(Also available at https://webmod.app/privacy)


Effective October 27, 2026

In short. Webmod does not collect anything from the pages you browse.
Beyond the account traffic needed to keep you signed in, the extension transmits
a page to us only when you click "Analyze and modify site" — and even then
the content of input fields is removed from the page structure and covered up in
the screenshot before anything leaves your browser. The page snapshot and
screenshot we receive are deleted from our servers
automatically after 24 hours; our AI provider, which needs them to write the
modification, keeps its own copy for a limited period of its own (see §5). Apart
from us, page content reaches only Anthropic (the AI that writes the
modification). Every recipient we engage directly, including our email provider,
is named in §5 with exactly what reaches it.

This service is offered to users in the European Economic Area and the United
States. Depending on where you are, different laws give you rights over your
data; §7 (GDPR) and §12 (United States / California) describe them.

Separately, the public website offers a waitlist: if you type your email
into the sign-up box and submit it, we store that email so we can tell you when
the extension launches. Apart from web server logs (§6), that is the only
thing the website stores; it happens only when you choose to submit the form, and
it is unrelated to the extension's collection boundary described above.

The controller of your personal data is Vítek Adam, IČO (business ID)
23991038, with registered seat at Slovenská 1766, 250 82 Úvaly, Czechia, a natural person doing
business under the Czech Trade Licensing Act and registered in the Trade Licensing
Register (živnostenský rejstřík). Wherever this policy says "we" or "us", it means
that person. "Webmod" means the browser extension, our public website and the
servers that run them.

For anything about your data, contact us by email at hello@webmod.app or by
phone at +420 735 635 191. We have not appointed a data protection officer, because
the law does not require us to.



This is the most important section of this policy.

While you browse normally, Webmod collects nothing from the pages you visit.
It does not read those pages, does not record your browsing history, does not track
you across sites, and sends us no page content. Modifications you have already
saved run entirely inside your browser and send nothing to us. A modification is
meant to communicate only with the website it changes, and only where the change
you asked for needs it — for example to load more items from that site; our safety
review is designed to reject a script that would send data to another server,
including ours (§11 explains what the review can and cannot guarantee).

Two things are, by necessity, not "nothing" — and we would rather name them than
let the headline overstate the case:

• Your account. When you are signed in, the extension exchanges the ordinary
  account traffic with our servers that any logged-in app does — issuing and
  refreshing your sign-in tokens, and the requests you yourself start. This never
  includes the content of pages you browse.
• The analysis you ask for. When you click "Analyze and modify site" on a
  page you have chosen, the extension captures that one page — both a screenshot
  of what is visible and the page's structure — and sends it to us. This is the
  only way page content reaches us, and it happens only on that click.

Apart from your account traffic, no page content is transmitted unless you click
"Analyze and modify site". There is no background collection, no telemetry on the
pages you visit, and no "anonymous usage statistics" gathered from browsing.


So that it can act on whatever page you are looking at when you click, the
extension asks, at install time, for permission to run on all sites, and it places
two small scripts on every page you open. We want to be plain about this, because
a broad permission can look at odds with "collects nothing":

• One script simply waits for your click. Until you press "Analyze and modify
  site", it reads nothing from the page and sends nothing anywhere. It has no
  network connection of its own.
• The other does one narrow technical job — it makes certain hidden parts of a
  page (so-called "closed shadow roots") capturable later if you ask for an
  analysis. It keeps this only in memory, writes nothing to disk, and transmits
  nothing.

Neither script collects, stores, or sends anything during ordinary browsing. The
broad permission exists only so that the capture can happen on the page you choose,
at the moment you choose it — not so that we can watch you. If you prefer, you can
restrict the extension's site access from your browser's extension settings;
analysis will then prompt for access to a site the first time you use it there.


When you click "Analyze and modify site", the following is transmitted:

| Data | Why | Retention |
|---|---|---|
| Screenshot of the visible page, with form fields covered up — see §3 | The AI needs to see the layout to understand what you are pointing at | 24 hours on our servers, then deleted automatically. Anthropic keeps its own copy for a limited period — see §5 |
| Page structure (HTML), with input content stripped — see §3 | The AI queries it to write selectors that target the right element | 24 hours on our servers, then deleted automatically. The parts the AI reads reach Anthropic — see §5 |
| The address of that page | To scope the modification to the right site and pages. In the record we keep with your modification we store only the site and path — the part of the address after ? is dropped before that record is written | The stored record is kept with the modification; see also §3 on the snapshot |
| Your description of the change and the area you marked on the page (only its position and size) | It is the request itself | Kept with the modification record; both reach Anthropic (§5) |
| Your answers to the AI's clarifying questions | They define what gets built | Kept with the modification record; your answers reach Anthropic (§5) |


Before any page structure leaves your browser, the extension strips the contents
of form fields — this happens on your own machine, in the extension, not on our
servers. If that stripping cannot complete, the capture is abandoned rather than
sent. Our servers then run the same stripping again on arrival, so a bug or an
outdated extension version cannot cause field content to be stored.

Specifically removed: text typed into text boxes and text areas, password fields,
the selected state of checkboxes, radio buttons and dropdowns, the contents of
file selectors, and the contents of rich-text editors.

Deliberately kept, because the AI cannot write a working modification without it:
element names and identifiers, CSS classes, the page's own styling, and the labels
and placeholder text the site's authors wrote.

In the screenshot, form fields are covered up. Before the screenshot leaves your
browser, the extension paints over the same fields it strips from the page structure
— text boxes and text areas, dropdowns, and rich-text editors — so that what has
been typed into them cannot be read. This covers fields on the page and inside its
shadow roots (the same reach as the structure stripping). Two things it does not
cover: fields inside a frame embedded from another website (such as an embedded
payment widget), because the extension cannot reach into another site's frame to
either strip or mask it; and, like the structure stripping, a checkbox's or radio
button's ticked state, which is not typed text. For anything in an embedded
third-party frame, treat it as visible in the screenshot. The rest of the page
appears in the screenshot as you see it on screen.

The address we keep is reduced to site and path. The record stored with your
modification keeps only the site and the page path; the part after ? — which
often carries what you searched for, or a token identifying your session — is
dropped before that record is written, and it is never written to our logs.

A page snapshot is a capture of the page as it was. The HTML snapshot and the
screenshot are a picture of that one page at the moment you clicked. Because they
are the page itself, they may still contain things the page was displaying —
including, in the page's own address or markup, a query string — even though the
separate address record we keep is reduced to site and path. This is why the whole
snapshot is held only briefly: on our servers, screenshots and page structure are
deleted after 24 hours. The parts of them that the AI reads are sent to Anthropic,
which keeps its own copy for a limited period of its own (§5), so our 24-hour limit
does not control that copy. Do not run an analysis on a page showing something you
would not want processed — a bank statement, medical records, private messages.

We keep a count of what was removed from each snapshot — how many fields were
emptied, and of which kind — so that this promise can be checked after the fact
rather than merely asserted. The count records that something was removed, never
what it was.


Stored by the extension in your browser, never sent to us and never synced to your
browser account:

• your sign-in tokens, until you sign out or remove the extension;
• the modifications you chose to keep — the script, its undo script, and the
  addresses it applies to — until you delete them;
• a short-lived marker so a dropped connection can resume where it left off;
• a request you have typed but not yet sent, so closing the extension does not
  lose it — until you send it or sign out;
• for a modification you rejected while it was being made, the first few lines of
  its code and what you said was wrong with it, so the conversation can be shown
  again when you reopen the extension — until you keep a result, cancel, or leave
  that conversation;
• your settings choices (how many clarifying questions Webmod should ask, and
  whether the Enter key sends your message or starts a new line).

Removing the extension deletes all of it.

Modifications run as user scripts, and we show you the script first. Your
browser requires you to turn on user-script support for this extension before
anything can run, and the modification is generated by an AI and shown to you
in full before it executes. You decide whether to run it.


To use Webmod you create an account in the extension, with an email address and a
password. We store the address and a cryptographic hash of the password — never the
password itself. The address is used to sign in, to
verify the account, and for essential service messages. We do not send marketing
email; the only exception is the one-time launch announcement, if you joined the
website waitlist yourself (§4a).

Signing in. When you sign in we issue two tokens: a short-lived one your
browser sends with each request, and a longer-lived one used to obtain fresh
short-lived ones. We store only a hash of the longer-lived token, so our database
cannot be used to impersonate you. Each use replaces it; if a token is ever used
twice — which is what happens when one has been stolen — every token from that
sign-in is cancelled immediately and you are asked to sign in again.

What we do not record about your sessions. We do not store your IP address or
your browser's user-agent string against your account. An IP address may appear
only in our web server's logs, which we delete after 30 days (§6).

Your preferences. We store the choices you make in the extension's settings
against your account: how many clarifying questions Webmod should ask before
building a modification, and whether pressing Enter in the message box sends
your message or starts a new line. Webmod picks the Enter setting for you once,
the first time you send a message with the keyboard — for example, if you send
with Ctrl+Enter (⌘+Enter on a Mac), or press Enter and then cancel straight away,
it assumes you want Enter to start a new line. Only the resulting choice is
stored, never your keystrokes, and you can change it in the settings at any time.
These are preferences about how the tool behaves, not anything about the pages
you visit, and nothing is stored here until you change a setting or that first
keyboard send happens.

Emails we send you. Only messages about the service: confirming your address,
resetting your password, and essential notices about your account and the service
(for example about a change to the terms or the outcome of a claim). Confirmation
and reset messages contain a single-use link that expires (24 hours for
confirmation, 1 hour for a reset). Resetting your password signs you out
everywhere. So that these emails arrive in the language you use Webmod in, we
store your interface language (English or Czech) with your account; the extension
sends it when you register, ask for a confirmation or password-reset email again,
or request a data export.

If your address stops working. Each message carries a short code identifying
which address it was sent to, so that when mail is returned undelivered we know
whose it was. The code contains only your own address and is only ever present in
mail sent to you. If an address permanently rejects our mail — because it no
longer exists, or because someone reported the message as spam — we add it to a
list and stop sending to it. Temporary failures, such as a full mailbox, do not
have this effect: they are retried, so a brief problem never permanently cuts you
off from resetting your password.

Communications with us. When you write to us or call us, for example with a
question, a complaint, a claim about defective service, or to exercise your rights
under §7, we process your contact details and the content of the communication so
that we can answer and deal with the matter. §6 says how long we keep it.


The public website — the page describing Webmod, separate from the extension —
offers a waitlist so we can let you know when the extension is available.

Nothing goes on the waitlist until you submit the form. If, and only if, you
type your email address into the waitlist box and click to join, your email is sent
to our own server and stored. The website uses no third-party analytics,
advertising or tracking tools; as with any website, though, the web server may
record technical data about each request, including the IP address (§6).

| Data | Why | Retention |
|---|---|---|
| The email address you typed | So we can email you once, when the extension launches | Until launch notices are sent, or until you ask us to remove it — whichever comes first |
| Which language the page was in (en/cs) | So the launch email can be in the same language | Stored alongside the address |
| The time you joined | To process sign-ups in order and spot abuse | Stored alongside the address |

We do not store your IP address with the waitlist entry. An IP address is
used momentarily, in memory only, to limit how many sign-ups one source can send
in a short window; it is never written to the waitlist.

The waitlist is stored on our own server, as a plain file, and is not shared with
Anthropic or with any advertiser or analytics provider. It is used
only to send the launch announcement, which we send through our email provider
(§5). To be removed before then, email us at hello@webmod.app and we will delete
your entry.

A waitlist entry is not an account. It does not sign you in, and it carries no
password. If you later create an account, that is the separate process described
in §4.


We run our application servers, database, cache, file storage and AI observability
tooling ourselves, on our own servers (below). Apart from us, only the recipients
listed in this section receive data, and only to the extent needed to run the
service. Anthropic processes page content for us as our processor under a data
processing agreement (Art. 28 GDPR) and may not use it for its own purposes, such as
advertising or training AI models; the one exception — temporary retention to
protect against abuse — is described below under Anthropic. Our email is run by
Seznam.cz under its own terms of service; what it processes from emails for its own
purposes is described below.


Entity. Our contracting partner is Anthropic Ireland, Limited (Ireland); it
processes the data as our processor under a data processing agreement. The
underlying model inference runs on Anthropic's infrastructure in the United
States (see §9 on transfers).

What it receives. The screenshot, your description, the area you marked, your
answers, and the parts of the stripped page structure the AI asks to see. This
is unavoidable:
Anthropic's model is what writes the modification. The AI does not receive the whole
page — it queries the stripped snapshot through a small set of lookups ("find
elements matching this", "what is inside this", "what contains this text"), so only
the parts relevant to your request are sent. Your email address and account identity
are never included. Each generated script is also sent back for a safety review
before you see it.

What Anthropic does with it. Anthropic processes API data as a service provider
and, under its commercial terms, does not use API inputs or outputs to train its
models. However, it keeps its own copy for a limited period: under its data
retention rules it deletes API inputs and outputs within 30 days at most,
except where it must retain them longer to comply with law or enforce its usage
policies — in a case of suspected policy violation it may keep inputs and outputs
for up to 2 years (and safety classification results longer). This copy is on
Anthropic's systems and is separate from our own deletion; our retention periods
control only the copy on our servers. Data is sent over an encrypted connection.

• Servers. We run our servers, website and backups ourselves, on our own
  hardware in the Czech Republic. We do not rent them from a hosting or cloud
  provider, so we do not entrust the data stored on them to anyone else.
• Email. Our email server is run by Seznam.cz, a.s. (Czech Republic)
  through its Email Profi service. We use it to send service emails (for example
  address verification and password reset) and the launch announcement, and to
  receive the messages you send to hello@webmod.app. Seznam therefore receives
  sender and recipient addresses, the content and headers of these emails, and
  technical data about their delivery (for example IP addresses and bounce
  notifications). Under its privacy policy, besides delivering and storing the
  messages, it also processes this data to keep its email service secure and
  working, for example to sort incoming and outgoing mail; for these purposes it is
  an independent controller. Details (in Czech) are at
  o-seznam.cz/pravni-informace/ochrana-udaju/email.

We do not sell personal data and do not share it with advertisers. We use no
third-party analytics, advertising or tracking tools, in the extension or on the
website. We disclose data to public authorities only where the law requires us to.


| Data | Retention |
|---|---|
| Screenshots and page structure (on our servers) | 24 hours, deleted automatically |
| Copies of inputs/outputs held by Anthropic | Set by Anthropic: up to 30 days, or up to 2 years if flagged for a usage-policy violation (§5) |
| Generated modification scripts | Kept with your account until you delete them |
| The one-time link that delivers a script | 10 minutes, or until used once |
| Live progress updates for a running job (in the cache) | 1 hour |
| Short-lived counters and keys (daily cost, running jobs, protection against repeated requests) | 48 hours at most |
| History of our workflow engine (job status and references to stored data; your answers to clarifying questions and any feedback you give when rejecting a result; for service emails also the recipient address and the message. Never the page content, screenshots, your request text, or the generated script — those are loaded from storage by reference) | 3 days after the job finishes |
| The conversation for a modification | Kept with your account until you delete it |
| Account preferences (e.g. clarifying-question level, Enter key behavior) | Kept with your account until you delete it |
| Account record | Until you delete your account |
| Usage records (daily token counts and cost; never message text) | Kept with your account until you delete it |
| Model-call traces in our observability tooling (model used, token counts, cost, session and account identifiers; never prompt or response text) | Kept with your account until you delete it |
| Export bundle with your data, and the link to download it | 24 hours, then deleted |
| Record of an address that permanently rejects our mail (address, failure class, count) | Deleted together with the account the address belongs to |
| Record of a completed deletion (pseudonymised, see §7) | 3 years from the deletion |
| Website waitlist entry (email, language, join time) | Until launch notices are sent, or until you ask us to remove it |
| Sign-in tokens (hashed) | Until they expire (30 days) or are cancelled |
| Email confirmation / reset tokens (hashed) | 24 hours / 1 hour, or until used |
| Application security and error logs (no IP addresses) | 30 days |
| Web server logs (IP address, time and requested address) | 30 days |
| Communications with us (questions, complaints, claims, requests under §7) | Until resolved, then 3 years (the general limitation period) |
| Encrypted backups of our systems (never screenshots, page structure or export bundles) | 30 days, then overwritten |

The 24-hour limit on page snapshots on our servers is enforced by the storage
system itself, not by a cleanup job that could fail to run. Screenshots, page
structure and export bundles are never included in our backups, so the limit
applies to them in full; the only copy it does not reach is the separate one
Anthropic holds under its own retention policy (§5). Our other data is covered by
ordinary infrastructure backups, which are overwritten within 30 days. Backups
exist only to restore the service after an outage or error; a deletion request (§7)
does not rewrite them, and they are not restored to undo it. But if we ever had to
restore a system from a backup, we would delete again any data whose deletion you
had requested.


If you are in the European Economic Area, the GDPR gives you rights over your data.
We apply these to everyone who uses the service, wherever they are. (If you are in
the United States, see also §12.)

• Access and portability. Request an export and receive a machine-readable
  copy of your account, your modification sessions and conversations, your page
  snapshot records, your generated scripts, and your usage history. Because an
  export contains everything we hold, it requires you to re-enter your password,
  is limited to one per day, and is delivered through an emailed link that works
  exactly once and expires after 24 hours. Your password itself is never included
  — we only ever hold an unreadable hash of it.
• Deletion. Request deletion and your data is erased from every live system
  that holds it: the database, the cache, file storage, the workflow engine, and our
  self-hosted AI observability tooling. Stored screenshots and page snapshots are
  erased immediately rather than left to expire. There are no soft deletes —
  records are removed, not hidden. Deletion requires your password and a typed
  confirmation, and cannot be undone.

Deletion begins the moment you confirm it. Most stores — the database, the
cache, and file storage — are cleared at once. Two are cleared a little behind
that: the workflow engine deletes its history asynchronously, which completes
within about a minute, and our observability tooling is cleared separately and,
if it is briefly unreachable, retried until the deletion completes, so that a
transient outage there cannot stop the rest of the erasure. All of it is erasure,
not hiding.

Two things deletion does not instantly reach. Routine encrypted backups of
our systems are not rewritten on request; they are not restored in the ordinary
course and are overwritten within 30 days. If we ever had to restore a
system from a backup, we would delete again the data you asked us to erase. And
any copy already sent to Anthropic is subject to Anthropic's retention (§5),
not ours — we do not separately hold that copy to delete. Everything within our
own live control is erased as described.

We keep one thing afterwards: a record that a deletion happened, containing a
one-way reference that cannot be turned back into your account, and counts of
what was removed. It does not identify you on its own, but because it is
pseudonymised data we treat it as personal data. It exists only so we can
demonstrate that the erasure took place, and we keep it for 3 years.
- Correction of inaccurate data and restriction of processing.
- Withdraw consent to the website waitlist at any time (see §4a and §8), with
no effect on anything else.
- Complaint to a data protection authority. In the Czech Republic this is the
Office for Personal Data Protection (Úřad pro ochranu osobních údajů, ÚOOÚ),
Pplk. Sochora 27, 170 00 Praha 7, uoou.gov.cz; elsewhere in the EEA you may
complain to the authority where you live or work, or where the alleged breach took
place.


Where we process your data on the basis of legitimate interests (§8), you have
the right to object at any time, on grounds relating to your particular
situation. We will then stop, unless we demonstrate compelling legitimate grounds
that override your interests, rights and freedoms, or we need the processing to
establish, exercise or defend legal claims. Send your objection to hello@webmod.app.


You can start an export or a deletion yourself, as described above. You can also
exercise any of your rights by email at hello@webmod.app or by phone at
+420 735 635 191. We reply without undue delay and within one month at the latest;
for complex requests we may extend this by two further months, in which case we
will tell you within the first month. So that we do not hand your data to someone
else, we may ask you to confirm that the account is yours, for example by replying
from the account's email address. Handling a request is free of charge.


We do not make decisions about you based solely on automated processing that
produce legal effects concerning you or similarly significantly affect you. The
automated safety review (§11) assesses the generated code, not you, and usage
limits apply equally to all accounts.


| Purpose | Basis |
|---|---|
| Running your account | Contract |
| Analysing a page you submitted and generating a modification | Contract |
| Other people's data that happens to be visible on a page you submit (e.g. names or posts on a social network) | Legitimate interests — providing the service you asked for; we do not seek such data out or use it otherwise, and it is deleted with the snapshot |
| Screening generated code for malicious behaviour | Legitimate interests — user safety |
| Usage records and model-call traces (no text) | Legitimate interests — enforcing usage limits, finding faults and controlling the cost of the service |
| Security logging and abuse prevention | Legitimate interests |
| Service emails (verification, password reset, account and service notices) | Contract |
| Recording addresses that permanently reject our mail | Legitimate interests — not sending mail to dead addresses and protecting the deliverability of our email |
| Keeping sign-in tokens and detecting stolen ones | Legitimate interests — account security |
| Handling questions, complaints and claims | Contract; for claims about defective service also a legal obligation |
| Handling your requests under §7 | Legal obligation |
| Record of a completed deletion | Legitimate interests — demonstrating that we carried out a deletion request |
| Backups | Legitimate interests — restoring the service after an outage or error |
| Website waitlist (storing your email to notify you at launch) | Consent — you give it by submitting the form, and may withdraw it any time by asking us to remove your entry |

We do not rely on consent for the service purposes above, because each is
necessary to provide the service you asked for or to run it securely. The website
waitlist is the one exception: it is optional and consent-based, and withdrawing
consent (asking to be removed) has no effect on anything else.

Do you have to give us your data? You cannot create an account without an email
address and a password, and we cannot build a modification without the page you
submit. No law requires you to give us any data.

If you appear on a page someone else submits. If your data is part of a page
that another user sends us for analysis (for example, your post on a social
network), this policy applies to it in the same way: on our servers it is deleted
with the snapshot, and you can exercise your rights under §7 with us. We do not
notify such people individually, because we do not seek out their data and cannot
identify them.

Special category data. We do not ask for, and have no use for, special category
data (such as health, biometric or political data). But because an analysis captures
a screenshot and the page's structure, such data can be captured incidentally if
it happens to be on the page when you click. We do not seek it out or use it, and
the deletion on our servers (§6) limits how long any incidental copy survives
with us — but we cannot guarantee a page never contains it, which is why we ask you
not to run an analysis on a page displaying information you would not want
processed. Where such data is captured this way, we rely on your explicit
consent (Art. 9(2)(a) GDPR): you give it by deliberately choosing that page and
clicking "Analyze and modify site", and you can withdraw it at any time by deleting
the modification and its data (§7). We process the data only to produce the
modification you asked for, and for no other purpose.


One recipient named in §5 processes data in the United States: Anthropic,
whose model inference runs there.

Our contract for the AI is with Anthropic Ireland, Limited (an EU entity), so
the data goes to Anthropic in the EU; the move to the United States, where the
model inference runs, is Anthropic's onward transfer of that data, which
Anthropic protects under the European Commission's 2021 Standard Contractual
Clauses (Art. 46(2)(c) GDPR) in its data processing addendum, published at
anthropic.com/legal/data-processing-addendum. We will send you a copy of the
Anthropic clauses on request.


Webmod is for adults. You must be at least 18 to use it (see the Terms of
Service), and you confirm this when you sign up; we do not otherwise verify age.
The service is not directed to children, and we do not knowingly collect
personal data from anyone under 18 — in particular, consistent with the US
Children's Online Privacy Protection Act (COPPA), not from children under 13. If we
learn that an account belongs to a minor, we close it and delete its data. If you
believe a minor has created an account, please let us know.


Passwords are hashed with bcrypt over a SHA-256 pre-hash. Traffic is encrypted in
transit. Secrets are never written to logs. Every generated script is screened by
an automated safety review before it reaches you; scripts that touch cookies,
stored data or the network get a stricter review than ones that only change
appearance. Generated scripts are delivered exactly once through a link that
expires after ten minutes and stops working after a single use; a second attempt
on the same link is refused. Apart from that link, a script can be retrieved from
our servers only through your data export (§7), which requires your password.

Automated safety review. Before any script reaches you, it is parsed and
checked for what it is able to do. A script that only changes how a page looks gets
a standard review; one that could read stored data, cookies, or send anything over
the network gets a stricter one, and is rejected unless every such capability is
clearly necessary. The review is designed to allow network access only to the
website the script is for, and to reject a script that would send data to any other
server, including ours. A script that fails review is never delivered. This review
is a safeguard, not a guarantee: it inspects the script, but it cannot be certain
of catching every possible way code could reach another server — for example by
adding an image or script element, changing a link or form target, navigating the
page, or loading a stylesheet. Your browser itself does not restrict where a script
you run may connect, which is one more reason we show you every script in full
before it runs.

Who sees your data on our side. Nobody on our side reads your snapshots,
requests, conversations or generated scripts, except where you explicitly ask us to
(for example, when we help with a problem you report), where it is necessary to
investigate abuse or a security incident, or where the law requires it.

Chrome Web Store. The use of information received from Google APIs will adhere
to the Chrome Web Store User Data Policy, including the Limited Use requirements.


This section applies to users in the United States and supplements the rest of this
policy. We extend the GDPR rights in §7 to everyone, including you.

Categories we collect, and why. We collect, and since Webmod launched have
collected, for the purposes already described above: identifiers (your email address; account
identifiers; IP addresses in web server logs); internet/usage information strictly
as part of an analysis you start (a screenshot and the stripped structure of the
page you chose, your request text and answers); the security records in §6; and
the content of communications you send us. The sources are you and the pages you
send us for analysis. We do not collect government IDs, precise geolocation,
financial-account numbers, or the contents of pages you merely browse. Of the
information the law treats as sensitive, we collect only your account log-in — your
email address together with your password, which we store only as a hash — and we
use it only to run and secure your account. Other sensitive information may appear
incidentally on a page you send us (§8); we do not seek it out or use it beyond
providing the service.

We do not sell or "share" your personal information, and we do not use it for
cross-context behavioural advertising. We disclose it only to the recipients named
in §5, and only so that we can provide the service to you; some of them also use it
for limited purposes of their own, which §5 describes.

Your rights. Whichever state you live in, you may request to know what we
have collected, access/port a copy, correct it, and delete it. These
rights correspond to the rights that the CCPA/CPRA gives California residents; the
right to limit the use of sensitive personal information does not apply, because we
use it only to provide and secure the service. We will not discriminate against you
for exercising them. Exercise them through the export and deletion tools described
in §7, or by contacting us; we respond within the timeframes in §7. An authorised
agent may act for you; we may ask for proof of their authority.

"Do Not Track" (CalOPPA). Some browsers can send a "Do Not Track" (DNT) signal.
There is no common standard for how to respond to one, so we do not respond to DNT
signals — but this does not change what we collect, because the extension does not
track your browsing in the first place (see §1), and we use no third-party
advertising or analytics trackers. We do not allow third parties to collect
information about your online activities over time and across websites when you use
our website or the extension.

Effective date. This policy is effective as of October 27, 2026. §13
describes how we notify changes.


We may update this policy, for example when we add a feature or change a provider.
Material changes will be announced by email before taking effect. Minor edits —
such as fixing a typo or clarifying wording — may be made without a separate
announcement. The version published here is always the current
one; we do not keep an archive of past versions.