--- title: "DocuGenerate Integration" description: "Connect AI agents to DocuGenerate, which merges structured data into Word and OpenDocument templates to produce finished documents. Agents manage templates, generate Word or PDF output from JSON or spreadsheet data, and download the results so paperwork gets produced and filed automatically." url: https://flowrunner.ai/integrations/docugenerate date_modified: 2026-08-17T14:28:03-07:00 --- # DocuGenerate [Documents](https://flowrunner.ai/integrations/category/documents-forms) Connect AI agents to DocuGenerate, which merges structured data into Word and OpenDocument templates to produce finished documents. Agents manage templates, generate Word or PDF output from JSON or spreadsheet data, and download the results so paperwork gets produced and filed automatically. [Verified](https://flowrunner.ai/integrations/verified "What does verified mean?") · 12 actions · API key · available [DocuGenerate website](https://docugenerate.com/) · [Platform Documentation](https://api.docugenerate.com/) · Capability data verified 2026-08-28 1. A deal moves to Closed Won in the CRM 2. Agent pulls the customer, term, pricing, and signatory fields from the deal record 3. Agent calls Get Template and matches every key it is about to send against the template's valid tags 4. Agent calls Generate Document to merge the data into the contract template as PDF 5. Agent calls Download Document to copy the finished file into FlowRunner file storage 6. Agent posts the document link into the deal channel with the values it merged 7. Contract owner confirms the merged terms, and the agent releases the document to the e-signature step ## What This Integration Enables DocuGenerate is a merge engine, not a document editor, and the distinction shapes how an agent should use it. Templates are Word or OpenDocument files carrying tags; the API parses a template on upload and tells you exactly which tags it found and which are broken. That tag list is the contract between the template and the flow. An agent that reads it before generating knows whether its data will land in the right places, and a template with even one invalid tag cannot produce a document at all until the file is corrected. FlowRunner agents run both generation paths: one data object per document, or a spreadsheet where each row becomes a document. They can produce Word, PDF, PDF/A for archival requirements, HTML, plain text, or a PNG of the first page for a thumbnail. Download Document copies the result out of DocuGenerate cloud storage and into FlowRunner file storage, a file the rest of the workflow can attach, email, or keep. The [connectors](https://flowrunner.ai/concepts/connectors) layer handles the mechanics. Where the document carries a commitment, the platform default is that a person reads it before it leaves. ### Without FlowRunner **Templates edited per deal**: Someone opens the last contract, saves a copy, and edits the parts that changed **Silent merge failures**: A renamed tag produces a document with a blank where a price should be, and nobody notices until the customer does **Documents stranded in a vendor account**: Finished files live behind links that stop resolving when the source record is cleaned up ### With FlowRunner **Data driven output**: The document is produced from the record that authorized it, with no retyping in between **Tags checked before the merge**: The agent reads the template's valid tags and stops when its data does not match them **Archived where the workflow lives**: Every generated file is copied into FlowRunner storage so downstream steps have a durable copy ## Use Case Scenarios ### Closed deal to signature-ready contract A deal closes in [HubSpot](https://flowrunner.ai/integrations/hubspot). The agent calls Get Template to read the contract template's valid tags, maps deal fields onto those exact tag names, and calls Generate Document with PDF output. It calls Download Document so the file is held in FlowRunner storage, then hands the stored file to [DocuSign](https://flowrunner.ai/integrations/docusign) or [SignNow](https://flowrunner.ai/integrations/signnow) for signature. The signed document goes back onto the deal record. Nobody opens Word. ### Batch letters from a spreadsheet An operations team needs personalized notices for several hundred accounts. The agent builds a spreadsheet from a [PostgreSQL](https://flowrunner.ai/integrations/postgresql) query, uploads it, and calls Generate Document From Spreadsheet against the notice template with Combine Into Single File turned off, so the result is a zip archive holding one file per row. The agent downloads the archive into FlowRunner storage and posts a sample of three individual documents for review before the print or mail step runs. ### Archival-grade records A finance or compliance workflow produces records that have to survive an audit years later. The agent generates PDF/A output, downloads it into FlowRunner storage, and files it into [SharePoint](https://flowrunner.ai/integrations/sharepoint) or [Amazon S3 / Cloud Storage](https://flowrunner.ai/integrations/s3) alongside the source data used to build it. Because Download Document runs at generation time, the archived copy does not depend on the document still existing in DocuGenerate months later. ## Human-in-Loop Highlight The genuinely destructive operation in DocuGenerate is not generating a bad document, it is Delete Template. Documents already generated survive the deletion and their download links keep resolving, but the template itself, the layout and merge-tag work that produces every future document, is gone for good. When an agent runs a template cleanup, it does not delete. It gathers the evidence and asks. It calls List Documents for each candidate template to see when it last actually produced anything, and posts to the document owner: "Template MSA v3 is unused since March. Its 412 generated documents and their links stay available either way. Delete the template, or keep it?" The sweep continues on that answer. Everything else in the cleanup runs unattended. 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 12 actions ### Templates 5 - **Create Template** Uploads a document file and turns it into a reusable template, returning every tag found in it split into valid and invalid. Used to onboard a new document type, and the returned tag list is what later steps map their data onto. - **List Templates** Returns every template in the account, newest first, with tags, page count, delimiters, folder path, and download links. Used to resolve a template by name or folder before generating. - **Get Template** Retrieves a single template with its extracted tags. Used as the pre-flight check: read the valid tags, confirm the data keys match, and stop before generating when they do not. - **Update Template** Replaces the template file, changes the delimiters, or updates the name or folder while keeping the template ID and its generated documents. Uploading a new file re-parses the tags, so the returned tag list should be re-read afterward. - **Delete Template** Permanently deletes a template. Documents generated from it are kept and their links keep working, but the template is gone for good, which is why it belongs behind a human decision rather than inside an unattended sweep. ### Documents 6 - **Generate Document** Merges one or more data objects into a template and produces a finished document in Word, PDF, PDF/A, OpenDocument, plain text, HTML, or PNG. Used as the main production step once the tag mapping is confirmed. - **Generate Document From Spreadsheet** Merges the rows of an uploaded spreadsheet into a template, producing one document per row without building the data objects by hand. Used for batch runs where the source is already tabular. - **List Documents** Returns every document generated from one template, newest first, with format, record count, and download link. Used for reconciliation and, before any template cleanup, to see how recently the template was actually used. - **Get Document** Retrieves a single generated document with its download link. Used to re-read a link produced earlier in a flow or to confirm a document still exists before sending it onward. - **Rename Document** Changes the logical name used to identify a document in the dashboard and in listings. The generated filename and the existing download link are unaffected. - **Delete Document** Permanently deletes a generated document and the file behind it, and the download link stops resolving immediately. Used for cleanup after the file has been archived elsewhere. ### Files 1 - **Download Document** Fetches a generated document from DocuGenerate cloud storage into FlowRunner file storage and returns a FlowRunner URL. Later steps can attach, email, or archive it without depending on the vendor link. ## Frequently Asked Questions ### What can FlowRunner do with DocuGenerate? FlowRunner agents can run Create Template, List Templates, and Get Template in DocuGenerate, plus 9 more actions. ### Does connecting DocuGenerate to FlowRunner require OAuth? No. DocuGenerate connects to FlowRunner with an API key, no OAuth flow required. ### Can DocuGenerate trigger a FlowRunner workflow automatically? DocuGenerate doesn't currently expose triggers in FlowRunner. It connects as an action step inside workflows started by another trigger. **Work at DocuGenerate?** This integration exposes DocuGenerate 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/docugenerate. Site index: https://flowrunner.ai/llms.txt