HTML to IMG API
Developer ToolsConvert live web pages or raw HTML and CSS into images and PDFs, render charts from data, and run PDF utilities. Agents produce visual output from markup without running a browser themselves.
What This Integration Enables
HTML to IMG API is the broadest of FlowRunner's rendering connectors, and the breadth is the argument for it. One connector renders raw markup to an image, renders the same markup to a PDF with real page geometry, and produces a Chart.js chart without you writing any HTML at all. It also loads a JavaScript-heavy page in a real browser and hands back the fully hydrated markup, then does document surgery on the result: merge, split, unlock, watermark, compress. A statement pipeline that would otherwise stitch together three vendors runs against one.
Templating is inline rather than stored. Placeholders written as {{name}} in the HTML are substituted from a parameter map on each call, so one template produces personalized invoices, reports, or certificates without a template registry to keep in sync. Results come back as a hosted URL, a base64 string, or bytes written straight into FlowRunner file storage. Hosted files are removed automatically after their retention window, which is the reason Save Result To File Storage exists and the reason a durable copy is a step rather than an afterthought. Agents run all of it unattended until they reach an operation that changes what a document is allowed to be, and then they ask.
Without FlowRunner
With FlowRunner
Use Case Scenarios
Month-end statement packets
Close completes in QuickBooks Online. The agent renders each customer statement with Generate PDF from a single HTML template, filling {{account_name}}, {{period}}, and the line item rows from the ledger. It checks each output for placeholder text that failed to resolve, merges the statement with its supporting schedule using Merge PDFs, calls Save Result To File Storage for the durable copy, and sends through Gmail. The finance lead sees a summary of what went out and what was held, not a folder of files to check.
Visual archive of a supplier portal that renders in JavaScript
A procurement workflow needs weekly evidence of supplier pricing from a portal that builds its page client side, where a plain HTTP fetch returns an empty shell. The agent calls Get Page HTML, which loads the page in a real browser and waits for a named CSS selector to appear, then passes the hydrated markup to Generate Image for a dated capture. Compress Image shrinks it, Save Result To File Storage keeps it, and the archive lands in Amazon S3 with the week's price table readable in the picture.
Draft board reporting with the draft marked as a draft
An agent builds the quarterly reporting pack: Generate Chart renders the revenue and margin series straight from a Chart.js data object, the charts are embedded in the report HTML, and Generate PDF produces the document. Because the pack is not final, Add Watermark stamps diagonal draft text across every page before it is shared. Split PDF pulls the two pages the audit committee asked for into a separate file. The reviewers get something they cannot mistake for the approved version.
Human-in-Loop Highlight
Unlock PDF is the operation on this connector that undoes another person's decision. Someone applied a password to that document on purpose, and Unlock PDF removes it and returns the result at a publicly reachable hosted URL, which can be chained directly into Merge PDFs and out to a recipient in the same flow with nobody in between. The mechanics make it worse: the unlocked file is fetched over a public link, so a protection that existed a second ago is now a URL. So the agent treats an encrypted source as a stop, not a step. It posts to the finance channel with what it knows: "Packet for Northwind includes bank-confirmation.pdf, which is password protected. Unlocking it and merging it into the outbound packet removes that protection from the copy the customer receives. The password is available in the flow. Unlock and merge, send the packet without it, or route the packet to you for manual assembly?" A controller answers, the decision is attached to the run, and the packet continues. The agent handled the ninety-nine documents that needed no judgment. It stopped on the one that did.
Agent Capabilities
10 actionsImage Generation
2- Generate Image Captures an image of a live web page or renders one from raw HTML and CSS. Placeholders written as {{name}} are replaced from the dynamic parameters map, which makes it a lightweight template renderer. Output can be a hosted URL, a base64 string, or binary bytes written to FlowRunner file storage.
- Generate Chart Renders a chart image from Chart.js data and options with no HTML involved. Supports line, bar, pie, doughnut, radar, polar area, bubble, and scatter. Used when a report or an outbound message needs a visual and the flow already has the numbers.
Document Generation
1- Generate PDF Renders a live web page or raw HTML and CSS into a PDF. The same {{name}} substitution applies, so one template produces personalized invoices, reports, or certificates. Page geometry is set by a named paper size or explicit dimensions, and headers, footers, and background graphics are toggled per call.
Image Editing
1- Compress Image Reduces the file size of an image supplied as a public URL or a base64 string and reports the original size, compressed size, and percentage saved. Used to shrink generated assets before emailing or archiving them.
Web Capture
1- Get Page HTML Loads a page in a real browser and returns the fully rendered HTML, including markup produced by JavaScript that a plain fetch would miss. Supports waiting for a CSS selector, custom headers, and a custom user agent. Used to capture single-page applications before rendering them.
PDF Tools
4- Merge PDFs Combines between 2 and 15 PDF documents supplied as public URLs into a single file in the order given. Used to assemble statement packets, application bundles, and reporting packs from separately rendered parts.
- Split PDF Extracts a selection of pages from a PDF and returns the result. The selection accepts a single page, a range, a comma-separated combination, or an open-ended range. Used to pull the section a specific reviewer asked for out of a larger document.
- Unlock PDF Removes password protection from a PDF and returns the unlocked document at a hosted link. The correct password is required. This is the operation the human-in-loop gate above is built on, because it strips a protection someone deliberately applied.
- Add Watermark Stamps repeating diagonal watermark text across every page of a PDF or across an image, with control over opacity, rotation, and font size. Used to mark drafts, previews, and confidential copies so a reader cannot mistake one for the final.
Files
1- Save Result To File Storage Downloads a hosted image, chart, or PDF produced by any generation operation into FlowRunner file storage and returns both URLs. Hosted files are deleted automatically after their retention window, so this is the step that makes a result durable.
Frequently Asked Questions
What can FlowRunner do with HTML to IMG API?
FlowRunner agents can run Generate Image, Generate PDF, and Generate Chart in HTML to IMG API, plus 7 more actions.
Does connecting HTML to IMG API to FlowRunner require OAuth?
No. HTML to IMG API connects to FlowRunner with an API key, no OAuth flow required.
Can HTML to IMG API trigger a FlowRunner workflow automatically?
HTML to IMG API doesn't currently expose triggers in FlowRunner. It connects as an action step inside workflows started by another trigger.
Start building with HTML to IMG API
$100 in credits. No card required. Connect in minutes.