Test evidence without Print Screen and Word
Run the cases.
Walkstamp organises the proof.
Upload your spreadsheet of cases, open each one already filled in and record the run. Walkstamp turns the screen changes into steps with the exact time and what was being said, builds the document and hands the script back with status, date and who ran it.
Try it with a video — free See how the case script works
No sign-up to try it, no install, no size limit. The video, the audio and the transcript are processed inside your browser.
The Monday of someone running a round of cases
This is not a feature list: it is the same work, in the same order, done two ways.
| Today | With Walkstamp |
|---|---|
| Open the case in the spreadsheet and copy system, ticket and owner into the document. | The link opens the case already filled in. Nothing is retyped. |
| Screenshot at every step, paste into Word, write down what happened. | Screen changes become steps on their own, with the exact time and what was said. |
| Rename the files and remember which attachment belongs to which case. | The receipt and the fingerprint stay tied to the case that produced them. |
| Go back to the spreadsheet and update status, date and who ran it, row by row. | The script comes back as .xlsx with all of that filled in. |
Why the raw recording does not solve it
An hour-long video is not attachable anywhere: it does not fit in a ticket, it does not fit in minutes, and nobody will watch it. Sending the raw file to an AI has the same problem by another route — some refuse video outright, and the ones that accept it sample one frame per second, blind: 3,600 images and over a million tokens, almost all of it the previous screen again. Taking screenshots by hand has the opposite flaw: it is laborious, and the time of each one is lost.
One document, in whatever format your routine asks for
The same recording comes out as any of these. You choose at the end, and you can download more than one.
See every output
Five use cases, one recording
Pick what you have to hand over. The tool gets the rest right on its own.
Choosing the use case is the first step inside the tool, and it is what sets the document layout, the identification fields, whether clock time appears, what each frame is called, and the ready-made prompt for the AI.
Test evidence
UAT and audit: one screen per step, with clock time, identification header and integrity verification.
Read the full case →Work instruction
The key user’s step-by-step, in the imperative and with no date stamp — a go-live contractual deliverable.
Read the full case →Meeting minutes
The screens of what was shown, the speech paired to them, and the full transcript at the end.
Read the full case →Usability test
Moderator and participant kept apart, moments marked live, and redaction to anonymise.
Read the full case →Context for AI
~60 screens instead of 3,600, with the speech of each stretch and a ready-made prompt.
Read the full case →One job, two destinations
One recording. Two things nobody enjoys doing by hand.
Screenshotting every step and writing the caption underneath. Or explaining to a model what happened in an hour of video. Same problem — turning time into a document — and that is why one tool solves both.
You ran it. Now you have to prove it.
A test with no evidence does not survive an audit. Today that means Print Screen → paste into Word → write the caption, forty times. Record once and get the document ready.
- One screenshot per step, no repeats of a still screen
- The clock time of every step — and your narration becomes its written note
- A header with case, system, ticket and result: it becomes the first page and the file name
- Test data is usually a copy of production — and here it never uploads: there is no server to receive it
The model reads the whole document.
A model that watches video samples one frame per second, blind. An hour becomes 3,600 images and over a million tokens — and you pay again on every question.
- ~60 screens instead of 3,600, because the rest were repeats
- Each image paired with the speech from that stretch
- Fits in any model, including the ones that refuse video
- The instruction text comes along, ready to paste
The bridge between the two: what you want is the same — what happened on screen, in order, with the time, and with what was said. Only the reader changes at the end: an auditor or a model.
Who else does this, what it costs, and where each one wins — including against us: see the comparison.
How it works
Three steps, no account
Pick the video
Drag the file onto the page. It is read straight from your disk — 300 MB or 3 GB makes no difference, because nothing is uploaded anywhere.
Bring the transcript
Transcribe it right there, with Whisper running on your own machine, or drag in a .vtt / .srt caption file you already have.
Generate the PDF and copy the prompt
Review the frames, discard the bad ones and download the document. Along with it comes the text that tells the model how to read that PDF — that is what changes the quality of the answer.
What it does differently
Keeps the change, not the interval
Instead of grabbing a frame every ten seconds, it scans the video and stops at the moments the screen actually changes, to within half a second. In a slide deck that means one frame per slide instead of forty of the same one.
Stamps the time
Every screen carries the moment it happened. That is what separates a pile of screenshots from a document that works as evidence — and what lets the AI cite the right moment.
Your video never leaves your computer
There is no server that receives your video. The video, the audio and the transcript are processed in your browser — including confidential meeting recordings.
Hands you the prompt too
The ready-made text tells the model to cite timestamps, separate what it saw from what it heard, and admit when the information is not in the document instead of inventing it.
Privacy
We do not have your files because we have no server
Most online video tools require an upload: your file goes to someone's machine, is processed there and stays stored for a period you do not control. Walkstamp does none of that — the page is a static file, and all processing happens in your browser.
Using the tool needs no account, and the video, the audio and the transcript never leave your browser. Paid plans do have an account — it keeps what billing and the case script require, and nothing beyond what you attach on purpose. Usage measurement is three marks with no identifier, and it honours Do Not Track. The only other connection is downloading the transcription model, and only if you use automatic transcription. Read the full policy.
A single case is free. Repeated work becomes a plan.
The three are told apart by the work they do, not by a feature list.
Free — one-off case
Turn any recording into evidence, with no sign-up and no size limit.
Personal — your own script
Upload the spreadsheet of cases, open each one filled in, and hand the script back with status, date and who ran it.
Team — coordinated runs
Split the round across people, follow who finished what, and deliver in the same format.
Frequently asked questions
Do I need to install anything?
No. It opens in the browser and works. An up-to-date Chrome or Edge gives the best experience, because automatic transcription uses the graphics card through WebGPU. On phones and tablets it opens, but no browser there can record the screen — that is a device limit, not a limit of the tool. In that case record with the phone's own screen recorder and bring the file in.
What is the maximum video size?
There is no limit imposed by us, precisely because there is no upload. The practical limit is your computer's memory. Long videos take longer to scan, and you can stop the scan whenever you want.
Which formats work?
MP4 (H.264), WebM and MOV in most cases. Formats the browser cannot decode — HEVC and some MKV files — will not open; those need converting to MP4 first.
Is the transcription any good?
It uses Whisper running locally. The fast model is reasonable and the accurate one is considerably better, at the cost of a larger download. If you already have captions from another service, drag the file in and skip this step.
Is it really free?
It is free, and the reason is structural: your computer does the work, so serving you costs me nothing. No ads, no data of yours being sold and no sign-up to use it — not out of generosity, but because there is no server that receives your video. Paid plans do have an account, and it never receives it either.
Which AI should I use the PDF with?
Any model that accepts attachments and reads images — Claude, ChatGPT and Gemini all work. Attach the PDF and paste the prompt the tool generates alongside it.
How do I check that nothing is really uploaded?
Three ways, each a minute long, none of them requiring you to trust me. Open the developer panel (F12), go to the network tab and process a video: you see every request the page makes — and none of them carries your file. Second: the code is a single page; right-click, view source, and it is all there to read. Third, the most direct: turn the internet off once the page has loaded and use it — the recording, the frames and the document still come out, because none of that needed a network. If it works with no network, it is not sending anything anywhere.
What is kept in the account?
The script fields you uploaded yourself — case, title, system, ticket and owner —, who completed each case and when, the name of the delivered file, its fingerprint and your note. Plus invoices and support tickets. And one file, if you choose: the complete .json of a session, attached by you to a case and deletable at any time. Video, audio, transcript and images never get here.
Can I import my current spreadsheet?
Yes: .xlsx, .csv, .tsv, .txt or pasted straight from Excel. Before saving, you check which column is the case code, which is the title, the system, the ticket and the owner — and fix whatever the reading got wrong. Up to 2,000 rows per script.
What about Jira, Zephyr and TestRail?
Only Jira, and it is worth being exact: it is not an integration, it is a summary ready to paste into the ticket comment, with the PDF attached alongside. There is no connection to Zephyr or TestRail today, and I am not going to claim otherwise.
What happens when I cancel?
The subscription stops at the end of the paid period, and from then the policy clock runs: account content is deleted within 90 days. The invoice survives — tax records have a statutory retention period longer than that, and it is not our choice — so the customer row stays, reduced to the minimum that keeps the invoice legible. Paying again before that resets the clock.
Do I have to upload my video to use Personal or Team?
No. The tool is the same on all three plans and still runs inside your browser. What the account receives is the script — the case fields and what was completed — and, if you want it, the .json of a session attached to a case. The video never uploads, on any plan.
How does per-person billing work?
Team is per seat, starting at 3 people. Each invited person takes a seat; blocking someone frees that seat for another person, with no new charge. Personal is a single subscription, for one person.
Record two minutes and look at the document
It takes less time than reading this page.
Open the tool