✨ Maia Gets Skills and @ Mentions, plus Pipeline Quality Enhancements

Welcome back to the Matillion New Features Blog! This week, we're excited to introduce powerful new capabilities for Maia and significant enhancements to pipeline quality management. From AI-powered table mentions and specialized sub-agents to automated quality fixes and reusable skills, these updates are designed to make your data workflows more efficient and intelligent. For a full list of recent changes, be sure to check our changelog updates.

🧠 Maia Update: Skills Are Here

We're happy to introduce skills — a way to extend Maia with reusable, specialized knowledge for your projects. Skills give Maia step-by-step guidance for specific tasks, so it follows your team's best practices every time.

Complex or domain-specific tasks often need detailed, consistent instructions that go beyond what an AI assistant knows out of the box. Skills let you encode proven approaches once and have Maia apply them automatically whenever they're relevant.

What's new:

• Built-in skills: Maia ships with curated built-in skills for common tasks, such as Iterator Components guidance for building pipelines that loop over data, retry on failure, and more. We'll be adding more built-in skills over time.
• Project skills: Create your own custom skills by adding a SKILL.md file to your project's .matillion/maia/skills/ directory. Encode project-specific patterns, conventions, or step-by-step procedures that Maia will follow.
• Automatic activation: When a task matches a skill, Maia activates it automatically — no manual steps needed. You'll see which skill is being applied as Maia works.

How to use it: Built-in skills work automatically. To create your own, add a folder under .matillion/maia/skills/ in your project with a SKILL.md file containing a name, description, and instructions. Maia will discover and use it in future conversations. For more information, read the Skills documentation.

🤖 Maia Update: @ Mention Your Database Tables in Chat

We're happy to introduce a new way to work with your data in Maia — you can now reference specific tables directly in chat using @ mentions.

Why is this helpful? If you already know which tables you want to work with, @ mentions let you reference them precisely — no ambiguity. And if you're not sure exactly what you're looking for, Maia still has you covered — just describe what you need and it'll find the right tables for you.

What's new?

• Table mentions: Type @ in the chat to reference specific tables directly.
• Search-driven navigation: Results load as you search — drill down through databases and schemas to find the table you need.
• Full path support: Build precise references using dot notation (e.g. @DATABASE.SCHEMA.TABLE).
• Multi-platform: Works with Snowflake, Redshift, and Databricks.

How do I use it? Type @ in the Maia chat input to open the mentions picker. Search and drill down through your databases and schemas to select the table you want to reference.

🎉 Pipeline Quality Reviewer Improvements

Three new pipeline quality features are now available to help you maintain higher code quality and streamline your development workflow within the Data Productivity Cloud.

• Maia Autofix leverages AI to automatically detect and fix common pipeline quality issues. When Maia identifies fixable problems such as incorrect component naming, missing start/end components, or unused components, it will plan the necessary fixes and request your confirmation before applying changes. This allows you to resolve common issues in seconds rather than spending time manually editing components across multiple pipelines.
• Review All enables you to validate multiple pipelines simultaneously. You can now review all pipelines in your project against quality rules to efficiently identify issues across your entire codebase, saving significant time when managing large projects.
• Pre-commit Review helps prevent quality issues from reaching your main codebase. By enabling pre-commit review via a checkbox in the commit modal, the system will automatically check for pipeline quality rule violations. Any attempted commits with error-level violations will be blocked, ensuring your main branch stays clean and maintains high quality standards.

For more information about all our pipeline quality features, read our documentation.

💬 We'd Love to Hear From You!

Let us know how these new features are improving your workflows—we're all ears! Feel free to add any comments or questions below.

Want to get involved?

Join the Matillion Community to stay up to date, share feedback, and help shape our product roadmap for future innovations.