--- title: "DocuWare Integration" description: "Archive and retrieve documents in DocuWare, Ricoh's enterprise document management platform. Agents upload files with index-field metadata, search cabinets with DocuWare Query Language, and download stored documents into FlowRunner storage for downstream steps." url: https://flowrunner.ai/integrations/docuware date_modified: 2026-08-04T10:04:16-07:00 --- # DocuWare [Storage](https://flowrunner.ai/integrations/category/storage) Archive and retrieve documents in DocuWare, Ricoh's enterprise document management platform. Agents upload files with index-field metadata, search cabinets with DocuWare Query Language, and download stored documents into FlowRunner storage for downstream steps. [Verified](https://flowrunner.ai/integrations/verified "What does verified mean?") · 6 actions · OAuth · available [DocuWare website](https://docuware.com/) · [Platform Documentation](https://developer.docuware.com/) · Capability data verified 2026-07-27 1. An invoice PDF arrives as an email attachment in the accounts payable mailbox 2. The agent extracts the vendor, invoice number, amount, and date from the document 3. Search Documents runs a DQL query on the invoice number to catch duplicates already in the cabinet 4. The agent scores its own extraction confidence against the cabinet's required index fields 5. An AP clerk confirms the index values for any document the agent marked low-confidence 6. Upload Document archives the file into the invoices cabinet with the confirmed index fields 7. The document Id and index values log to the AP tracking sheet and the team channel ## What This Integration Enables DocuWare's premise is that a document without index metadata is just a picture of information. The index fields are what make a cabinet queryable, and populating them accurately for every incoming document is precisely the repetitive, judgment-light work that agents should own. FlowRunner agents read incoming files, extract the metadata the cabinet requires, check for duplicates with DocuWare Query Language, and archive documents where the whole company can find them. On the way out, they pull stored documents into FlowRunner storage so contracts, invoices, and records can flow into signature, approval, and reporting steps. - Archive incoming PDFs, invoices, and scans with searchable index-field metadata - Query cabinets with DQL by status, date, and field values, no folder archaeology - Pull stored documents into FlowRunner storage for signature, forwarding, or attachment steps - Bridge the archive with spreadsheets, email, and chat for document-driven workflows - Navigate multi-organization systems and their cabinets programmatically The archive is the record of record, which is why FlowRunner's [human-in-the-loop](https://flowrunner.ai/concepts/human-in-the-loop) gate sits on the indexing decision, not after it. An agent that files a thousand clean invoices unattended earns its keep; an agent that guesses on the ambiguous ones spends it. The gate keeps the distinction sharp. ### Without FlowRunner **Scanning is the easy half**: Documents get digitized fast and then sit unindexed, because typing metadata is the part nobody wants **Findability depends on the filer**: A misfiled invoice is functionally lost until someone stumbles across it at audit time **Documents leave the archive by email**: Getting a stored contract to a signer means downloading, attaching, and losing the paper trail ### With FlowRunner **Every document arrives indexed**: Agents extract index-field values and archive with searchable metadata in the same motion **The cabinet answers questions**: DQL queries by status, date, or field value replace folder spelunking **Retrieval is a workflow step**: Download Document pulls file bytes into FlowRunner storage where any downstream step can use them ## Use Case Scenarios ### The AP mailbox that files itself An attachment lands via [Gmail](https://flowrunner.ai/integrations/gmail-service). The agent identifies it as an invoice, extracts vendor, invoice number, amount, and date, and runs Search Documents with a DQL condition on the invoice number to catch a duplicate submission. Clean documents get archived with Upload Document, index fields matching the cabinet's database field names exactly. The document Id logs to [Google Sheets](https://flowrunner.ai/integrations/google-sheets) and a notification posts to [Slack](https://flowrunner.ai/integrations/slack). The cabinet stays current with zero manual indexing on routine invoices, and the duplicate check runs on every single document, which is more than any busy clerk can honestly promise. ### Contracts routed to signature from the cabinet Renewal season. The agent runs Search Documents for contracts with a DQL condition on expiry date, then calls Download Document on each match, landing the file bytes in FlowRunner storage with a usable URL. Each contract routes to [Dropbox Sign](https://flowrunner.ai/integrations/hellosign) for signature, and the signed version comes back through Upload Document with index fields linking it to the original. The archive holds both ends of the story, and nobody emailed an attachment to make it happen. ### The audit request answered in minutes An auditor asks for all invoices from a vendor across a date range, the kind of request that used to consume an afternoon and still miss the misfiled ones. The agent queries the cabinet with a DQL condition combining the vendor field and DWSTOREDATETIME bounds, pages through the results, downloads each document into FlowRunner storage, and assembles the set with a manifest of document Ids and index values for the auditor's file. What used to be an afternoon of cabinet searching becomes a request the ops team fulfills before the follow-up email arrives. ## Human-in-Loop Highlight Upload Document is a write into the cabinet your company treats as the record of record, and its index fields are the document's only address. An invoice archived with a mistyped INVOICE\_NO or the wrong vendor value does not look broken; it looks archived. It simply stops being findable by the queries everyone uses, which in an audit or a payment dispute is the same as being lost. So FlowRunner agents grade their own extraction before they file. High-confidence documents archive straight through. Anything ambiguous, a handwritten total, a vendor name that fuzzy-matches two masters, an invoice number format the agent has not seen, goes to a person with the document image and the proposed index values side by side. The clerk corrects or confirms, and only then does Upload Document commit the record. Ten seconds of human review at filing time beats an hour of DQL forensics at audit time. Agent processes routinely Detects exception requiring judgment Clear match Continues automatically Ambiguous Routes to human via preferred channel Human decides Agent resumes with decision ## Agent Capabilities 6 actions ### Organizations and Cabinets 2 - **List Organizations** Lists the organizations available to the authenticated user with Id, Name, and default flag. Used to scope cabinet discovery in multi-organization systems. - **List File Cabinets** Lists the file cabinets the user can access, with Id, Name, Color, and whether the cabinet is a document tray. The starting point for every document operation. ### Documents 4 - **Search Documents** Searches a cabinet with a DQL condition on index fields, with paging via Start and Count. An empty condition lists recent documents. The query surface behind dedup checks, audit pulls, and routing decisions. - **Get Document** Retrieves a single document's full metadata by document Id: index-field values, content type, title, size, sections, and store date. - **Upload Document** Stores a file into a cabinet with its index-field values, fetching the file from a URL and posting it with the index JSON. Returns the created document's Id. The write this page's human gate exists for. - **Download Document** Downloads a document's file bytes into FlowRunner storage and returns a URL for downstream steps, with DocuWare choosing the right representation for multi-section documents. ## Frequently Asked Questions ### What can FlowRunner do with DocuWare? FlowRunner agents can run List File Cabinets, Search Documents, and Get Document in DocuWare, plus 3 more actions. ### Does connecting DocuWare to FlowRunner require OAuth? Yes. DocuWare connects to FlowRunner with OAuth 2.0, so agents authenticate without handling raw credentials. ### Can DocuWare trigger a FlowRunner workflow automatically? DocuWare doesn't currently expose triggers in FlowRunner. It connects as an action step inside workflows started by another trigger. **Work at DocuWare?** This integration exposes DocuWare to AI agents on every FlowRunner plan, including through MCP, at no cost to you. [See what FlowRunner offers integration partners](https://flowrunner.ai/integrations/partners), including how to keep this page current. --- Markdown version of https://flowrunner.ai/integrations/docuware. Site index: https://flowrunner.ai/llms.txt