Files Tab
The Files tab on every request. Document-group folders, the four file-match states, OCR/text-readability status, full-text search across uploads, and one-click download of every file in the request.
The Files tab on a request is where every uploaded document lives, organized by document group. Click in and you see one folder per group (Time Cards, Payroll Records, Employee Records) with a count of the files inside each. From there you can drill into a group to see each file's match state, check OCR/text-readability, run a full-text search across every upload on the request, and download the whole set when you need a local copy.
This page walks the Files tab in the order you actually use it: pick a folder, read the file states, confirm the documents are text-readable, search, and download.
What you'll learn
- How files are organized into document-group folders on the Files tab.
- The four file-match states (Unmatched, Matched (unlocked), Matched (locked), and Excluded) and what each one means.
- How the Text button surfaces OCR / text-readability status.
- How Search scans every text-readable document on the request.
- How Download pulls every file at once.
Document-group folders
The first thing you see in the Files tab is a row of folders, one per document group defined on the Workflow tab. Each folder shows:
| Element | What it tells you |
|---|---|
| Folder name | The document group (e.g., Time Cards, Payroll Records, Employee Records). |
| File count | The number of files uploaded into that group. |
Click a folder to open the file viewer for that group. Inside the viewer, each file carries a state badge that tells you where it sits in the workflow.
File-match states
Every file inside a document-group folder lands in one of four states. The state is what tells you whether the file is wired into a selection, whether a human has signed off on the match, or whether it's been pulled out of scope.
Unmatched
The file wasn't matched to any selection during the workflow run. Either the Match Agent couldn't find a confident pairing, or the file was uploaded after the last run.
Matched (unlocked)
The agents reviewed the file and believe they found the proper match based on the selection criteria, but no human has reviewed and locked it yet. Open the match, confirm it, and lock it.
Matched (locked)
A human reviewed the file, confirmed the match, and locked it to the selection. Locked matches are sticky; subsequent reruns won't shuffle them.
Excluded
The file was uploaded by the client but isn't relevant to the workflow. Auditors can move a file into Excluded to keep it on record without letting it drive any of the workpaper logic.
The Excluded state is the right answer for files the client uploaded "just in case." They stay attached to the request for the audit trail, but they don't participate in matching or extraction.
The Text button: OCR / text-readability
Each file goes through an OCR / text-readability pass before the agents will read it. The Text button on the file viewer surfaces where each file is in that pipeline:
Uploaded
The file landed but text extraction hasn't started yet.
Processing
Agentive is running the file through OCR. Scanned PDFs and images take longer than digital PDFs.
Text readable
The file's text has been extracted. The agents can read it, and the workflow can run against it.
Don't run the workflow against files still in Uploaded or Processing. The agents have no text to read yet and the run will skip those documents. Wait until the Text button shows Text readable across the document group, then click Run on the Workflow tab.
Search across documents
The magnifying-glass icon opens a full-text search across every text-readable document on the request: every document group, not just the one you're viewing.
Click the magnifying glass
Open the search box from the Files tab toolbar.
Type a keyword and press Enter
Type the term you're looking for: a check number, an employee name, a contribution amount, a date. Press Enter.
Read the results
Agentive scans every text-readable file across every document group on the request, finds the hits, and surfaces the matches (including the document, the group, and the location of the hit) so you can jump straight to the page that mentions the term.
Search only reads files that are Text readable. If you're missing hits you'd expect, double-check the Text status on the documents you thought should match. They may still be in Processing or have failed OCR.
Download every file
The Download button on the Files tab pulls every file uploaded to the request as a single download. Keep a local copy for the audit file, hand off to a reviewer who isn't in Agentive, or archive the evidence offline.
The download preserves the document-group folder structure, so the local copy mirrors what you see in the Files tab.
Common situations
Continue your journey
← The Memo tab
Give the AI its standing context for the request.
The Workflow tab →
Build the workpaper: document groups, AI Prompt Columns, formulas, and testing attributes.
How is this guide?
Memo Tab
The structured context (Purpose, Source, Scope, Accounting Policies, Procedures, Testing Attributes, and Conclusion) that tells the AI what the workpaper is doing, what rules to apply, and what it's already done. Captured with the workflow and exported with the audit file.
The Workflow Tab
Build a workpaper end to end. Pick a selections vs. no-selections shape, add selections, wire up document groups (new or shared from the engagement), configure the Source and Workpaper sections, set match inputs, add AI Prompts, Formulas, and Testing Attributes, and run. The complete reference for the Workflow tab.