Transmission and processing
The website and its API use HTTPS for communication between your browser and the service. Documents are sent to the server to run the selected tool. The service and its hosting provider can therefore process their content; this is not end-to-end encryption.
File access
The account session associates uploaded files with their owner. The engine checks ownership before each operation. Storage is not exposed as a public directory.
Results are available through temporary links containing an access token. Anyone who obtains such a link can download the result while it is valid. Keep these links confidential and avoid publishing them or including them in a public report.
Sign-in
Sign-in uses a one-time code sent by email, valid for fifteen minutes, together with a verification phrase shown in the browser. No password is stored. Code requests and attempts are rate-limited by the platform.
An email informs you of every new sign-in to your account, with the device, the IP address and the country of the connection.
From your account, you can add an authenticator app as a second factor. Its 6-digit code is then requested after the email code; recovery codes, shown once, let you sign in if you lose the app.
Sign out on a shared device.
Temporary retention
The cleanup rule makes each file eligible for deletion 30 days with an account after it is created on the server. Original files and results each have their own creation date. Cleanup is scheduled every 15 minutes; an error or service interruption may delay it.
Account sign-in uses a one-time code sent by email. The profile and history are private. Links created from your space are valid for at most 15 minutes and never beyond the file expiry.
Closing the page or clearing your browser data does not immediately delete files already sent. Downloads saved on your device remain under your control.
Processing safeguards
The service checks received formats and applies limits to size, file count, pages and resource use. An invalid or overly complex document, or one exceeding these limits, may be rejected.
HTML and Markdown conversions do not execute document scripts or load external resources. Their rendering supports a subset of formatting; it does not reproduce a complete web page.
The engine’s error messages are limited to diagnostic information. They do not copy document content or names. Metadata and infrastructure logs remain necessary for operation; their scope is detailed in the privacy policy.
Choosing suitable documents
Keep your originals and check the results. If a document is subject to specific secrecy, location or hosting obligations, check that the service’s framework meets them before sending it. This page is not a security or compliance certification.
Storage location alone does not rule out all access from abroad. Read the information on hosting and transfers.
Reporting a problem
In the event of abnormal access, exposed data or a suspected vulnerability, use the security contact page. Describe the problem with an example without real data and limit your checks to your own data.