Search And Filter By Content
Once a value is pulled into a metadata field, you can search and list files by it in the Files.com app and API — find every file where the country code is US, not just every file whose name happens to contain it.
Route And Filter Files By What’s Inside Them
Files.com reads field values out of structured files — XML, JSON, EDI, HL7, CSV — and turns them into custom metadata you can search, filter, automate, and route on. A file lands with a country code buried in its XML, Files.com pulls that value into a metadata field, and routes the file on it automatically — with no parser to write and no separate content engine to run.
See Inside The File, Not Just Its Name
Most file platforms only know a file by its outside: its name, its size, when it showed up. That’s why so much routing ends up leaning on naming rules that partners never follow the same way twice. File extraction lets Files.com look inside the file and act on what’s actually in it.
Take a common job: pull a country code out of an inbound XML file and send the file to a different place depending on the value. Today that means a custom Lambda to parse, extract, and re-route. With file extraction you point at the field once, and from then on the value is part of the file’s record — something you can filter, automate, and audit on.
What The Metadata Unlocks
Once a value is extracted into metadata, the rest of the platform can act on it the same way it acts on any other file attribute.
Trigger Automations On It
Extracted values can kick off an automation rule. “Move files where country is US to one folder, files where country is CN to another” is a rule you set once, not a custom script you have to maintain.
Route By What’s Inside
Send a file somewhere based on a value buried inside it, not on its filename. No more relying on naming rules that partners never follow the same way twice.
Keep It In The Audit Record
The content classification becomes part of the immutable audit record, not a one-time lookup that disappears. How a file was sorted is recorded alongside everything else that happened to it.
Extraction Is The First Half Of The Pipeline
Extraction and transformation are usually paired: pull a value out of a file, then use TransformScript to reshape the file based on it and route the result. Together they’re the engine behind the broader extract, transform, and verify story — turning a file that just arrived into data your downstream systems can use right away.
Extraction usage consumes Transformation and AI Credits, included per plan and scaling by tier. See what your plan includes on the pricing page.
Zip And Unzip Without Downloading First
Extraction reads inside a file without pulling it down. Zipping works the same way. Select a set of files and folders and bundle them into a single .zip for delivery, or select an archive and extract it in place — all on the platform, with nothing downloaded to a laptop first.
Zip and unzip run as background jobs, so a huge archive or a deeply nested folder tree processes without tying up your session. You can browse a .zip to see every filename and its uncompressed size before you extract, and pull out one file or the whole archive. It works from the web app, the CLI, and the API alike, and every zip and unzip lands in the audit trail like any other file action.
File Extraction Questions
What teams ask about which file types extraction reads, what the metadata can do, how zip and unzip work, and how it differs from AI-driven extraction of messy inputs.
File extraction lets Files.com look inside a structured file and pull specific field values out into custom metadata attached to that file. Instead of only knowing a file’s name, size, and arrival time, Files.com can read a value from inside it — like a country code in an XML payload — and make that value filterable, automatable, routable, and part of the audit record.
File extraction works on structured files: XML, JSON, EDI X12, EDIFACT, HL7 v2, and CSV. These are formats where the field locations are deterministic, so the value can be pulled reliably every time.
Not under file extraction itself, which is for structured files. Pulling fields out of messy or unstructured inputs like PDFs, scans, and emails is handled by Run-Time AI inside TransformScript, a separate capability. File extraction is the deterministic path for files where the field paths are known.
Extracted values become custom metadata fields on the file. From there the metadata is searchable and filterable in the UI and API, usable as a trigger for automation rules, available as a routing input, and recorded in the audit trail. It behaves like the file’s standard attributes — first-class, not a side note.
No. You declare that for files matching a pattern, a given field should be extracted into a named metadata field, and from then on Files.com populates it automatically the moment a file arrives. There is no per-file-type extraction pipeline to build and keep alive, and no separate content-inspection engine to bolt on.
Yes. Select files and folders and choose Save as Zip to bundle them into a single .zip archive, or select a .zip and choose Unzip to extract it — all server-side, with no download required. Zip and unzip run as background operations, so large archives and deeply nested folder structures process without blocking your work, and both are available in the web app, the CLI, and the SDKs and API.
Unzip supports archives containing up to 10,000 files. Before extracting, you can browse a .zip archive to see each file name and its uncompressed size, and you can extract either a single file or the entire archive. Zip and unzip require read permission on the source content and write permission on the destination folder.
Make File Content First-Class Metadata
Start a free trial, point Files.com at the field you care about, and watch it become searchable, filterable metadata the moment a file arrives — ready to route and automate on.
No credit card required • Free for 7 days • Live in minutes