ComPDFKit PDF Editor
Document ProcessingComPDFKit PDF Editor edits PDFs through the ComPDF Cloud API without opening a file. Agents merge, split, rotate, compress, and watermark documents, add or remove pages, compare two versions, and convert to PDF/A for archiving.
What This Integration Enables
There is a fact about this connector worth stating before anything else: nothing in it changes the file you send. Every operation produces a new document and leaves the input alone, including Delete Pages, which is a page selection rather than a destructive call. That is unusual for a document toolkit, and it is what makes this a comfortable place to let an agent work at volume. Merges, splits, rotations, compressions and archival conversions can run unattended because the worst outcome is a bad output sitting next to a good original, not a good original that is now gone.
So FlowRunner agents use it for the assembly work that eats afternoons. Merge PDFs joins a cover letter, a scope document and an appendix into one pack in a fixed order. Split PDF cuts a scanned batch into one document per record at page ranges you name. Compress PDF shrinks something that will not fit a mailbox, Convert PDF Standard produces the PDF/A version for anything that has to survive an archive, and Compare PDFs marks what changed between two versions so a reviewer reads a diff instead of a document. Add Watermark and Remove Watermark are the pair that carries weight, because in most teams a watermark is not decoration. It is the status of the document.
Without FlowRunner
With FlowRunner
Use Case Scenarios
-
The pack that assembles itself
A proposal moves to ready on the deal record in HubSpot. The agent pulls the cover letter, the statement of work, the rate card and the case study appendix from the client folder in Box, calls Merge PDFs in the order the proposal template specifies, and stamps the result with Add Watermark. Rotate Pages fixes anything that came in sideways from a scanner, and Compress PDF runs only when the merged pack is over the mailbox limit, with the size reported before and after rather than compressing everything by reflex. The draft pack lands back in the folder and a link goes to the account lead. Because the merge takes between two and five files, a longer pack is assembled in stages, which the flow handles rather than a person discovering it.
-
One scanned batch becomes one document per record
A client returns twelve signed change orders as a single scanned file. Page selections in this API are one entry per output, separated by semicolons, so the agent submits the boundaries it derived and gets twelve documents back in one call. Each piece is filed against its own record, with the reference written into Airtable so the delivery team can see which change orders came back and which are still outstanding. Extract Pages does the narrower version of the same job when only the signature pages are wanted for a countersignature file, and Insert Blank Pages is there for the print-side requirement that a section must start on a right-hand page.
-
Show the reviewer the change, not the document
Before a revised scope goes back to a client, the agent runs Compare PDFs between the version that was sent and the version coming back. Text comparison is the default and is right for anything with a text layer; overlay comparison handles scans and drawings where there is no text to compare. The marked-up result goes to Slack with the pack attached, so the account lead reads three highlighted changes rather than fourteen pages. That is the digital andon cord in its most ordinary form: the work continues, and the one thing a person needs to look at is put in front of them.
Human-in-Loop Highlight
This connector has no operation that destroys anything, which makes it easy to say where the single consequential call is. Remove Watermark takes the DRAFT stamp off a document, and once it is off, the resulting file is indistinguishable from the version a team meant to send. In most agencies the watermark is doing real work: it is the only thing on the page saying this has not been approved yet. Taking it off is not a formatting step, it is an assertion that somebody signed off, and the API will make that assertion in one call without asking anyone. So FlowRunner's human-in-the-loop step sits exactly there. The agent posts to the account channel: "The Northwind proposal pack is assembled, 14 pages, stamped DRAFT. Compare PDFs found three changes since the version they saw: the scope section, the rate table, and a new page 9. Removing the watermark produces the copy that goes to the client. Release it, or send it back?" The account lead answers. Everything upstream of that question ran without anyone watching, which is the point.
Agent Capabilities
15 actionsPages
6- Merge PDFs Joins several PDFs into one, in the order you list them. The vendor's limit is between two and five files per call, so longer packs assemble in stages.
- Split PDF Cuts a PDF into several documents at the page ranges you name. Page selections are one entry per output, so a single call can turn a scanned batch into one file per record.
- Extract Pages Pulls the pages you name into a new PDF and leaves the original alone. Used when only the signature pages or a single schedule are wanted.
- Delete Pages Removes the pages you name. Despite the name, this is a page selection: the file you send is untouched and the result is a new document.
- Insert Blank Pages Adds blank pages at the position you choose. Used for print requirements where a section has to start on a particular side.
- Rotate Pages Turns the pages you name by 90, 180 or 270 degrees. The usual fix for pages that came off a scanner sideways.
Document
3- Compress PDF Makes a PDF smaller. Quality runs from 1 to 100, and lower means a smaller file with more visible loss, so a flow reports the size before and after rather than compressing by default.
- Convert PDF Standard Converts a PDF to a PDF/A archival standard. The step before anything is filed that has to still open in ten years.
- Compare PDFs Compares two PDFs and marks what changed, either by reading their text or by overlaying the pages. Text comparison is the default; overlay handles scans and drawings with no text layer.
Watermarks
2- Add Watermark Puts a text or image watermark on the pages you choose, tiled across the page if you want it repeated. The way a draft announces that it is a draft.
- Remove Watermark Takes the watermark off a PDF. The one operation on this connector that changes what a document claims about itself, and the reason this page has a human gate.
Results
4- Download Result Downloads a processed 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 task, including the download URL of each file once it has finished.
- List Tasks Lists recent tasks on the account, for reconciliation and for tracing what a flow actually submitted.
- Get Credit Balance Reads how much credit is left on the account. 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 Editor?
FlowRunner agents can run Merge PDFs, Split PDF, and Extract Pages in ComPDFKit PDF Editor, plus 12 more actions.
Does connecting ComPDFKit PDF Editor to FlowRunner require OAuth?
No. ComPDFKit PDF Editor connects to FlowRunner with an API key, no OAuth flow required.
Can ComPDFKit PDF Editor trigger a FlowRunner workflow automatically?
ComPDFKit PDF Editor doesn't currently expose triggers in FlowRunner. It connects as an action step inside workflows started by another trigger.
Other Document Processing integrations
Start building with ComPDFKit PDF Editor
Free plan, no card required. Connect in minutes.