow-up budget idea to the TYPO3 Community Budget (EXT:seal Ecosystem Expansion - Advanced Search & AI Vector Integration), with a clear goal to take the Seal extension ecosystem from a promising foundation to a production-ready search solution. I am happy to report, all three milestones have been reached.

The first milestone focused on the Index extension, the central indexing framework that sits beneath the entire SEAL ecosystem. Version 2.2.0 brought significant improvements both in terms of coverage and code quality.

New extension support was added for:

• Address List (tt_address): Address and contact records.
• Container Content Elements (container): One of the most widely used container organization extensions.
• TYPO3 CMS Content Blocks (content_blocks): For modern, structured content elements.
• Calendarize — Event Management (calendarize): For event-based content.

These extensions represent real-world content patterns that appear in the vast majority of modern TYPO3 projects. By supporting them natively in the Index extension, the need for project-specific indexing customization is significantly reduced.

Beyond new features, the release also brought numerous tests, bug fixes and structural improvements, enhancing both the stability and testability of the codebase. The documentation was revised and expanded, making it easier for developers to get started.

The feedback from the community on the Index extension has been very positive. The extension is already being used in production setups, and the improved stability has been well received.

The second milestone was arguably the most visible one: the SEAL extension. With this release, the extension moves from an early-stage integration to a solid search solution.

One of the most requested features is now available: faceted search. Users can now filter search results by tags, enabling the kind of refined, multidimensional search experience that modern websites demand. Whether it is filtering blog posts by category or narrowing down products by attribute, faceted search is a cornerstone of professional search UX.

The SEAL extension introduces geographical search capabilities, including a radius and distance search, that finds content within a given distance from the location of the indexed records. This opens up entirely new use cases for TYPO3. From store finders to event maps to location-aware content delivery.

A new search statistics feature now collects data about what users are actually searching for. The results are surfaced through basic dashboard widgets, giving editors and administrators genuine insight into search behavior — without needing a third-party analytics tool.

The 1.0.0 release also includes a meaningful test suite covering both indexing and search processes, and the documentation has been significantly improved to help new users get up and running with the SEAL extension and the Index extension together.

The third and final milestone completes the picture: a properly stabilized and production-oriented release of the SEAL AI extension (seal_ai), the AI-powered vector search integration for TYPO3.

When I first published the SEAL AI extension last year, it was clearly labelled a proof-of-concept. There was no tagged release of Symfony AI at the time, and the ecosystem was moving fast. That situation has changed, and the SEAL AI extension has evolved with it.

• Updated Symfony AI alignment: The extension has been migrated to the current state of symfony/ai, ensuring compatibility and stability with the underlying platform.
• DSN-based configuration: Switching or extending vector backends is now as simple as configuring a DSN in your TYPO3 Site Configuration. This significantly reduces the complexity of setting up and managing AI vector stores, and makes the extension much more approachable for agencies and integrators.
• Google SGE-like output: When a user performs a search, the SEAL AI extension now generates an AI-synthesized summary at the top of the first results page — similar to the AI Overviews seen in Google Search. This gives users an immediate, contextual answer before they dive into individual results.

Vector search, also known as semantic search or embedding-based search, understands the meaning behind a query, not just its keywords. A search for "car" will also surface content about "automobile" or "vehicle". A question like "How do I reset my password?" can return relevant help articles even if none of them contain those exact words.

This is not a distant future technology. It is here, it works, and with the SEAL AI extension it is now accessible to the TYPO3 ecosystem in a structured and sustainable way.

This work deliberately aligns with the strategic direction of TYPO3 v14. Native support for common extensions and a unified DSN-based configuration directly address the goal of reducing complexity, cutting down the amount of custom code projects need to write. Better integration between TYPO3 content structures, search abstraction, and AI services improves interconnectivity across the ecosystem. And with advanced search features like faceted and geographical search alongside AI-powered vector search, the SEAL ecosystem enables future-proof, scalable architectures for both small projects and enterprise-grade TYPO3 installations.

I would like to sincerely thank the TYPO3 Association and all community members who voted for this budget idea. Without the Community Budget program, this work simply would not have happened at this pace or at this level of quality.

I also want to thank everyone who has already tried the extensions and shared feedback. Your real-world usage scenarios directly shaped what was built in this round.

All three extensions are available in the TYPO3 Extension Repository and on Packagist (Composer is required for the SEAL AI extension to install platform and storage provider):

• Index extension: TYPO3 Extension Repository | GitHub
• SEAL extension: TYPO3 Extension Repository | Packagist
• SEAL AI extension: TYPO3 Extension Repository | Packagist

I am looking forward to your feedback. Whether it is a bug report, a feature idea, or simply a message to say you got it running. The best place to reach me is via the GitHub repositories.