Finding your way around the Atlas File System

The File System is home base in Atlas — where data modules, record schemas, topics, records and submissions live. A tour of the breadcrumb, views, search and the details panel.

Almost everything you build in Atlas lives in one workspace: the File System. Once you’re comfortable with it, the rest of the platform is easy to navigate, because every kind of item — modules, record schemas, topics, records, submissions — is found the same way.

What lives in the File System

  • Data Modules — every top-level classification in your organization.
  • Inside each module: its record schemas, sample schemas, topics and resources.
  • Inside each topic: its data records, samples and any nested resources.
  • On the Advanced tier, Data Collection submissions are part of the same file system — a submission is just another thing you can find in it.

The breadcrumb: where am I?

The breadcrumb along the top always shows where you are — for example, Data Modules › Properties › 42 Juniper Lane. Use it to move up and down your hierarchy: select any level to jump back to it.

Each level lists the items it contains, grouped by kind.

Views, search and sorting

  • Grid and list views — switch with the toggle at the top right. Grid is good for browsing; list is good for scanning names and dates.
  • Search — narrows the current kind of item, so you can find one topic among hundreds.
  • Sort — by name or date.

The details panel

Select any item to open its details panel, which shows the actions available on it. For a data module, that’s open it, edit it, change its security policy, or archive it. Other items offer the actions that make sense for them, such as adding a child or starting a pipeline.

A useful habit: select first, then act. The details panel is where Atlas puts everything you can do to an item, so you rarely need to hunt through menus.

Labels, not names

If a classification has a label, the File System shows the label in place of the name. A module named Service Properties with the label Properties appears as Properties. That keeps the workspace readable while names stay precise. See Naming classifications well.

The New menu follows your model

Inside a module or a topic, New offers what that location can hold. Inside a topic, the record schemas its classification defines appear as new data records — so choosing what to create and choosing what it’s made from are the same decision.

A typical navigation path

Finding one property’s latest record follows a predictable path:

  1. Open Data Modules in the navigation.
  2. Select Properties — the label shown for the Service Properties classification.
  3. Search within topics for 42 Juniper.
  4. Select the topic and open it.
  5. Inside the topic, its data records, samples and resources are listed.
  6. Open the record you need.

The breadcrumb along the top shows the whole path, so you can jump back to any level.

Finding things quickly

  • Search within a kind. Search narrows the current kind of item, so search for topics inside a module, or record schemas inside a classification.
  • Sort deliberately. Sorting by date surfaces recent work; sorting by name is better for browsing a long list.
  • Use list view for scanning names and dates, and grid view for browsing.
  • Use labels so modules read the way your team talks.
  • Use the details panel rather than hunting for menus — it shows the actions available on the selected item.

Where each kind of thing lives

Looking for… Where it is
A classification Data Modules
A record schema Inside its classification
A sample schema Inside its classification
A topic Inside its classification
A data record Inside its topic
A resource’s topics Inside the parent topic
A public form submission Data Collection (Advanced)
A topic with no owner Unassigned Topics

Frequently asked questions

Why does a module show a different name than I gave it? If a classification has a label, the File System shows the label instead of the name.

Why doesn’t the New menu show what I expect? The menu is built from what the classification defines at that location — inside a topic, that’s its record schemas as new data records.

Can I move an item? The structure follows your model: a record belongs to its topic, and a topic to its classification. Changing where something belongs generally means creating it in the right place and archiving the original.

How do I take something out of use? Archive it from its details panel, which preserves history rather than destroying it.

Habits that make the File System fast

  • Start from the breadcrumb. Jumping to a parent level is usually quicker than searching from scratch.
  • Search within the kind you want. Searching topics inside a module beats scanning everything.
  • Give labels to long names so lists stay scannable.
  • Keep descriptions current — they’re the fastest orientation for anyone new.
  • Archive rather than accumulate. A module full of retired topics makes finding current ones slower for everyone.

These are small habits, but with hundreds of topics they’re the difference between finding something in seconds and hunting for minutes.

Keeping the File System tidy

  • Use labels so modules read the way your team talks.
  • Archive, don’t delete, items you no longer use, so history stays intact.
  • Keep Unassigned Topics empty by assigning every topic an owning group. See Why every topic needs an owner.
  • Model before you build, so the hierarchy reflects the business. See Modeling your business in Atlas.

Ready to build your business graph?

Start with reusable schemas today. Scale into enterprise APIs when you need them.