What’s new with Flatfile

With reference field filtering, users can now configure dropdown fields that show only relevant options based on the value of another field in the same record.

Learn how to configure your filters in this step-by-step guide.

Flatfile's new Prepare agent helps you restructure incoming data files so they're easily mapped to your destination schema.

Import any CSV or Excel file and prompt the agent to:

• Add, remove, split, combine, or format columns

• Populate columns with calculations of other fields

• Create conditional fields

• Unpivot data (from wide → long)

Learn more about Prepare in the user guide here.

Flatfile’s developer docs have been completely restructured to make it easier to set up and configure Flatfile. The new experience offers a clear, more progressive path for learning, improved navigation, and built-in support for AI and LLM integration.

Ask the on-page AI assistant any product question, open any page in ChatGPT or Claude as context, or copy the page in Markdown to paste in your tool of choice.

Explore the new docs here.

Smart Extract is a new agent within Flatfile designed to help you extract structured, usable data from files that don’t follow a standard tabular format.

Smart Extract works with more complex files that contain: 

• Global or local metadata

• Multiple header rows

• Multiple tables within a sheet

• Summary or total rows

• Shifted data tables

• Hierarchical data

You’ll be prompted to use Smart Extract when Flatfile detects these characteristics during the initial data upload and extraction process. 

Learn more about Smart Extract and how to get started in this article.

Transforms that are applied to your data—whether from the Transform agent's recommendations or by user prompting—can now be saved to reuse on future imports. Any transforms you choose to save will be stored in the Saved Transforms library and can be run individually or all at once to quickly clean and prepare incoming data.

Learn more about Saved Transforms and how to get started in this article.

You no longer have to visit a separate interface to prompt AI to transform your data. With Inline Transform, you can select the three dots next to any field and type your prompts right there. You'll see a preview of any changes to your data and be able to edit or approve.

Learn more about Inline Transform and how to get started in this article.

With the release of Analyze and Suggestions, Flatfile now recommends actions to resolve validation errors and formatting issues in your workbooks.

When you click Analyze, the Transform agent will read your data, consider your schema, validation rules, and prior decisions, and determine where it has enough context to recommend actions.

As the analysis runs, you'll see sparkle icons appear next to column headers where the agent is proposing actions. Clicking the icon will open the Suggestions for that field and allow you to edit, preview, and apply the transform.

Learn more about these features and how to get started in this article.

Use Flatfile's newly released MCP server with tools like Claude Code, Claude Desktop, and Cursor to interact directly with your Flatfile data. Ask questions, add sheets, create sample data, and more—all with whatever tool you’re most comfortable with.

Learn more about the server and how to get started in this guide.

We've improved our filtering experience within sheets to allow users more control over the data you see. You can now apply both basic and complex filters to view errors, missing values, specific records or values, and more.