> ## Documentation Index
> Fetch the complete documentation index at: https://docs.hired.tools/llms.txt
> Use this file to discover all available pages before exploring further.

# Publishing and exporting

> A real PDF for the forms that want a file, an unlisted link for the ones that want a URL, and exactly what each one exposes.

Application forms ask for one of two things and never agree on which. Hired does both,
and the difference between them is worth understanding before you send either.

<CardGroup cols={2}>
  <Card title="A link" icon="link">
    `publish_resume`, or **Share → Create a link** in the editor. An unlisted address
    anyone holding it can read without signing in.
  </Card>

  <Card title="A file" icon="file-pdf">
    `export_resume_pdf`, or **PDF** in the editor. A real PDF, rendered server-side,
    downloaded to your machine.
  </Card>
</CardGroup>

## Publishing a link

```
https://your-instance.example.com/r/staff-engineer-stripe-k7m2qx4bnp8t
```

Or just ask: *"publish my Stripe resume and give me the link."*

### The privacy model is the address, and nothing else

The slug is long and random, so it cannot be guessed or walked. The page tells search
engines not to index it. It is listed nowhere. That is the whole model — there are no
per-viewer permissions and no passwords, because everyone you would send this to is
someone you already decided to send it to.

<Warning>
  **If the resume uses a template that draws a photo — Classic, Modern, Compact or
  Editorial — and has `showPhoto` on, your face is on that page**, visible to anyone
  holding the link. Decide that before you publish. An assistant is instructed to mention
  it, and to offer to turn the photo off first.

  Harvard, the default, never renders one whatever `showPhoto` says, so a default resume
  carries no face.
</Warning>

Your private notes on the resume are never shown on the public page.

### Republishing and withdrawing

Publishing a resume that is already published returns the **same** URL, so it is safe to
ask twice.

`unpublish_resume` — or **Withdraw** — **destroys** the address rather than pausing it.
The page starts returning "not found" for everyone immediately, and publishing that
resume again later gives a completely different link. That is deliberate: the reason to
withdraw is usually that a URL went somewhere you regret, and a pause you can undo does
not fix that.

Nothing about the resume itself is deleted either way.

## Exporting a PDF

Open a resume and press **PDF**. The file downloads. There is no print dialog and no
margin setting to get wrong.

The server renders the same page you would have printed by hand, so what you see is what
you get: exactly 8.5in × 11in, with the Harvard template's half-inch margins. The output
is **real selectable text**, not an image, so an applicant tracking system can read it.

Ask instead and you get the thing you actually wanted to know:

> Export my Stripe resume.

`export_resume_pdf` hands back a download link *and* the page count, measured from the
rendered document rather than estimated. "Does this fit on one page?" is answered before
you send it rather than after.

<Note>
  The download URL opens in your browser, where you are already signed in. It is not a
  public link and nobody else can fetch it.
</Note>

### No webfont is fetched

The serif stack is Tinos → Times New Roman → Liberation Serif, which are metrically
identical. The document renders the same on macOS, Windows and Linux with nothing to
download and nothing to expire.

### If your host has no headless browser

Server-side rendering needs a Chromium on the machine. The Docker image ships one, so a
compose deploy has working one-click PDF from first boot. A stock Railway deploy does not.

Where it is missing, the **PDF** button says so and the fallback still works:
**⋯ → Open print view**, then your browser's **Save as PDF** with margins set to
**None**. Same document, one more step. `export_resume_pdf` reports the same thing rather
than failing silently.

[Getting one-click PDF on a host without Chromium →](/self-hosting/pdf)

## Checking length before either

`preview_resume_text` renders a document to plain text and estimates its length
**without saving**. Use it while there is still something to cut:

> Show me that resume as text and tell me if it fits on a page.

For the measured answer, export the PDF — the page count comes back with it.

## Which resume went where

An application can have a resume attached, and `diagnose_search` reports the response rate
of each resume so you can see which one is actually working. That is only true if you
attach them, which is one field on the application — `resumeId`, or the cell in the
pipeline table.

<Tip>
  Never edit a resume that is already attached to an application. That is the version you
  actually sent, and it is the record of what a recruiter read. `duplicate_resume` first,
  then tailor the copy.
</Tip>