Skip to content
Nappai Documentation

Storage Search

Storage Search looks for files in your NappAI storage and returns the ones that match every filter you set: a piece of the name or a pattern, the file type, a range of modification dates and a minimum or maximum size.

It acts on the storage of the person who runs the flow, the same one you see on the Storage page, so it needs no credential. It only reads: it never changes anything.

When to use it

  • An AI agent has to find a client’s contract or invoice without knowing which folder it is in.
  • You want to process every morning the PDFs saved the day before.
  • You need to find the largest files to clean up.
  • You want every spreadsheet in a folder and its subfolders.

How it works

  • Where it searches. By default, in the Files section (the files folder) and all its subfolders. A path that names no section, such as reports, is inside that section; you can also search other sections, for example pycode_data/sales. See Files.
  • The name. Type a piece of the name (invoice finds Invoice-0042.pdf) or a pattern with * and ? (invoice-2026-*.pdf). Upper and lower case do not matter.
  • Filters add up. A file appears only if it matches every filter you fill in; the ones you leave empty do not filter.

Configuration

FieldWhat it does
Selection ModeHow you point to the folder: By Selector (you pick it in your storage) or By Path (you type its path). Defaults to By Selector.
Search In / Search In PathThe folder to search in, for example reports. Empty means the Files section. With By Selector you see Search In; with By Path, Search In Path.
NameA piece of the file name or a pattern such as invoice-*.pdf. Empty matches any name.
File TypeWhich files to find: all, document, pdf, spreadsheet, presentation, image, audio, video, archive or code. Defaults to all.
Modified AfterOnly files changed on or after that day, as YYYY-MM-DD, for example 2026-10-01.
Modified BeforeOnly files changed on or before that day, for example 2026-10-31. The whole day is included.
Include SubfoldersAlso searches inside subfolders. On by default.
Max ResultsThe most files to return, between 1 and 1000. Defaults to 100.

Advanced options

FieldWhat it does
Minimum Size (MB)Only files at least this big, in MB. 0 means no minimum. Defaults to 0.
Maximum Size (MB)Only files at most this big, in MB. 0 means no maximum. Defaults to 0.

The dates can also carry a time, for example 2026-10-01T14:30; without a time zone, the time is taken as UTC.

Output

The Data output returns one item per file found, with its name (name), its path (path), its size in bytes (size), its extension, its file type and its modification date (modified). If nothing is found, it returns an empty list.

Connect the result to Storage Files with Batch Mode to read, move or share each file found, or to an agent so it summarises them.

The component can also be connected to an Agent as a tool, so the agent looks for files when it needs to. See Tool Mode. In that case Search In Path is always used, and the agent writes it along with the filters.

Example

A company wants to send its accountant, every Monday, the previous week’s PDF invoices.

  1. Schedule the flow to run on Mondays. See Scheduler.
  2. Add Storage Search with Search In Path set to invoices, File Type set to pdf, and the previous Monday and Sunday in Modified After and Modified Before.
  3. Connect its output to Storage Files with Get File, Batch Mode on and File Path Mapped to path.
  4. Connect the files to the attachments of Gmail Message.

Tips

  • Start from the most specific folder you can: the search is faster and the results more precise.
  • If you expect many results, adjust Max Results: the search stops when it reaches the limit.
  • Turn off Include Subfolders when you only care about the files in that folder.

Security

  • Only your storage. The component searches the storage of whoever runs the flow and cannot leave it.
  • Read-only. Searching never changes, moves or deletes anything, so it is safe to give it to an agent as a tool.