/api/drive/*
Google Drive document extraction should not begin with a download folder full of renamed copies. Connect the Google account once, choose the files in Google Picker and review the extracted rows beside each source. When the data is ready, write a new XLSX, CSV or JSON file back to Drive. The original stays where it was.
Google Drive account and file access
- Manual Drive workflows create two extra transfers for every document: download before extraction and upload after export.
- Local copies lose the shared folder context, which makes the right source and the latest result harder to trace.
- A finished workbook belongs back in the team Drive, not in one person's Downloads folder or an email attachment.
- The connection should request a clear Google scope and remain easy to revoke.
How the Drive connection works
Dynamite Docs- 01Connect Google once
- 02Browse and import
- 03Extract and review
- 04Export back to Drive
Connect Google once
One OAuth connection can grant the exact scope used by each feature: drive.file for Google Sheets push, Drive exports, and Drive imports via Google Picker. The connection is per account, encrypted, and revocable from the workspace.
Browse and import
The standard Google Picker lets you securely browse, search, and select one or multiple files directly from your Drive. Selected files stream to the client for the normal upload pipeline, so R2 persistence and browser-only storage work exactly as they do for a drag-and-drop.
Extract and review
The imported document uses the normal extraction pipeline with confidence-scored rows, saved correction patterns, hosted page caps, and the same file limits as other uploads.
Export back to Drive
Push a finished extraction to your Drive as an XLSX workbook, CSV file, or JSON file. The endpoint uses the same format builders as direct downloads, then creates a new Drive file.
What the Drive connection moves
Dynamite Docs- Drive import
- PDFs, images, Sheets, Docs, and more via Google Picker
- Drive export
- finished extractions as XLSX, CSV, or JSON files
- Google Picker
- secure multi-file selection with standard Drive UI
One shared OAuth connection powers both directions, with tokens encrypted using the same BYOK encryption that protects your API keys:
A realistic example
A typical audit-adjacent loop with Drive:
| Step | Where it happens | Outcome |
|---|---|---|
| Client drops a trial balance PDF into a shared Drive folder | Google Drive | Document waits, no download needed |
| You import it from the Drive panel in Dynamite Docs | Dynamite Docs workspace | File streams into the normal pipeline, parsed as rows |
| Review low-confidence balances and export | Dynamite Docs Data Studio | Clean .xlsx written back to your Drive |
What the Drive connection does and does not
Does
- Opens Google Picker for user-controlled browsing and multi-file selection.
- Streams selected files to the client so the standard upload pipeline decides storage and extraction.
- Exports finished extractions back to your Drive as new XLSX, CSV, or JSON files.
Does not
- Does not scan or index your entire Drive. Only the files you select are fetched.
- Does not overwrite files in your Drive; exports are written as new workbooks.
- Does not keep your OAuth token unencrypted. Tokens are AES-GCM encrypted like every other secret.
Why Drive is more than a file picker
Import reuses the upload pipeline, so Drive-sourced files get identical storage, extraction, and review behavior, no second-class path to maintain.
Export closes the loop your team already has. The results go back to the Drive where the source documents and the shared spreadsheets already live.
One scoped, revocable OAuth connection with encrypted tokens keeps the integration useful without a broad, unexaminable scope grant.
Setup and delivery questions
No. Both use the same encrypted Google connection. If your connection predates a scope needed by a visible feature, you will be prompted to reconnect for that missing scope.
The panel lists importable files by default, PDFs, images, Sheets, Docs, and other formats the extraction pipeline understands. Non-importable types are filtered out of the list.
Exactly where an upload would: 500 MB of R2-backed storage on Starter, 2 GB on Pro, 10 GB on Scale, and browser-only storage on Free. The import feeds the existing pipeline.
Disconnect from the workspace. One action revokes the shared connection for both Drive and Sheets.
Extraction guides
Related integrations
Push reviewed AI-extracted tables to Google Sheets
Export reviewed AI tables from Dynamite Docs to Google Sheets. Connect once, push rows to a fresh spreadsheet tab, and keep document fields separate.
POST /api/sheets/export
Read the integrationExport reviewed AI extraction results to CSV, JSON, and Excel
Export reviewed AI data from Dynamite Docs to CSV, JSON, and Excel. Choose table downloads or Word and PDF exports from the reviewed document state.
Export API (?format=csv|json|xlsx)
Read the integrationExtract document data using LLMs through a REST API
Extract document data using LLMs through the Dynamite Docs REST API. Submit files, retrieve structured rows, and manage scoped keys within plan limits.
/api/v1/extractions, /api/v1/infer, /api/tokens
Read the integrationDocument workflows it feeds
Extract invoice line items for AI accounting
Extract invoice line items for AI accounting in Dynamite Docs. Check descriptions, quantities, prices and tax against the source before entry.
Extract bank statement data for AI bookkeeping
Extract bank statement rows for AI bookkeeping in Dynamite Docs. Check dates, debits, credits and balances before exporting reconciliation data.
From the blog
Pull documents from Drive and send results back.
Connect Google once from the workspace, import a PDF straight from your Drive, and export the finished extraction back as an .xlsx, no downloads in between.