--- title: "ActiveMerge Integration" description: "ActiveMerge fills document templates with your data and returns a finished PDF, Word, or EPUB file. Agents generate contracts and letters from records, merge several files into one, render images from templates, and send the result straight out for signature." url: https://flowrunner.ai/integrations/activemerge date_modified: 2026-09-04T14:07:24-07:00 --- # ActiveMerge [Documents](https://flowrunner.ai/integrations/category/documents-forms) ActiveMerge fills document templates with your data and returns a finished PDF, Word, or EPUB file. Agents generate contracts and letters from records, merge several files into one, render images from templates, and send the result straight out for signature. 23 actions · API key · available [ActiveMerge website](https://activemerge.com/) · [Platform Documentation](https://app.activemerge.com/docs) · Capability data verified 2026-08-13 1. A candidate moves to Offer in the applicant tracking system 2. Agent reads the compensation, start date, manager, and legal entity from the record 3. Get Workflow Placeholders confirms the offer pack has a value for every placeholder it expects 4. Generate Workflow Documents builds the offer letter, the NDA, and the benefits summary with no signers attached 5. Merge Documents joins the pack into a single file so it can be read the way the candidate will read it 6. The merged proof posts to the People Ops channel with the values the agent used 7. A person reads the pack and releases Send For Signature, which is the call that reaches the candidate ## What This Integration Enables ActiveMerge is built for the moment when one set of facts has to become several documents at once. A new hire needs an offer letter, an NDA, and a benefits acknowledgment. A new customer needs an order form, a data processing addendum, and a welcome pack. ActiveMerge models that as a workflow rather than a loop, so FlowRunner agents supply the data once and get back every document the workflow defines, then merge them into a single file and send that file out for signature without leaving the connector. The generation surface is honest about being asynchronous, which suits agents better than it suits people. Every generate, merge, and image call hands back a job, and the agent polls it. That gives a flow somewhere to stand between "the document exists" and "the document has been sent," and FlowRunner uses that gap deliberately. Agents read placeholder definitions with Get Template Placeholders and Get Workflow Placeholders, render, merge, and save. The step that puts a document in front of a real person is where [human-in-the-loop](https://flowrunner.ai/concepts/human-in-the-loop) belongs, and this connector makes that step separable. ### Without FlowRunner **Assembly by copy and paste**: Each offer pack is rebuilt from last week's version, and last week's version had a different bonus clause **Errors found by the recipient**: A wrong start date is discovered when the candidate replies asking about it **Files that quietly vanish**: A generated document sits in a downloads folder until somebody needs it and cannot find it ### With FlowRunner **Assembly from the record**: One data set produces every document the workflow defines, keyed on placeholder names the agent reads first **Errors found before the send**: The pack is generated, merged, and reviewed as one artifact while it is still recallable **Files filed on purpose**: Save Generated Document keeps what matters, in the folder the originating record points at ## Use Case Scenarios ### The onboarding pack that assembles itself A candidate reaches the offer stage in [Workable](https://flowrunner.ai/integrations/workable) or [Greenhouse](https://flowrunner.ai/integrations/greenhouse). The agent reads the compensation, the start date, and the reporting line, calls Get Workflow Placeholders to confirm the offer workflow has a value for every field it expects, and runs Generate Workflow Documents. One data set produces the offer letter, the NDA, and the benefits summary. Merge Documents joins them into one file, Save Generated Document files the result, and the pack goes to People Ops for review before anything is sent. What used to be three documents assembled by three people from three templates becomes one call and one review. ### Certificates and badges from a spreadsheet Training completions land in a sheet in [Google Sheets](https://flowrunner.ai/integrations/google-sheets). On a schedule, the agent calls Get Credits to confirm the run will finish, points Generate Images at the sheet with Field Mapping saying which column feeds which placeholder, and polls Get Image Job until the run completes. Download Image Job pulls everything into file storage, and the files land in [Google Drive](https://flowrunner.ai/integrations/google-drive-api) in a folder per cohort. The flow never loops over rows, because the platform batches natively, and the agent's job is to confirm the mapping and watch the balance rather than to iterate. ### Chasing a signature without chasing everybody A contract has been out for signature for a week. Rather than a blanket reminder, the agent walks List Signature Requests for anything still open, reads Get Signature Request for each one to see which signer has stalled, and posts the specific names to the account owner in [Slack](https://flowrunner.ai/integrations/slack). If the deal has been renegotiated, the owner calls for Cancel Signature Request and the agent regenerates from the new figures. When the last signature lands, Download Signed Document files the executed copy against the record in [HubSpot](https://flowrunner.ai/integrations/hubspot). ## Human-in-Loop Highlight ActiveMerge hides its irreversible step inside a parameter. Generate Document with the Signers field empty produces a file and nothing else. The same call with Signers filled in produces the file and emails it to those people for signature, and no part of the API distinguishes the two beyond the presence of that value. A flow that builds signer data from a record is one bad lookup away from sending a draft contract to a real person, and no draft state exists to catch it. So FlowRunner agents on this connector never populate Signers. They generate clean, merge the pack, save it, and post the merged proof with the values it used: "Offer pack for Dana Reyes is ready. Start date March 3, base 118,000, signing bonus clause included. Send to dana@example.com for signature?" A person answers, and only then does the agent call Send For Signature, which takes literal addresses and is the only way a document leaves the account. The one-call convenience is real. It is also exactly the call worth not making. Agent processes routinely Detects exception requiring judgment Clear match Continues automatically Ambiguous Routes to human via email Human decides Agent resumes with decision ## Agent Capabilities 23 actions ### Generation 5 - **Generate Document** Fills a template with your data and produces a PDF, Word, or PowerPoint file, returning a job to follow. Leave the signer field empty so this stays a generation call and nothing reaches a recipient. - **Generate Workflow Documents** Runs a workflow, which produces every document that workflow defines from one set of data. The action behind onboarding packs and customer welcome sets. - **Get Generation Job** Reads a generation job. Generation is asynchronous, so this is how a flow knows the file is actually ready. - **List Generation History** Lists past generation jobs with filters for status, template, and date. Used for reconciliation and for finding the run that produced a given document. - **Merge Documents** Joins several documents into one, in the order listed. Used to turn a pack into a single file a person can read end to end before it is sent. ### Templates and Workflows 4 - **List Templates** Lists the document templates on the account, so flows adapt to the template library rather than hard-coding identifiers. - **Get Template Placeholders** Lists the placeholder names a template expects, which is what the data has to be keyed by. The step that turns a guess into a mapping. - **List Workflows** Lists the workflows on the account. A workflow produces several documents from one set of data. - **Get Workflow Placeholders** Lists the placeholder names a workflow expects across all of its templates, so an agent can check the whole pack for gaps before rendering any of it. ### Files 3 - **List Files** Lists the files and folders in the account's storage. - **Save Generated Document** Keeps a generated document permanently. Without this call the generated file expires, so anything that has to survive gets saved deliberately. - **Download File** Downloads a stored file into flow file storage, where later steps can attach it or forward it. ### Signatures 5 - **Send For Signature** Sends an existing document out for signature. Every signer is emailed as soon as this succeeds, which is why this page's human gate sits directly in front of it. - **List Signature Requests** Lists signature requests, optionally narrowed to one state. The sweep behind targeted follow-up. - **Get Signature Request** Reads one signature request with each signer and how far they have got, so a reminder can name the person who is actually holding things up. - **Cancel Signature Request** Stops a signature request. Nobody can sign it afterward, so this is used when terms change rather than as routine cleanup. - **Download Signed Document** Downloads the signed document into flow file storage, ready to file against the originating record. ### Images 5 - **Generate Images** Produces images from an image template, one per row of data. Used for certificates, badges, and any per-record visual asset. - **Get Image Job** Reads an image generation job and how far along it is. - **Download Image Job** Downloads everything an image job produced and puts it in file storage. - **List Image Templates** Lists the image templates on the account. - **Get Image Template** Reads an image template with its placeholders, which is what the field mapping is keyed by. ### Account 1 - **Get Credits** Reads how much credit is left. Worth calling before a large run, and useful as a connection check since it needs no identifiers. ## Frequently Asked Questions ### What can FlowRunner do with ActiveMerge? FlowRunner agents can run Generate Document, Generate Workflow Documents, and Get Generation Job in ActiveMerge, plus 20 more actions. ### Does connecting ActiveMerge to FlowRunner require OAuth? No. ActiveMerge connects to FlowRunner with an API key, no OAuth flow required. ### Can ActiveMerge trigger a FlowRunner workflow automatically? ActiveMerge doesn't currently expose triggers in FlowRunner. It connects as an action step inside workflows started by another trigger. **Work at ActiveMerge?** This integration exposes ActiveMerge 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/activemerge. Site index: https://flowrunner.ai/llms.txt