n8n-nodes-printfriendly

This is an n8n node for PrintFriendly. Give it a web page address, some HTML or some Markdown. It returns a PDF, a Word file, Markdown, JSON, HTML or plain text.

Use it to save web pages as PDFs, to turn articles into Markdown for an AI agent or a summary, or to build documents from HTML or Markdown inside a workflow.

• Installation
• Credentials
• Operations
• Errors
• Limits
• Compatibility
• Support
• Development
• Resources

Installation

The package name is n8n-nodes-printfriendly.

n8n Cloud

n8n Cloud installs verified community nodes from the nodes panel. Only the instance owner or an admin can install a node. Everyone on the instance can then use it.

1. Open a workflow and open the nodes panel (select + or press N).
2. Search for PrintFriendly.
3. Under More from the community, select the PrintFriendly node.
4. Select Install.

If PrintFriendly does not appear, n8n has not verified the package yet. Use a self-hosted n8n instead.

Self-hosted n8n

In the n8n interface:

1. Go to Settings > Community Nodes.
2. Select Install.
3. Enter n8n-nodes-printfriendly.
4. Tick the box to accept the risks of community nodes, then select Install.

From the command line (for queue mode, Docker or any setup without the interface option):

cd ~/.n8n/nodes   # create it first if it does not exist
npm install n8n-nodes-printfriendly

Then restart n8n. See the n8n guide to installing community nodes for more.

Credentials

The node uses one credential, PrintFriendly API.

1. Get a free API key at printfriendly.com/me/api-key. The free plan includes 100 documents a month and needs no card.
2. In n8n, create a PrintFriendly API credential.
3. Paste the key into API Key.

The key is sent as Authorization: Bearer <key>. Leave API Base URL at https://api.printfriendly.com unless support tells you otherwise.

Test on the credential checks the key and does not use up a document. The API has no endpoint that only checks a key. So the test reads a job ID that was never issued. A known key gets 404 job_not_found, and the test passes. An unknown key gets 401 invalid_api_key, and the test reports a rejected key. Any other answer (a firewall block, a rate limit, an outage) fails the test.

Operations

All operations are on the Document resource.

Convert

Turns one source into one output.

Options (all optional, and left out of the request unless you add them):

What comes back is the API's JSON answer:

• PDF and Word: file_url (works for 48 hours), expires_at, size_bytes, document (title, author, date, site, excerpt, image, word count) and usage.
• Markdown, HTML and text: content and document.
• JSON: the whole article in document.
• Run Asynchronously: status: "queued", job_id, status_url and expires_at.

With a PDF or Word output, switch on Download File. The node then fetches file_url and adds the file to the item as binary data. The MIME type is right (application/pdf or the Word one). The file name comes from the document title. The JSON answer stays on the item, so file_url is still there. The download does not send your API key.

Example: a web page to a PDF file. Import this with Workflow > Import from File (or paste it onto the canvas). Pick your credential on the PrintFriendly node. Connect the output to anything that takes binary data, such as an email attachment, Google Drive, S3, Slack or Read/Write Files from Disk.

{
  "name": "Web page to PDF",
  "nodes": [
    {
      "parameters": {},
      "name": "When clicking Execute workflow",
      "type": "n8n-nodes-base.manualTrigger",
      "typeVersion": 1,
      "position": [0, 0]
    },
    {
      "parameters": {
        "resource": "document",
        "operation": "convert",
        "sourceType": "url",
        "url": "https://en.wikipedia.org/wiki/Pangram",
        "output": "pdf",
        "downloadFile": true,
        "binaryPropertyName": "data",
        "options": { "pageSize": "a4" }
      },
      "name": "PrintFriendly",
      "type": "n8n-nodes-printfriendly.printFriendly",
      "typeVersion": 1,
      "position": [240, 0]
    }
  ],
  "connections": {
    "When clicking Execute workflow": {
      "main": [[{ "node": "PrintFriendly", "type": "main", "index": 0 }]]
    }
  }
}

Read Page

A shortcut for the common case: a URL in, the article out, as Markdown or as JSON. It is a Convert with the source fixed to a URL, so it counts as one document. The only option is Include Links. It keeps link and image addresses. Leave it off for shorter Markdown, which suits AI agents and summaries.

The node can be used as a tool by an AI agent. Read Page is the operation made for that.

Example: read an article as Markdown. Set Operation to Read Page, URL to https://en.wikipedia.org/wiki/Pangram and Output Format to Markdown. The item that comes back has content (the article as Markdown) and document (title, author, date, site, excerpt, image and word count). Pass {{ $json.content }} to the next node, for example an AI Agent or a Summarization Chain.

Get Job

Reads a conversion that was started with Run Asynchronously. Enter the job_id from Convert. The answer has status (queued, processing, succeeded or failed), and:

• when succeeded: result, which is exactly what a synchronous Convert returns, usage included;
• when failed: error with code, message and retryable.

A failed job is returned as data, not thrown, so a workflow can branch on status. While a job is unfinished, the API asks for about 2 seconds between reads. Download File works here too. Once the job has succeeded with a PDF or Word file, the node attaches it as Convert does. Only the API key that started a job can read it. Jobs are kept for 48 hours.

Example: a slow page in the background. In a Convert node, add Run Asynchronously under Options. The item that comes back has a job_id. Add a Wait node (5 seconds), then a second PrintFriendly node. Set its Operation to Get Job, Job ID to {{ $json.job_id }} and switch on Download File. If status is still processing, wait and read the job again.

Errors

Failed calls are free. Only a successful conversion counts against your plan.

When the API refuses a request, the node stops with an error. The error title carries the API's code. The description says what to do, quotes what PrintFriendly said and gives the request ID to quote to support. With Continue On Fail on, the item carries the error instead.

A response that is not JSON at all (for example a firewall block page) is reported as PrintFriendly did not answer with JSON, with the HTTP status.

Limits

These come from the API documentation.

• Documents: the free plan includes 100 a month. One successful call is one document, whatever the output. Paid plans include more.
• Free plan PDFs carry the line "Made with the PrintFriendly API" at the foot of every page. HTML output carries it once at the end. Markdown, JSON, text and Word have none. Paid plans have none.
• Request rate: 10 requests a minute on the free plan, 120 on paid plans. Reading a job has its own limit: 120 a minute on the free plan, 600 on paid plans.
• Time: a synchronous call answers within 27 seconds, or with render_timeout. An asynchronous job has up to 55 seconds.
• Background jobs: 3 unfinished jobs per key on the free plan, 20 on paid plans. Jobs and their results are kept for 48 hours.
• Size: HTML up to 2 MB of markup (50 MB with embedded images). Markdown up to 1 MB.
• Files: PDF and Word links work for 48 hours. Download them if you need them longer.
• Use is governed by the Acceptable Use Policy.

Compatibility

Built against n8n 2.x (n8n-workflow 2.x) with the @n8n/node-cli toolchain. The node is declarative. It has no runtime dependencies and uses no file system or environment access.

The credential test relies on the ignoreHttpStatusErrors except list, which older n8n versions do not understand. On those versions the test may report success for a rejected key. The first run of the node then shows the real answer.

Support

• Questions about the API, your plan or your key: printfriendly.com/contact or support@printfriendly.com.
• Bugs and feature requests for this node: GitHub issues.

When you report an error, include the request ID from the error description. Do not post your API key.

Development

Use Node.js 24 or later. Current n8n-workflow releases depend on isolated-vm, which needs Node.js 24 and does not compile on Node.js 20.

npm install
npm run build   # compile and copy icons into dist/
npm run lint    # n8n's community-node lint rules, including eslint-plugin-n8n-nodes-base
npm test        # build, then run the unit tests in test/
npm run dev     # a local n8n with the node loaded (newer n8n versions need Docker or Podman)

scripts/live-check.mjs runs the node's own request builder, error mapping and file download against the real API, without an n8n instance. By default it uses two documents (one Read Page, one Convert to PDF with Download File). It also runs free checks of the credential test and of three refused calls.

PRINTFRIENDLY_API_KEY=your-key node scripts/live-check.mjs
# only the steps that cost no document:
PRINTFRIENDLY_API_KEY=your-key node scripts/live-check.mjs --steps probe,errors

Maintainers: see PUBLISHING.md for how to release a new version.

Resources

• PrintFriendly API documentation
• OpenAPI spec
• n8n community nodes documentation

License

MIT