--- title: "ComPDFKit PDF Converter Integration" description: "ComPDFKit PDF Converter turns PDFs into Word, Excel, PowerPoint, HTML, CSV, JSON, Markdown, or images and converts those formats back. Agents make scans searchable with OCR, pull PDF data into structured output, and hand editable files to the people who need them." url: https://flowrunner.ai/integrations/compdfkit-pdf-converter date_modified: 2026-09-04T14:07:24-07:00 --- # ComPDFKit PDF Converter [Document Processing](https://flowrunner.ai/integrations/category/documents-forms) ComPDFKit PDF Converter turns PDFs into Word, Excel, PowerPoint, HTML, CSV, JSON, Markdown, or images and converts those formats back. Agents make scans searchable with OCR, pull PDF data into structured output, and hand editable files to the people who need them. 9 actions · API key · available [ComPDFKit PDF Converter website](https://www.compdf.com/) · [Platform Documentation](https://www.compdf.com/guides/api-reference) · Capability data verified 2026-08-13 1. A counterparty returns a signed agreement as a scanned PDF 2. Make PDF Searchable runs recognition over the scan so the filed copy has a text layer 3. Agent compares the page count and clause headings against the version that was sent out 4. Convert PDF produces the editable copy for the next redline round 5. Agent files the searchable PDF and posts the editable copy with any pagination difference called out 6. The contracts lead decides whether this document may be reflowed before it goes back to the counterparty ## What This Integration Enables Format is usually treated as a property of a document. It is closer to a routing decision. A contract that arrives as a scan needs a text layer before anything can check it, the same contract needs to be editable before a counterparty can mark it up, and the archived version needs to be a PDF again. ComPDFKit PDF Converter does all three directions: PDF into Word, Excel, PowerPoint, HTML, CSV, text, JSON, Markdown or images; those formats back into PDF; an image straight into any of them without a PDF in between; and recognition to make a scan searchable in place. That bidirectionality is what makes it useful as a flow component rather than a utility. FlowRunner agents convert on arrival so the rest of the workflow only ever handles one format, pull numbers out as CSV or JSON instead of parsing text that was never meant to be parsed, and render pages to images when a person needs a preview rather than a file. Conversions are synchronous, so the result comes back in the response and there is usually nothing to wait for. FlowRunner's [connectors](https://flowrunner.ai/concepts/connectors) are built and verified against each vendor's official API, so the format pairs a flow can rely on are the ones ComPDF actually serves, and List Supported Conversions is the vendor's own answer when one is refused. ### Without FlowRunner **Format is somebody's afternoon**: A person opens a converter, picks settings, saves the output, and files it in the right place **Scans that nothing can read**: A signed PDF arrives as an image, so search, clause checks and archiving all wait on retyping **One set of defaults for everything**: The same conversion settings get used on a pitch deck and an executed agreement ### With FlowRunner **Format decided by the destination**: Files are normalized on arrival, so every later step in the flow handles one shape of document **Scans carry text the day they land**: Recognition runs before filing, so the archive is searchable without anyone retyping a page **Settings chosen per document**: Layout and recognition are picked from what the document is and where it is going next ## Use Case Scenarios ### One format in, whatever the sender had Client documents reach a shared mailbox as Word files, spreadsheets, HTML exports and photographs of paper, in no predictable mix. The agent calls Convert To PDF on anything that is not already a PDF and Convert Image on the photographs, so the rest of the intake flow has exactly one shape of file to deal with. The normalized document goes to [SharePoint](https://flowrunner.ai/integrations/sharepoint) or [Box](https://flowrunner.ai/integrations/box) with the matter reference attached, and the intake record updates in [HubSpot](https://flowrunner.ai/integrations/hubspot). Downstream steps stop carrying a branch for every format a client might have used. ### Numbers out of a document, not out of a sentence A fee schedule or a rate exhibit arrives as a PDF table. Rather than reading the text and guessing at columns, the agent converts the document straight to CSV or JSON and writes the rows into [Google Sheets](https://flowrunner.ai/integrations/google-sheets) for review, or into [NetSuite](https://flowrunner.ai/integrations/netsuite) once they are confirmed. When a conversion pair is refused, List Supported Conversions is the vendor's own answer on what is possible, so the flow reports a real reason instead of retrying. Before a large batch, Get Credit Balance tells the agent whether it can finish, which is a better place to find out than halfway through. ### A signed scan that has to become searchable and then editable An executed agreement comes back as a scan from the counterparty's own printer. Make PDF Searchable produces a PDF whose text can be selected and searched, which is the copy that gets archived, indexed and quoted from. For the next amendment round, Convert PDF produces an editable copy. Those two outputs serve different purposes and carry different risks, and the agent treats them differently: the searchable archive copy is filed automatically, and the editable copy waits, because turning an executed document into something people can rewrite is a decision with a person's name on it. ## Human-in-Loop Highlight Convert PDF offers a layout mode, and the default is reflowable, because reflowable is genuinely easier to edit and is usually right for a Word copy. It is exactly wrong for an executed agreement. Reflowing repaginates the document, and an agreement's pagination is not cosmetic: amendments cite page and clause, initials sit against specific pages, and a schedule that lived on page 7 is now somewhere else with no error anywhere to say so. There is no clean reverse either, since Convert To PDF on the reflowed file produces a third pagination rather than the original one. So the agent does the conversion, counts the pages on both sides, and stops. It posts to the contracts channel: "The countersigned MSA converted from 18 pages to 15 in reflowable mode. Amendment 2 cites page 11 clause 9.3. Send the reflowable copy for markup, reconvert in fixed layout, or keep the redline in the PDF?" A person answers, because the question is about what this document is, and that is not something a page count can settle. FlowRunner's [human-in-the-loop](https://flowrunner.ai/concepts/human-in-the-loop) step is the difference between a conversion service and a workflow that can be trusted with a contract. 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 9 actions ### Conversions 4 - **Convert PDF** Turns a PDF into Word, Excel, PowerPoint, HTML, CSV, text, JSON, Markdown or images. Layout mode is a real choice here: reflowable is easier to edit, fixed keeps the page looking exactly as it did. - **Convert To PDF** Turns a Word, Excel, PowerPoint, HTML, CSV, text or image file into a PDF. The normalizing step that lets an intake flow stop branching on format. - **Convert Image** Turns an image straight into Word, Excel, PowerPoint, HTML, CSV, text or JSON, without producing a PDF in between. - **Make PDF Searchable** Runs recognition over a scanned PDF and gives back a PDF whose text can be selected and searched. Run it before filing anything that arrived as an image, or the archive holds a picture of the answer. ### Results 5 - **Download Result** Downloads a converted file and stores it in file storage. ComPDF returns a download URL that does not last, so this is how an output is kept. - **Get Task** Reads a conversion task, including the download URL of each file once it has finished. Used for anything long running. - **List Tasks** Lists recent conversion tasks on the account, for reconciliation and for tracing what a flow submitted. - **List Supported Conversions** Lists every source and target format pair the API supports, as the vendor reports them. The first thing to check when a conversion is refused. - **Get Credit Balance** Reads how much conversion credit is left. Needs no identifiers, so it works as a connection check and as the guard before a batch run. ## Frequently Asked Questions ### What can FlowRunner do with ComPDFKit PDF Converter? FlowRunner agents can run Convert PDF, Convert To PDF, and Convert Image in ComPDFKit PDF Converter, plus 6 more actions. ### Does connecting ComPDFKit PDF Converter to FlowRunner require OAuth? No. ComPDFKit PDF Converter connects to FlowRunner with an API key, no OAuth flow required. ### Can ComPDFKit PDF Converter trigger a FlowRunner workflow automatically? ComPDFKit PDF Converter doesn't currently expose triggers in FlowRunner. It connects as an action step inside workflows started by another trigger. **Work at ComPDFKit PDF Converter?** This integration exposes ComPDFKit PDF Converter 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/compdfkit-pdf-converter. Site index: https://flowrunner.ai/llms.txt