The agentification of Test Data Management is here. Meet the Structural Agent.

TL;DR: Tonic.ai announces the launch of the Structural Agent, an intelligent AI copilot that fuels AI-native software development by transforming how teams configure and provision anonymized test data. In bringing agentic AI to Tonic Structural, we’ve turned complex data masking into a conversation. No more manual mapping, no more SQL guesswork—just high-fidelity, privacy-safe data in a fraction of the time.

Why the future of TDM is agentic

Let’s be honest: configuring test data has historically been a bottleneck. Even with the best tools in place, getting a dataset ready for development can be a time-consuming, manual process of hunting through legacy schemas, writing complex SQL, and meticulously mapping generators to thousands of columns. Configuration accounts for an estimated 80% of the total work required in Test Data Management. It is the single biggest hurdle to development velocity.

At Tonic.ai, we believe your time is better spent building products than shaping data transformations.

Earlier this year, we took our first step toward an AI-driven workflow by introducing LLM-enhanced sensitive data detection. By leveraging the contextual understanding of Large Language Models, we enabled Tonic Structural to catch 50% more sensitive data than standard pattern-matching alone.

But detection was only the beginning. Today, we’re taking the next leap: the agentification of Test Data Management. By agentifying the configuration phase, we are effectively eliminating 80% of the manual work, turning complex data masking into a simple, plain-language conversation completed at the speed of AI.

Meet the Structural Agent: Your shortcut to anonymized data

The Structural Agent isn't just a chatbot; it’s an active member of your engineering team. It understands your schema, knows our product suite inside and out, and—most importantly—takes action on your behalf.

By combining the domain expertise of LLMs (powered by Anthropic’s Claude) with Tonic.ai’s industry-leading masking algorithms, the Structural Agent allows you to describe your needs in plain language and watch the configuration happen in real-time.

How the Structural Agent works with you

We designed the Agent to remove the friction of the "blank page" problem. Here is how it streamlines your workflow:

• Intelligent PII clustering: Instead of overwhelming you with a list of 1,000 columns, the Agent groups scan results by PII type—such as emails or first names—presenting a high-level summary of your risk profile.
• Bulk generator application: Once risks are identified, the Agent proposes best-practice presets. You can apply the recommended generators to every relevant column across your entire workspace in a single click.
• Natural language adjustments: Need to shift a date range or change an output format? Just ask. The Agent handles the granular settings across columns so you don’t have to track them down manually.
• In-app expertise: Ask the Agent how a specific feature works or where the most sensitive data lives in your environment. It provides instant context without you ever leaving the workspace.

The Structural Agent removes the repetitive manual effort that has defined data masking for a decade. It’s like having a Tonic expert sitting next to you, ready to build data on demand.

Built for trust and scale

Innovation shouldn't come at the cost of integrity. The Structural Agent operates within your existing Tonic permissions, ensuring your data boundaries remain exactly as you’ve defined them.

Crucially, the Agent is schema-aware. It understands referential integrity, meaning it won’t mask a foreign key without first securing its primary key, preventing broken applications and ensuring your database stays consistent. Every action taken by the Agent is logged, providing the auditability required for highly regulated environments.

This is only the beginning

The launch of the Structural Agent marks a fundamental shift in how we approach the Tonic platform. We’re already hard at work on the next phase of agentic features, including:

• Proactive subsetting: The Agent will soon recommend the optimal ways to reduce your dataset size while preserving the relationships that matter for your testing.
• Automated schema handling: When your source schema changes, the Agent will detect the drift, suggest new generators, and alert your team to ensure no sensitive data slips through.
• External pipeline integration: We are exploring ways to bring Agent capabilities directly into your CI/CD pipelines and scripts, allowing you to interact with Tonic where you already work.

Ready to accelerate your development?

The Structural Agent is available now.

• Tonic Cloud users: You’re already connected. The Agent is live and powered by Claude.
• Self-hosted customers: You can connect your own LLM (we recommend Anthropic’s Claude for the best experience) to start using these features today.

Sign up for a free account or request a demo to see the future of Test Data Management in action.