Privacy Policy

Understand how MyIP.dog handles connection information, network queries, maps, analytics, browser storage and third-party diagnostic services.

Last updated: 2026-09-29

Scope of this policy

This Privacy Policy describes the public network tools and informational pages at MyIP.dog. The service helps you see the public address used for a connection, inspect public network records, and perform selected diagnostics. The public tools do not require registration, payment, a personal profile, or an uploaded identity document. References below to a browser mean the software you use to open the website; they do not mean that every operation remains entirely on your device.

Different tools have different data flows. A subnet calculation can happen locally, while a DNS query needs a resolver and a remote connectivity test needs a measurement provider. Read the disclosure beside a tool before starting it. Our data sources and methodology explain the providers, measurements, and limitations in more detail. This Privacy Policy should be read with those explanations, especially when you are testing a network belonging to an employer or another person.

Connection information

When you visit a website, the network serving it receives connection information, including a public IP address. MyIP.dog uses Cloudflare to deliver the website. On a supported Cloudflare request, the homepage uses the address observed at the edge and associated network metadata to display your result in the initial page response. Available metadata can include an estimated country, region, city, time zone, autonomous system number, network organization, and approximate coordinates.

An IP address may identify a shared gateway, VPN exit, proxy, mobile carrier, or other intermediary. It is not a verified identity or an exact home address. We do not request browser GPS permission to draw the homepage map. For public IPs, the homepage automatically asks IPWhois for approximate location and network details. The queried IP is sent from our server to that provider. Successful lookups are cached in bounded server memory for up to ten minutes, without a database history; unavailable fields stay unmeasured. The homepage automatically makes separate IPv4-only and IPv6-only browser requests to ipify, which receives the connection address for each request. If the application cannot obtain a usable server observation, the browser may also use ipify to identify the connection used for the map and network details. These address requests omit credentials and referrer information. Results remain in the current tab without an account-based address history.

Personalized application responses use headers instructing browsers and intermediary caches not to store them. This helps avoid serving one visitor's result to someone else. It does not remove records maintained independently by your network provider, the hosting platform, a third-party service, or software on your device. A response header is not a promise that no infrastructure logs exist.

Information entered into tools

DNS, reverse DNS, hostname, IP registration, domain registration, and ASN tools process the address or name you submit. The application validates the input and sends the relevant query to the stated source. Depending on the tool, these sources include Google Public DNS, IANA bootstrap directories, the responsible RDAP registry, and RIPEstat. The query target is needed to obtain the result. Do not enter passwords, private access tokens, confidential internal hostnames, or personal information unrelated to the diagnostic.

Some results come from public registration data and can contain an organization's name or public contact information. Their appearance in a result does not establish that the information belongs to you, that it is current, or that you have permission to republish it. Missing registration fields do not establish that a person is anonymous. Corrections to source records generally need to be requested from the registry or data provider responsible for the record.

The subnet calculator and CIDR calculator perform their address arithmetic within the browser. They do not need to contact addresses in the calculated ranges. Their inputs and calculated results are not submitted to a calculation server by those tools. Copy buttons write only the result you choose to your system clipboard. Other applications, browser extensions, and clipboard synchronization software may be able to read clipboard contents according to their own permissions and settings.

The email header analyzer parses pasted header text in the current tab. It does not send those headers to an analysis endpoint, perform automatic address lookups, or save the input in browser storage. Editing clears the previous result, and Clear or leaving the page discards the working state. The optional copied report contains original header fields and may include names, email addresses and message identifiers; review it before sharing. Message-body content after the first blank line is excluded from the report.

The password and PIN generators use the browser's cryptographic random source after you request generation. Generated values stay in the current page's memory and are not uploaded, put in URLs, attached to analytics events or written to local/session storage. Editing settings, Clear and leaving the page discard the results; hiding the tab conceals visible values. An explicit Copy action places the original value on your system clipboard, even if it is hidden on screen. Clearing the page does not erase clipboard history or copies saved in other applications. We do not offer a password vault or recovery history for generated values.

The random IP address generator creates address strings in the browser using a built-in range policy and browser cryptography. Optional CIDR filters, settings and generated batches are not sent to a generation service, analytics event or destination probe, and are not written to URLs or browser storage. Changing generation settings, Clear or navigation discards the working output. Explicit copying transfers it to your system clipboard, where separate retention rules apply. Ordinary page delivery and analytics remain as described above.

Email breach query preview

The data breach check remains a preview until its subscription is configured and verified. It requires a deliberate submission, and shows the configured query mode before accepting an address. Full-address mode sends your normalized email through our service to Have I Been Pwned. Hash-range mode keeps the complete email and SHA-1 digest in the browser, sending only the first six hexadecimal characters through our service to that provider. The browser matches returned suffixes locally and discards unrelated entries immediately. A fixed public catalog request supplies incident details without uploading the matched incident names. Hash prefixes reduce disclosure but do not make requests anonymous.

This tool does not save emails, hash ranges or matches in application logs, databases, URLs, analytics events or browser storage. Working results remain in tab memory. Editing clears results; Clear, hiding the tab or leaving the page removes the input and results. Cancellation cannot retract requests already sent. Ordinary hosting and provider processing still apply. The server caches only public incident metadata for up to one hour. Shared abuse counters contain short-lived pseudonymous connection identifiers and a constant subscription cooldown key, not email addresses or queried prefixes. HIBP's privacy policy governs its processing. No password or mailbox access is requested.

Active network diagnostics

Some tests make additional network requests when you start them. IPv4 and IPv6 checks contact address-family endpoints. The WebRTC test asks the browser to gather connection candidates using the disclosed STUN service. The speed test embeds OpenSpeedTest: loading its panel connects your browser to the provider and supporting services, exposing your public address and ordinary request information. Pressing Start transfers measurement data to and from its servers and consumes bandwidth. OpenSpeedTest displays and processes results under its own policies; we do not import them into an account-based diagnostic archive or promise that the provider retains no records.

Ping, traceroute and HTTP response header checks use Globalping measurement probes. HTTP checks also disclose the URL path and query, selected addresses, request settings and returned headers through public measurements. Do not submit secret links, tokens or personal information; clearing our page does not delete an upstream measurement. A submitted destination and selected test settings are sent to that service. Measurement identifiers and source links are shown with results. Treat remote measurement destinations and shared result links as information disclosed to the measurement provider; do not use them for confidential targets. A port check connects from the service runtime to a validated public destination. The destination can observe that connection even when the test sends no application payload.

The IP comparison feature makes direct browser requests to the destinations named in its results. Each destination observes its own connection. This Privacy Policy does not promise that different destinations will see the same address. VPN routing, split tunneling, protocol selection, and provider availability can change the observations. Unavailable or preview diagnostics are labeled; a missing result does not establish that a connection is private or secure.

Maps and external resources

The homepage downloads map resources from OpenFreeMap and renders them using MapLibre. The map provider receives requests for the resources needed to draw the visible area. Requested map tiles can correspond to an approximate area inferred from your network location. These are third-party requests even though the map appears within our page. The visible attribution identifies the map data contributors and resource provider.

Map failure does not prevent the primary IP result from being displayed. You can also block external resource requests through browser controls or extensions, although doing so may disable maps or particular diagnostics. Following an external source link leaves MyIP.dog and makes a new request to the linked website. External providers have their own privacy practices and retention policies.

Analytics, preferences and storage

The website loads Google Analytics to understand page usage and improve the service. Analytics requests can include the page URL, page title, referrer information, browser and device information, and identifiers managed by the analytics service. Network requests to Google also disclose connection information to Google. The application does not deliberately attach tool inputs or diagnostic result objects as custom analytics events. Avoid placing confidential information in a page URL or sharing a URL that contains it.

The browser stores your chosen color theme so that the interface can remember it. The application may also set a referral-source cookie when a URL includes a supported campaign source. That cookie lasts up to 30 days. Google Analytics may use its own cookies. You can manage cookies, clear site storage, or block analytics through your browser and related privacy tools. Clearing local storage resets preferences; blocking analytics does not prevent local subnet or CIDR calculations.

Hiding an IP or result changes its visibility on the screen. It does not alter your network address, erase an upstream request, or create a VPN. Similarly, clearing a tool's interface removes its displayed state but does not instruct independent providers to delete information they already received. Keep those distinctions in mind when recording a screen or sharing a diagnostic with support staff.

Local browser signal inspection

The browser fingerprint test collects selected browser, screen and preference values only when you choose Inspect browser. Optional graphics collection draws a fixed Canvas sample and reads WebGL properties. The resulting values and Canvas digest stay in the tab; the tool does not send them to a fingerprinting provider, include them in analytics events, write them to browser storage or create a persistent visitor identifier. Clear, hiding the tab or leaving the tool discards its working snapshots. Copying a report deliberately places those details on your system clipboard, whose retention is outside this page's control. Ordinary page delivery and site analytics remain separate from collecting a report and can independently receive browser and device metadata as described above.

Selected checks and combined reports

The privacy scan runs only the checks you choose. Site address, external address-family and WebRTC checks observe this browser's connections. Reverse DNS and mail blacklist checks send the separately entered target through the disclosed resolver and list providers. Six fixed TCP connection attempts require explicit permission confirmation; destinations can observe those attempts. Choosing Use connection IP makes a fresh request to this site and fills the input without authorizing a TCP check. Local browser signals and optional graphics follow the local collection behavior described above.

The combined report stays in tab memory. Stop preserves returned observations and cancels pending work. Clear, editing settings, hiding the tab or leaving the page discards the working report. Neither action deletes requests already received by other providers. Default copying includes coverage, states, times and counts while omitting addresses, PTR names and browser values. A separate detailed-report choice includes those values on your clipboard; review them before sharing. The tool does not put its target or report in URLs, browser storage or custom analytics events.

Retention, security and requests

The public application does not create an account-based archive of your diagnostic history. Temporary request-limit counters and in-progress state support operation and abuse prevention. TCP and Globalping diagnostics also use shared counters stored in Cloudflare Durable Objects. The coordinator processes a trusted connection address and, for TCP cooldowns, the destination address and port. It stores keyed hashes instead of those raw addresses, together with counts and expiry times. A shared exit address can mean several people use the same allowance. These hashes are pseudonymous operational identifiers, not a promise of anonymity. Counters expire within an hour and are removed on later requests or periodic cleanup, normally within five further minutes; platform failures can delay cleanup, and platform backups have their own retention. Clearing the page does not reset a server-side allowance. Hosting, analytics, DNS, mapping, registry, and measurement providers may retain their own operational records under their policies. We do not claim a single retention period covering those independent services. If a particular provider's retention is important to your use, review its policy before starting the associated operation.

We use HTTPS and validate supported inputs. We restrict private and special-use destinations where a tool initiates a remote request, and apply request limits to reduce abuse. These measures reduce specific risks; they do not make a browser session, third-party result, or network configuration immune to failures. Do not send secrets or rely on the website as a storage system for sensitive information.

Privacy rights and available requests depend on the applicable law and the data involved. A request concerning a public registry record or a third-party measurement may need to be directed to that source. For information about this website and its current scope, see About. The operator's direct contact details are being finalized for this launch. The date above identifies this version of the Privacy Policy; material changes will be reflected in the published policy and its update date.

Password strength analysis

The password strength checker loads its model from our static assets only after you request a check. Your candidate is processed in a dedicated browser worker, without sending it or a hash prefix to a checking service. We do not put candidates or results in URLs, analytics events or local/session storage. Stop, Clear, hiding the tab or leaving the page terminates active analysis and clears the working input and result; this is not a promise of forensic memory erasure. An explicit copied summary contains aggregate scores, length and pattern categories, with no password or matched fragments. Clipboard retention, extensions and browser-managed data remain outside this tool's control. Page delivery and ordinary site analytics remain separate as described above.

Telephone number analysis

Phone number lookup processes a single number locally after an explicit request. Its numbering rules load from our static assets; the entered number is not sent to a lookup API, carrier, analytics event or call/SMS service. Inputs and results are not placed in URLs or local/session storage. Editing discards the previous result and cancels pending analysis. Clear, hiding the tab or leaving the tool also removes the working input. Any pending module completion is ignored. A copied base number or report intentionally includes the normalized number; a report can also include an extension. Clipboard history, browser-managed input data and other applications are outside this tool’s control. Ordinary page delivery and site analytics remain separate as described above.

IP reputation catalogs

The IP reputation check sends the entered public IP to our service for comparison with cached CINS Army and Emerging Threats Compromised IPs files. The publishers receive shared file downloads without visitor lookup addresses. A separate Cloudflare Durable Object catalog stores those public address lists, file timestamps, content fingerprints and refresh state, not a visitor lookup history. Results stay in the current page, with no URL parameters or saved browser history of queries. Copying a report includes the checked address. Hosting infrastructure processes requests as described above.

Proxy checker catalogs

The proxy checker sends the entered public IP to our service to compare it with public source lists. We download catalogs from GitHub-hosted X4B and monosans repositories and from Tor Onionoo; these fixed catalog requests do not include the entered IP. Source providers receive the downloading server request under their own policies. Cloudflare Durable Objects retain public network ranges and endpoint/exit observations, source timestamps and refresh state, but not a visitor query history. Upstream proxy credentials, ports and GeoLite2 fields are not retained. The report remains in the current page until cleared, changed or left. Copying explicitly includes the checked IP and source evidence.