Privacy Policy

This privacy policy explains how the Image Tools Kit website (referred to in this document as “the site”, “we”, or “the operator”) collects, uses, stores, and protects information when you use our free, browser-based image tools: compress, convert, resize, crop, metadata removal, color analysis, comparison, and accessibility checks. This policy applies to everyone who visits the site and uses any tool it offers, regardless of whether they access the site from a desktop browser, a tablet, a mobile phone, or any other device.

We want to be direct about one thing at the very start: our service is designed so that, for the vast majority of its features, your images never leave your device at all. The architecture of the site treats your privacy as a default condition rather than an optional setting. We do not monetize your files, we do not analyze your pictures, and we have no business interest in seeing the content you process. The purpose of this policy is to document, in specific and honest detail, exactly what happens to your data at every stage of your interaction with the site, including what we deliberately refuse to collect.

Please read this policy carefully. It is written in plain language but it is also a legal document, and by using any of our tools you agree to its terms. If you do not agree with any part of this policy, please do not use the site. If you have questions that this document does not answer, or if you believe we have failed to honor any commitment made here, you can contact us at imagetoolskit@gmail.com and we will respond within a reasonable time.

The privacy-by-default principle

The design of Image Tools Kit is organized around a single principle: privacy by default. This principle has four parts in practical terms.

Local first: the code that performs almost every image operation runs inside your web browser, on your device, using computing power from your own hardware. The default technical path for nearly every tool is that your file is read into your browser’s memory, processed by the browser’s Canvas API and related image primitives, and then offered back to you as a download. No upload occurs, no server round-trip occurs, and no copy of your file is ever created on our infrastructure.
Collection is minimal: we do not ask you to create an account, we do not require an email address, and we do not ask for your name, your location, or any other personal details to use any tool. The only data points we encounter are those required to operate a well-run website: page-load events, an aggregated picture of browser types, and basic security infrastructure logs.
Collection is aggregated: where we do measure website activity, we use an analytics approach that is aggregated and privacy-respecting. We deliberately avoid identifiers such as persistent cookies, fingerprinting scripts, or personal identifiers that would let us associate any specific person with any specific tool usage.
Data minimization is continuous: anything that is collected at all has a defined, short purpose and a defined lifetime. Temporary uploads are deleted immediately after processing. Logs expire on a fixed schedule. Aggregate analytics are retained only for as long as they remain useful for basic maintenance and improvement decisions.

We believe that privacy should be the quiet default of a web service, not a premium feature or a checkbox the user must find. Every decision described in this policy flows from that belief, and the sections below document the concrete consequences of that decision for the data life cycle.

Information we do not collect

The clearest way to describe our data practices is to start with what we do not collect, because for most tools that list is nearly everything.

Images and visual content

For the client-side tools described below, we do not see, receive, transmit, or store your images. The bytes of any file you load into a client-side tool are read by your browser and processed locally. We cannot view your images, we cannot recover them, and we have no technical mechanism to do so because no copy is ever sent across the network. This includes photographs, documents, screenshots, graphics, GIFs, and any other visual content you choose to process.

File metadata

We do not collect the metadata embedded in your files. That includes EXIF data, GPS coordinates, camera make and model, timestamps, lens information, author names, ICC color profiles, or any other metadata fields that image files frequently carry. In fact, one of our tools exists specifically to help you strip such metadata from your images. Because the metadata removal tool runs entirely in your browser, the metadata it removes (for example, your location or the serial number of your camera) is removed from your copy only, on your device, and is never transmitted anywhere. When we say we do not collect file metadata, we mean it literally: the metadata never appears on any server we control.

File names and paths

When you select a file through the file picker or drag and drop it onto the page, your browser tells our JavaScript code the name of the file so the tool can display it in the interface. That name exists only in your browser’s memory and in the local rendering of the page. It is not sent to us. We do not log file names, we do not store them, and we have no way to associate a file name with a visitor. Filenames can be revealing in unintended ways, and we deliberately treat them as if they do not exist on our side.

Clipboard and paste data

Some tools allow you to paste an image directly from your clipboard. That pasted content lives entirely in your browser’s memory for the duration of the session. We do not read, log, or retain clipboard contents, and the clipboard itself is managed by your operating system and browser under their own privacy rules, not ours.

Personal identifiers

We do not collect your name, email address (unless you deliberately email us), postal address, phone number, social media profiles, or any other identifier that could be used to contact you or identify you as a natural person in connection with tool usage. There is no account system, no sign-up form, and no marketing database tied to the tools.

Client-side processing explained

Client-side processing is the technical foundation of the privacy promises in this policy, so it is worth describing plainly. When you load an image into a client-side tool, the following happens inside your browser:

Your browser reads the file from your local disk into memory using the File API. At this stage the file still exists only on your device.
The image is decoded and drawn onto an off-screen canvas element using the standard browser Canvas API. The decoding is performed by your browser’s built-in image decoders; the resulting pixels exist only in the memory that your browser has allocated on your device.
The tool performs its operation on those pixels or on that canvas, whether that means re-encoding at a lower quality level, converting to another format, scaling the dimensions, cropping to a region, or analyzing color values.
The resulting image is exported from the canvas as a new blob or data URL and, in most cases, presented to you as a downloadable file. When the operation is complete, the temporary objects in memory become eligible for garbage collection by the browser.

At no point in this pipeline is a network request made to our servers carrying the image bytes. The JavaScript that runs the tool issues no request whose payload includes your file. If you watch the network panel in your browser’s developer tools while using a client-side tool, you will see requests for the page and its static assets, but you will not see your image being transmitted. This is not a privacy setting you must activate; it is the only way the tool’s code operates.

Two practical consequences follow. First, the tools keep working even if your internet connection drops after the page has loaded, because processing does not depend on a server. Second, the processing capacity available to you is that of your own device, so very large images will behave as they would in any desktop application: they may use significant memory, and the browser may take longer to decode them. This trade-off is the deliberate price of a genuinely private service, and we believe it is worth paying.

Temporary-upload tools

A small number of tools cannot run entirely in the browser because of technical limitations of the Canvas API or of browser memory limits for certain operations. These tools are clearly and prominently labeled “temporary upload” in the interface, and they operate under the most restrictive data handling practices we can design. This section describes those practices in full.

Which tools are in this category

If a tool displays the “temporary upload” label, it belongs to this category. You will always know before any upload begins, because the label appears next to the tool’s name and again on the upload screen before you select a file. We do not silently route any tool to a server. If a tool is not labeled “temporary upload”, then it is client-side and your file never leaves your device.

What happens to an uploaded file

When you use a “temporary upload” tool, the file you select is transmitted over an encrypted HTTPS connection to our temporary upload endpoint. Once the transmission is complete, the following protections apply by design:

Random naming: the uploaded file is immediately assigned a freshly generated random name. It is never stored under its original file name, and the random name has no relationship to you, to the original filename, or to any identifier associated with your browser session.
Strict size cap: uploads are hard-capped at 25 megabytes. Files larger than the cap are rejected by the server before any processing begins, even if your browser sent the entire payload. We accept no file larger than this limit under any circumstances.
MIME magic-byte validation: the server does not trust the file extension or the declared content type. It inspects the leading bytes of the file (the magic bytes) to verify that the content actually matches an allowed image type. Files that do not match are rejected and discarded immediately. This check is a security control that keeps unexpected content out of the processing pipeline.
Single use: an uploaded file is used for exactly one processing job. It is never reused, never served back to any other user, and never held for later reference, training, quality review, or any other secondary purpose.
No public exposure: uploaded files are never served back to the public internet. The file is only ever readable by the processing pipeline, and only for the moments required to perform your job. No URL that yields the uploaded content is ever made available to anyone, including the person who uploaded it.
Programmatic deletion: immediately after processing completes, the file is deleted by an automated routine. Deletion is unconditional and does not depend on a storage quota, a cleanup schedule, or any human action. The deletion step runs as a mandatory part of the processing pipeline, not as an optional housekeeping task.

Storage duration

Strictly speaking, the storage duration for a temporary upload is measured in seconds. The file rests on a short-lived, ephemeral media volume while the processing job runs, and it is then deleted programmatically. In the event of a processing failure that interrupts the pipeline, a defensive sweep deletes any orphaned temporary files on a fixed schedule measured in minutes, not days.

The result of your processing job (for example, the converted or compressed output image) is generated after the original file is deleted. That output follows the same rule: it is transmitted back to your browser once and is not retained on any server afterward. The temporary endpoint has no storage tier that accumulates user content over time.

Because uploads of this type exist on a server for a short time, and because even that brief existence is unavoidable for the tool to function, we want to be unambiguous: if your image is sensitive, confidential, or legally protected (such as medical records, identity documents, trade secrets, or images of other people), do not use a “temporary upload” tool. Use the equivalent client-side tool instead, or process the image on your own machine. The client-side options give you complete local control, and for sensitive material that is always the appropriate choice.

No reuse and no training

We want to rule something out in writing: an uploaded image is never used for training machine-learning models, never added to any dataset, never reviewed manually, never published, and never combined with images from other users. There is no review queue, no moderation pipeline that surfaces user content to staff, and no analytics event that records what your image contained. The only meaningful data we could derive from a temporary upload, namely its size, its type, and the tool selected, is reported only in anonymous, aggregated form and is never tied to the content of the file.

Data we do collect

We are not a data-free website; no website is. This section lists the narrow set of data points we do encounter, honestly and completely.

Page-view events: aggregate counts of when pages on the site are viewed, including which tool pages are visited. These events contain no personal identifiers.
Coarse usage aggregates: non-identifying summaries such as “the compress tool was used approximately X times today” or “visitors from browser type Y completed Z conversions”. These are rolled-up numbers, not records about individuals.
Site health data: technical signals such as response times and error rates that let us detect when the site is slow or broken. These signals are not tied to any particular user’s identity.
Basic security infrastructure logs: described in the dedicated section below.

Importantly, the analytics data we collect is never joined with any identifier that could point back to a specific person, and it is never joined with any information about the content of the images you process. There is no cross-referencing, no enrichment, no demographic append, and no selling or sharing of any kind.

Analytics and cookies

Aggregate-only analytics

The site uses a privacy-respecting, aggregate-only analytics approach. The analytics tool does not plant persistent tracking cookies, does not perform browser fingerprinting, and does not build a profile of your behavior over time. What we see is the shape of aggregate traffic: roughly how many visits a page received over a period, which regions those visits broadly came from at a coarse level if the browser discloses that information, and which general category of browser and device was used. The analytics reports answer questions like “did the new user interface change reduce abandoned sessions?” They cannot answer the question “what did an individual visitor do on the site?”

Cookies and local storage

We do not use cookies for advertising, tracking, personalization, or cross-site identification. The site may use functional, non-tracking mechanisms required for ordinary operation, such as the minimal internal state needed to keep the page responsive. Where your browser stores any small piece of state on our behalf (for example, to remember nothing more than a user interface preference), that state is limited, local, and not readable by third parties. To the extent the analytics tool sets a non-essential cookie, you can refuse or delete cookies through your browser at any time, and the site will continue to work normally.

Because the analytics service we use is designed to avoid persistent identification, turning it off changes very little: we would simply see fewer aggregate events. There is no targeting behavior on this site that you need to escape, and no hidden tracking layer that changes the meaning of this paragraph. If you would like to opt out of the aggregate analytics events, you can disable JavaScript in your browser or install a content blocker; either approach is entirely compatible with using the site.

Logs and security monitoring

Like every publicly reachable website, the infrastructure behind Image Tools Kit keeps technical logs. These logs exist to detect and respond to attacks such as brute-force attempts, malformed request floods, and automated abuse. A log entry will typically contain the IP address of the requester, the requested URL, a time stamp, user-agent information, and the HTTP status code returned.

We take several measures to keep these logs privacy-respecting. Logs are used only for security analysis and operational diagnosis. Raw logs are retained for a limited period and are not combined with analytics data, not combined with any tool-usage data, and not used to track the browsing habits of any individual. We do not intentionally map logs to real-world identities, and we do not share logs with any third parties except as required by law.

If you believe you have found a security vulnerability in the site, or if you observe behavior that suggests one, we encourage you to report it responsibly to imagetoolskit@gmail.com. Please do not test intrusive methods against the live site that could disrupt service for other users; contact us first, and we will coordinate a safe and private way to assess your finding, including making arrangements for you to test in a controlled environment if appropriate. We treat security reports seriously and will acknowledge valid reports.

Third parties and processors

We keep our processing chain deliberately short, and we are explicit about it.

No image-processing processor: no third-party service performs, hosts, inspects, or touches any image you process with a client-side tool, because those images never travel over the network. For “temporary upload” tools, the images are handled only by the operator’s own endpoint as described above; we do not forward uploaded files to any third-party processing service, API, or cloud function.
Hosting provider: the site’s static assets and the temporary endpoint are hosted by a commercial infrastructure provider. That provider operates the servers and storage on our behalf. It has no authorization to use or disclose user image content, and its incidental technical access is limited to what is required to run the ephemeral upload pipeline.
Analytics provider: the aggregate analytics service receives only the aggregate, non-identifying events described in this policy. It acts as a data processor on our instructions, and we maintain a data-processing agreement with it as required by applicable law.
No advertising networks: the site does not display third-party advertising and does not integrate advertising trackers, remarketing pixels, social widgets, or any other third-party script whose purpose is to profile visitors.

Because the set of processors is small and their roles are narrow, we can make a simple statement: no third party in our processing chain ever sees the content of an image you process with a client-side tool, and no third party retains any copy of a temporary upload after the processing job completes. There is no data brokering, no licensing of your content, and no hidden vendor that quietly receives the files you handle.

Your rights and choices

Because of the architecture described above, we hold essentially no personal data about you. Nevertheless, we describe your rights here, and we take them seriously for the small amount of information that does implicate them.

Access

You have the right to request a copy of the personal data about you that we hold. In practice, the answer will almost always be that we hold none. There is no user record, no profile, and no account that could serve as a container for personal data about you. You can make an access request at imagetoolskit@gmail.com, and we will confirm, in writing, whether any personal data about you is being processed and what it is.

Deletion

You have the right to request deletion of personal data about you. Since we do not create accounts, maintain user profiles, or link analytics to individuals, there is normally nothing of yours for us to delete. If you have emailed us, we will delete the emails and any associated records you ask us to remove, subject to legal obligations that may require retention. You may also, at any time and without contacting us, exercise the practical equivalent of deletion yourself: clear your browser cache, delete your downloads, and stop using the tools. Since client-side processing holds your files only in browser memory until you download them, deleting your copy ends the entire data life cycle with respect to you.

Correction

You have the right to have inaccurate personal data corrected. Because we do not hold personal data as a matter of design, there is almost nothing to correct. If we hold any information about you, for example in records of correspondence, we will correct it promptly on request.

Data portability in context

Data portability under the law generally means the right to receive personal data you have provided to a controller in a structured, machine-readable format and to transmit it elsewhere. In the context of this service, the personal data you provide is effectively nil, and the files you process are always in your own possession. You already have full portability by design: the images you make with our tools are yours, stored on your device in standard image formats, and free to move anywhere. Nothing on our side is structured to receive a portability request, because there is no dataset about you to extract.

Complaints

You have the right to lodge a complaint with a supervisory authority responsible for data protection in your jurisdiction if you believe we have not honored your rights. We hope it never comes to that, and we will do our best to resolve any concern directly first. You can raise concerns at imagetoolskit@gmail.com, and we will respond substantively.

We will never condition your ability to use the site on a waiver of any of these rights, and we will never charge a fee for exercising them except in the narrow circumstances where the law permits a reasonable charge, such as for manifestly unfounded or excessive requests. We will respond to valid requests without undue delay and, in any event, within the time limits set by applicable law.

Children’s privacy

The service is intended for use by adults and young people who are old enough to make informed decisions about the material they process. The site is not directed to children, and we do not knowingly collect personal data from children. The architecture of the service makes this a moot point in practice: there is no registration, no email harvesting, no social features, and no mechanism through which a child could supply us with personal data that we would then hold.

If you believe that a child has provided personal data to us through a contact channel, please tell us at imagetoolskit@gmail.com and we will delete the information as quickly as we can. Because we hold almost no structured personal data, the resolution is normally immediate: any correspondence is removed, and there is nothing else for us to purge.

International data transfers

The amount of data this site transfers across borders is close to zero for the features that matter, because client-side processing sends no image data anywhere. The small volume of data that does exist, mostly aggregate analytics events and technical logs, may be processed in infrastructure facilities that could be located in more than one country. We ensure that any such processing is covered by appropriate transfer safeguards, including standard contractual clauses where required, and that the processing remains subject to the commitments in this policy regardless of the jurisdiction in which the servers physically sit.

We have deliberately designed the service so that no image ever needs to cross a border to be processed in the client-side case, and a temporary upload travels no further than the operator’s own endpoint before being deleted. As a result, international data transfer is one more area where our answer is mostly “nothing moves”.

Data retention

Retention is where a privacy policy often becomes vague, so we will be precise. Different categories of data have different, fixed lifetimes.

Temporary uploads: deleted programmatically immediately after the processing job completes, with a defensive orphan sweep running on a schedule measured in minutes for any files that remain after an interrupted job. No uploaded file is ever moved to a long-term storage tier.
Client-side files: no retention by us of any kind. The file exists in your browser’s memory only until you leave the page or the browser reclaims the memory; the download, if you make one, lives on your device under your control.
Aggregate analytics: retained in rolled-up form for as long as they are useful for basic maintenance and improvement decisions. Individual event records are not kept, and the aggregate summaries are not tied to individuals even while they exist.
Security logs: raw logs are retained for a limited period (weeks, not years) and then automatically purged. Longer retention is applied only where a specific, documented security incident requires it.
Correspondence: email you send us is retained as long as needed to respond and resolve any follow-up, then deleted on request or when it is no longer relevant.