You've photographed a document with your phone. Now you need the text, so you download the PDF and re-upload it to the OCR. Then the text needs translating: you copy it, open another tool, paste it. Then you want a summary: copy, open, paste. Four useful steps, seven transfer operations. Most of the time doesn't go into the processing: it goes into moving files from one tool to another.
The hidden cost of "download and re-upload"
Every manual hop between two tools costs more than it seems. There's the time of the download and the fresh upload, which on a PDF of a few dozen megabytes isn't negligible. There's the mess in the Downloads folder, where document(3).pdf and final-text-real.txt pile up. And above all there's the risk of losing the thread: after three hops it's no longer obvious which version is the current one.
There's a less visible cost too: every time you reopen a file from disk, your train of thought breaks. The cognitive work of "what was I doing and why" has to be rebuilt, and across a sequence of four tools that happens three times.
What a handoff area does
A workspace isn't an archive: it's a temporary drop-off point. One tool leaves its result there — a file or a block of text — and the next tool picks it up ready to go, without a round trip through the disk.
The practical difference lies in what's being carried. A file keeps its original format and is what you need when the next step works on the whole document. Text is the result of an extraction — typically from OCR or a summary — and is what you need when the content matters more than the layout.
That distinction determines what makes sense next: a file can take translation, summarisation or page edits; text takes formatting, rewriting or export into a document. Which is why the tools offered as a next step change depending on what you just sent.
The chains people actually use
| Goal | Sequence |
|---|---|
| A paper document becomes editable text | Scanner → OCR → Editor |
| A foreign-language paper needs understanding fast | PDF search → Summary |
| A scanned contract has to be read in another language | Scanner → OCR → Translation |
| Meeting notes become a presentation | Formatter → AI slides |
| A book found online needs summarising | PDF search → Summary → Editor |
These are all sequences in which no single step makes sense on its own: scanning without OCR produces an unusable image, running OCR without a destination produces text that sits in the clipboard. The value is in the chain, not in the individual links.
How to connect two tools
- Work normally with the first tool until you have a result.
- Send the result to the workspace instead of downloading it.
- Open the workspace: you'll find the item waiting, showing which tool it came from and the next steps consistent with its type.
- Pick the next tool: it receives the content already loaded, with no file selection.
- If at some point you need the file on disk, you can still download it from the workspace.
The limitation to know before you start
There's one aspect to keep in mind, and it's deliberate: the content in transit lives only in the memory of the browser tab. Reloading the page clears it, exactly as would happen if you abandoned any of the tools mid-task.
It's a choice consistent with the fact that these tools require no sign-up and don't retain documents: if the handoff survived closing the tab, that would mean it had been saved somewhere. The practical rule that follows is simple: complete the chain in one session, and if you have to stop, download the file from the workspace before closing.
Frequently asked questions
Does the workspace store my documents?
No. It's a temporary handoff area: the content stays in the browser tab's memory for as long as it takes to pass it to the next tool, and it's cleared by reloading the page. It isn't an archive and shouldn't be used as one.
What can I transfer from one tool to another?
Two kinds of content: a file, which keeps its original format, or a block of text, typically the result of an extraction or a summary. The kind determines which follow-on tools are offered.
Why do the suggested steps change every time?
Because they depend on what you sent: a file takes operations on the whole document, text takes operations on the content. The tool the item came from is also excluded from the suggestions, so the step you've just done isn't offered again.
Can I download the file instead of continuing?
Yes: the waiting content can be downloaded from the workspace at any time. That's the thing to do if you have to interrupt the work, since the handoff doesn't survive closing the tab.
In summary
The time lost in a document workflow rarely sits in the individual processing step: it sits in the downloads and re-uploads between one step and the next, and in the thread lost along the way. A handoff area removes that cost by connecting the tools directly, offering at each point the steps consistent with the content type. The workspace holds the content in memory for the duration of the session — so it's best to complete the chain without reloading the page — and links together all the PDF and AI tools, from the scanner to OCR through to the summary.