> ## Documentation Index > Fetch the complete documentation index at: https://docs.adopt.ai/llms.txt > Use this file to discover all available pages before exploring further. # File Explorer > A per-client document library of raw source files that both agents and people can search during a run. Not all data is structured. A close depends on bank statements and invoices; a tax engagement depends on source PDFs and spreadsheets. File Explorer is where those raw documents live — the counterpart to the [Data Store](/accounting/data-store), which holds structured rows. Where the Data Store has tables, File Explorer has files. ## What Is File Explorer? File Explorer stores source documents — `.pdf`, `.xlsx`, `.csv`, `.doc` — organized into folders called **Document Stores**. These are the files an agent can search and reason over during a run, and the same files appear to the person in the conversation, so everyone sees exactly what the agent is working from. File Explorer is a scoped, searchable document library — a shared drive that both the agent and the human can open, per client. ## How to Access File Explorer Click **File Explorer** in the left navigation, under the **Adopt Primitives** group. Acct File Explorer ## Document Stores The left panel lists your **Document Stores** — the folders that group a client's source material. Use the **+** button to create a new store. Everything you see is scoped to the workstream selected in the **Workstream** picker at the top right, so switching clients switches the documents in view. ## Inside a Document Store Selecting a store opens its file list. Each file is shown with the details you need to manage it: | Column | What it shows | | ------------ | --------------------------------------------------------------------------------------------------------- | | **Filename** | The document's name, with a type icon. | | **Type** | The file format — CSV, PDF, and so on. | | **Size** | The file size. | | **Status** | Where the file is in processing — for example, **Uploaded** (ready) or **Queued** (still being ingested). | | **Uploaded** | When the file was added. | A **New folder** button lets you organize documents into sub-folders within a store. ## How Files Get There Documents reach a store in two ways. You can upload them directly with the **Upload** button. Or an intake agent or pipeline can populate the store automatically — for example, pulling a client's R\&D source files from a shared drive into the workstream's document library, which is exactly what the R\&D Tax Credit Intake agent does. Because a document's **Status** reflects ingestion progress, a file marked **Queued** isn't yet searchable by an agent. Wait for it to reach **Uploaded** before running an agent that depends on it. ## Why File Explorer Matters Accounting and tax work is document-heavy, and the source documents are often where the real answers live. File Explorer keeps each client's raw material tidy, searchable, and — critically — visible to both the agent and the reviewer. When an agent cites a figure from a bank statement, the reviewer can open that exact statement in the same view. No guessing about what the agent read. ## Next Steps 1. [See how structured data is handled in the Data Store](/accounting/data-store) 2. [Understand how Workstreams scope each client's documents](/accounting/workstreams) 3. [Watch files appear in a live run](/accounting/end-user-app)