Your data got leaked.
Find it. Remove it.
Free scan against real breach databases, then a guided removal centre for the data brokers and people-search sites that sell your name, phone and address.
Action Center
Full protection at a glance: your exposure, your removals, and the one most useful thing to do next — LeakGuard works out which, so you never operate a dashboard to stay protected.
Recent activity
Account
FULL PROTECTION — SAVED DETAILS, REMOVAL, MONITORING
Welcome back
An account is optional — the quick scan always works without one. An account just saves your details encrypted so you never type them twice, and unlocks monitoring, consent controls and a full data export. Free, like everything here.
Create a new password
Choose a new password for your account. Setting it signs out every device signed in to this account.
Signed in as
My household
Protecting family too? Add a person here, then use the little “whose detail” menu on any saved detail below to mark it as theirs. A person here is just a name label for grouping — they get no account, no login, and we store nothing about them beyond the label you type.
My saved details
Stored encrypted — we can only ever show them masked here. These are the details LeakGuard can watch and clean up for you, so you never type them again.
What each kind gets checked against: emails against the breach databases; phone numbers, names and addresses against public web mentions; usernames against public profiles on major platforms plus web mentions. A mention or a matching handle is a lead for you to review — never proof on its own.
My domains
Add a domain you own and LeakGuard can watch its public footprint (DNS records and certificates). Monitoring only starts after you prove the domain is yours: add the TXT record we show you to your domain's DNS, then press Verify. Until then the domain is never looked at.
My permissions (consent)
Nothing here runs unless you switch it on. Switching off is instant and recorded.
Monitoring
With the Monitoring permission on above, LeakGuard re-checks your saved details on the schedule below — using only the sources it already checks (the breach databases and the public web mentions from your scans). It never watches “the whole internet” — no service honestly can. When a check finds something new, or something that was there is no longer found, it lands in your notifications below and — with the Notifications permission on — the check's summary lands in your inbox too.
Monitoring is currently off — switch on the Monitoring permission above and scheduled re-checks will start.
Need a break without switching the permission off? Pause keeps your Monitoring permission exactly as it is — scheduled re-checks simply wait until you resume.
Timeline
Your scans, findings and removals, newest first — built from the actual records, not a separate log.
Notifications
Everything LeakGuard has flagged for you. If email delivery is not set up or a message could not be sent, the entry says so honestly instead of pretending it reached you.
Exposure map
Your saved details, where your latest scan actually found them, and the removal case working on each place — drawn only from real scan findings and real cases, never guessed. A line from a place to a broker exists only when that place is the broker (or sits on the broker's own site). Details are shown masked, exactly as everywhere else here.
Search exposure
What a search engine shows about you: the public web pages your scans found that mention one of your saved details, in one place. These are public search results — a page mentioning your detail is a lead to review, never a confirmation that the page is about you, and LeakGuard never presents it as one. That is why every entry here carries the weakest confidence we assign. Open a result and judge it yourself; if it really is you, the removal tools below are how it gets taken down at the source.
Privacy policy checker
Paste another service's privacy policy — or give the address of its policy page — and LeakGuard runs a fixed keyword checklist over it: selling or sharing, advertising and tracking, retention, your rights, contact details, security practices, children's information, policy changes and sharing with other companies. It is an automated keyword checklist, not a legal verdict: a phrase being present does not prove a company behaves that way, and a phrase being absent does not prove the opposite.
API access
For your own scripts: an API token lets a program read your LeakGuard data
(action center, scans, removal cases, notifications, timeline) without signing in here.
Tokens are read-only — they can never change, submit or delete anything —
and you can revoke one at any time. We store only a hash of each token, so the raw value is
shown exactly once, when you create it. Send it as
Authorization: Bearer lg_…. Treat it like a password.
Your new token — copy it now. You won't see this again; if you lose it, revoke it below and create a fresh one.
Full scan
One tap scans every detail you have saved against the breach databases and scores your whole profile — no re-typing, nothing else to fill in. It needs the Scanning permission above to be on. Details nobody can check yet are reported honestly as “no checks available yet”, never as “safe”.
Removal
One tap opens a removal case with every data broker in the registry, and LeakGuard works through them for you — submitting the forms it can and preparing letters for the rest. It needs the Automatic removal permission above to be on. A case only ever says Removed after LeakGuard has gone back and checked the listing is really gone — “request sent” is not the same thing, and we won't pretend it is. Brokers whose forms demand a CAPTCHA or a sign-in are handed to you below with the exact step to finish — no tool can honestly do those for you.
Two-factor authentication
Status: –. Use any authenticator app (Google Authenticator, Authy, your password manager). Codes are single-use — a code that worked once never works again.
Copy this setup text into your authenticator app (choose “enter a setup key” / scan alternative), then type the 6-digit code it shows to switch two-factor on:
Passkeys
A passkey signs you in with this device's fingerprint, face or PIN — no password to type, and it is phishing-resistant by design — there is no password to phish or leak: LeakGuard stores only the passkey's public half, never anything that could sign in as you. Passkeys are optional — your password (and two-factor, if it is on) keeps working exactly as before. Signing in with a passkey asks your device to confirm it is really you (fingerprint, face or PIN) every time.
Your password is asked once more to add or remove a passkey, so a borrowed session cannot change how you sign in.
This browser or device does not support passkeys, so you cannot add one here. Passkeys you added on other devices are still listed above, and you can remove them from here.
Download my data
Everything LeakGuard holds about you — account, saved details (shown in full, they are yours) and your complete permission history — as one file. Because the file shows your details in full, downloading asks for your password once more.
My exposure & removal report
A readable summary of your exposure score, findings and removal cases, with your details shown masked — made for keeping or sharing. It is generated when you open it and never stored; print it or save it as a PDF from your browser.
Change password
Delete account
This deletes your account, every saved detail and every session. It cannot be undone — download your data first if you want a copy.
Admin
Platform totals for the owner only — counts and system health. No user's details, emails, findings or letters ever appear here; the audit trail records actions and counts, never personal data.
Recent activity (audit)
1 Leak scan
QUICK SCAN — ONE EMAIL ADDRESS, RIGHT NOW
No account, nothing stored: type one email address and see its breaches in seconds — if the free server was asleep, the first check can take about a minute while it wakes. For the full protection — your details saved once, removals run for you and monitoring that keeps watch — create the free account above instead. Either way the honest limits below apply: brokers, people-search sites and Google results can be removed; a breach dump already copied to Telegram, dark-web forums or torrents cannot.
Breaches containing this email
Types of data exposed
Password check
What to do next
Saved details are stored encrypted and make future scans one tap. Nothing is saved unless you press this button.
2 Agent Mode (zero-token, automatic)
Fill in your details once and press the green button — the agent does everything else with plain scripts, no AI, no tokens: it probes all 40 brokers live, submits every removal form it can fill, and prepares the rest. You watch it work. The only steps ever handed back to you are the ones no tool can honestly do for you: pressing Send on the email-broker letters, and any broker hiding behind a CAPTCHA or sign-in — each listed with the exact step to finish.
Your removal plan
Optional preview: Probe visits a broker's opt-out page and reports its real form and blockers. The green button below runs all of this — and submits — automatically.
Probe result
🏠 Blocked by a broker's bot wall? Run it from your own device
A few brokers refuse every request that comes from a hosting network — but the same pages open normally from a home connection. The LeakGuard Local Runner is one small Python file (nothing to install) that runs the same one-click removal run on your own computer: it submits the forms it can fill, writes the email brokers' letters as ready-to-send files, and walks you through only the brokers that demand a human, one page at a time. Your details never leave your machine except inside the removal requests themselves.
Needs Python 3. Run it with: python3 local_agent.py
⚡ Remove my data everywhere
One press. The agent probes all 40 brokers (quick pass, then a deep pass for the stubborn ones), submits every form it can fill, and writes erasure letters for the brokers that only accept email requests. Brokers that hide behind CAPTCHAs or bot walls get listed honestly in the final report — no tool can pass those automatically, and any service claiming otherwise is not telling the truth.
Pressing the button is your confirmation: removal requests will be submitted to the brokers in your name.
Run complete
Email brokers only accept requests from your own email address, so the agent writes each letter and opens it in your mail app — your only job there is pressing Send.
3 Manual tools (advanced)
These companies sell your data. Each one legally has to remove it when you ask — under India's DPDP Act §12, GDPR Art. 17, or CCPA. Work through the list: open the opt-out, use the letter generator if they ask for a written request, and mark your progress. Progress is saved only in your browser.
Show the manual tools — only if you want to do it by hand
Deletion request letter generator
Pick a broker below and tap “Letter” — the letter is generated with that broker's name. Or generate a generic one:
4 Google exposure
Check what a stranger finds when they Google you, and get it removed from search:
“Results about you” is Google's own tool: it finds search results showing your address, phone or email and lets you request their removal from Google Search. Note: that hides the result on Google — removing it at the broker (Agent Mode / Removal Centre) deletes it at the source.
The honest limits
- Can be removed: data brokers, people-search sites, and Google search results — they must answer a legal erasure request.
- Cannot be removed: a breach dump already copied to Telegram, dark-web forums or torrents. No tool on earth can delete every copy. Anyone promising that is lying.
- So the real defence is both: remove what's removable, and change any password that appears in a breach — everywhere you reused it.
Trust & security
Exactly what LeakGuard stores, what it never stores, who else touches your data, and how long anything is kept. Every claim on this page describes the code that is actually running — nothing here is aspirational.
What we store — and how
- Your saved details (emails, phones, names, addresses, usernames, domains) are encrypted with AES-256-GCM envelope encryption: every record gets its own fresh random encryption key, and that key is itself encrypted under a master key that lives only in the server's environment — never in the database, never in the code.
- Your account email is looked up by a keyed one-way fingerprint (HMAC), stored encrypted the same way as your saved details, and only ever displayed masked (like r•••@example.com). It is never stored in plain text.
- Scan results and removal cases are stored against your account so monitoring and the timeline work. The anonymous Quick Scan stores nothing at all — its results are computed per request and forgotten.
What we never store
- Passwords. Account passwords exist only as Argon2id password hashes — we could not read yours back if we wanted to.
- Sign-in sessions and API tokens. Only their SHA-256 hashes are stored, so a database copy alone contains nothing anyone could sign in with.
- Passwords you check for breaches. The check uses k-anonymity: only the first 5 characters of the password's SHA-1 hash are sent to Have I Been Pwned, the match is made locally, and the password itself is never sent, stored or logged.
- Payment details. LeakGuard is free and has no payments — we hold no card or bank information about anyone.
Who else processes data
- Render — hosts the app. Neon — hosts the database (which holds only the encrypted forms described above).
- Brevo — sends account email (password resets, and monitoring alerts if you switch notifications on). It sees the recipient address and the message.
- XposedOrNot and Have I Been Pwned — the breach databases your email is checked against; checking means they receive the address being checked, the same as if you used their own sites.
- DuckDuckGo — public-web discovery searches. Domain checks read public DNS via Cloudflare and public certificate records via crt.sh. Username checks read the public profile pages of the platforms themselves.
There is no advertising, no analytics tracking, and no selling or sharing of data with anyone beyond the processors above.
Where data lives
- The app runs on Render in Oregon, United States.
- The database is hosted by Neon in AWS ap-southeast-1 (Singapore). It stores the encrypted and hashed forms described above.
- Email and scan sources receive only the specific data each request needs, as described in “Who else processes data” and in the full privacy policy. LeakGuard does not choose a different region per user; these are fixed locations for the current free services.
How long things are kept
- Sessions: deleted 7 days after they expire or are revoked.
- Password-reset tokens: deleted 7 days after they are used or expire.
- Notifications: 90 days.
- A deleted account: everything it owns is hard-deleted 30 days after deletion — saved details, scans, findings, removal cases, settings, household and tokens. A count-only audit trail (actions and numbers, never values or personal data) is kept.
The honest limits of removal
- Can be removed: data brokers, people-search sites, and Google search results — they must answer a legal erasure request, and a removal only ever counts as done here after LeakGuard has gone back and verified the listing is gone.
- Cannot be removed: a breach dump already copied to Telegram, dark-web forums or torrents. No tool can delete every copy — anyone promising that is lying.
Your rights, in practice
- See everything: Privacy Center → “Download my data” exports your account details, saved details, scan history and findings, removal cases, notifications, settings and consent history as one file (JSON or CSV). Passkeys and API tokens appear as names and dates only — never the keys themselves.
- Withdraw permission: every consent switch in the Privacy Center takes effect instantly and is recorded.
- Delete everything: Privacy Center → “Delete account”. You are signed out immediately and your data becomes inaccessible; the 30-day window above exists so an accidental or malicious deletion can still be undone by support, then the hard delete runs.
- Extra protection: optional two-factor authentication (authenticator app) and optional passkeys (sign in with your device's fingerprint, face or PIN — LeakGuard stores only the passkey's public half) are available in the Privacy Center.
Responsible disclosure
Security research in scope covers this live website and the public LeakGuard repository. Please act in good faith: test only with accounts and data you own, do not access or change another person's data, do not disrupt the service or run high-volume automated attacks, and give us a reasonable chance to fix a problem before publishing details. If you follow these rules, LeakGuard treats your research as authorized for the purpose of this disclosure policy and will not ask a platform or provider to act against you for the report itself.
LeakGuard does not offer a paid bug bounty or other rewards. Reports are still welcome and credited only if you ask to be named.
Security contact
Found a security problem? Email the project security contact: forapikeyonly2008@gmail.com. Our machine-readable contact file is at /.well-known/security.txt.
Privacy policy
Effective date: 7 October 2026. This policy describes what the LeakGuard code running on this site actually does. It claims no certification or formal compliance status.
Information stored for an account
- Account email: stored as a keyed HMAC lookup value, an AES-256-GCM encrypted value, and a masked display form — not in plain text.
- Saved details: emails, phone numbers, names, addresses, usernames and domains you save are stored the same way: encrypted, with an HMAC lookup value and a masked display form.
- Passwords and sign-in: account passwords are stored only as Argon2id hashes. Session, reset and API tokens are stored only as SHA-256 hashes. Two-factor secrets are encrypted. For a passkey, only its public key is stored.
- Scan and removal records: findings record the provider, source, exposed field names, confidence and an evidence hash — not the raw saved detail. Removal cases record the broker, status and evidence. If a removal letter is generated, it necessarily contains the details you supplied for that letter and is shown only in your account. When you start a removal run under the automated-remediation consent, LeakGuard also sends that letter to the broker for you, as your authorized agent, through the email processor listed below — the letter states that LeakGuard acts as your agent, and the broker's replies are addressed to your own email address, not to LeakGuard.
- Settings, permissions and notifications: your consent history, monitoring choices and in-app notifications are stored so those features work.
What is not stored in logs
Passwords are never stored in plain text. Application and audit records are built not to contain plaintext saved details, passwords, sign-in tokens or vault keys. The audit trail uses actions, identifiers for records inside the system, and counts; it contains no email or saved-detail values.
Anonymous Quick Scan
The Quick Scan needs no account and creates no server-side scan record. The email is processed during that request and the result is returned to your browser; it is not saved as an account scan. If you also check a password, only the first five characters of its SHA-1 hash are sent to Have I Been Pwned and the match is made by LeakGuard. Your browser keeps a small last-scan summary (the email, breach count, score and date) in its own local storage on your device so it can show a change next time; that summary is not a server-side LeakGuard record.
Processors and scan sources
| Service | Role | Location / data involved |
|---|---|---|
| Render | Application hosting | Oregon, United States; runs the LeakGuard service |
| Neon | Database | AWS ap-southeast-1, Singapore; stores the encrypted and hashed account data above |
| Brevo | Email delivery | Receives the recipient email address and message when LeakGuard sends a reset or notification email, or a removal letter it sends to a broker as your authorized agent |
| XposedOrNot | Email breach source | Receives the email address being checked |
| Have I Been Pwned | Password exposure source | Receives only a five-character SHA-1 prefix, never the password |
| DuckDuckGo | Public-web discovery | Receives the public search query used for a scan |
| Cloudflare DoH and crt.sh | Domain information | Receive the domain being checked for public DNS and certificate records |
There is no advertising, analytics tracking, or sale of account data in LeakGuard.
Cookies and browser storage
LeakGuard sets no advertising or analytics cookies and uses nothing that tracks you across other websites. It uses only:
- One sign-in cookie (essential): a first-party cookie named
lg_sessionkeeps you signed in for up to 30 days. It holds only a random session token — the server stores just a digest of it — and is marked HttpOnly, Secure and SameSite=Lax, so page scripts cannot read it and it is only sent over HTTPS. Signing out deletes it. - On-device storage: your browser keeps a few items on your own device only — your removal-progress marks and a short summary of your last Quick Scan — so the page can show your progress when you return. The server never receives them; clearing your browser storage removes them.
- Offline shell: a service worker caches the site's static files (the page, styles and scripts) so LeakGuard loads quickly. It never caches account data or API responses.
Retention, export and deletion
- Saved details stay until you remove them or delete the account.
- Sessions and password-reset tokens are purged seven days after expiry or revocation; notifications are kept for 90 days.
- Deleting an account signs you out and makes the account inaccessible. Its owned data is hard-deleted after 30 days. A privacy-free audit record of actions and counts may remain; it contains no email or saved-detail values.
- In the Privacy Center you can download your data as JSON or CSV after entering your password again. The export includes your account email and saved details in full because they are yours, plus your scan history and findings, removal cases, notifications, settings and consent history; passkeys and API tokens are listed by name and dates only, never the keys themselves. Protect the file.
Contact
For privacy questions or requests, use GitHub Issues. Do not post passwords, tokens, keys, or a full data export in a public issue.
Terms of use
Effective date: 7 October 2026.
Use LeakGuard only for data you may act on
You may use LeakGuard only for your own personal data, or for another person's data when that person has authorized you to act for them. Do not use it to investigate, expose, harass or act on another person's information without authorization.
Your account and devices
Keep your password, authenticator, passkeys and devices secure. You are responsible for actions taken through your signed-in account. Tell us through the support page if you believe your account has been misused.
No guarantee of removal
LeakGuard can submit and track removal requests, prepare letters, and verify a listing again. The broker or site holding a listing decides whether and when to act. A submitted request is not a completed removal; LeakGuard uses “verified removed” only after a later check supports it. Copies of breach data already spread to Telegram, dark-web forums, torrents or private collections cannot be recalled by LeakGuard.
Service availability
LeakGuard is a free service running on free infrastructure. It is provided as-is, may sleep when idle, may be unavailable, and may change. Scan sources are third-party services and may be incomplete, delayed or unreachable. Do not rely on LeakGuard as your only protection: use unique passwords and two-factor authentication on important accounts.
Acceptable use
Do not bypass security controls, CAPTCHAs, login walls or rate limits; do not attack the service; and do not use LeakGuard for unlawful activity. We may suspend activity that threatens the service or another person's data.
Support
Support is provided on a best-effort basis. There is no promised response time.
How to ask for help
Open an issue in the public repository: github.com/xorudra/leakguard/issues. Say which page or feature you used, what you expected, what happened, the approximate time, and the type of device and browser. If an error message showed a request ID, include that request ID — it contains no personal data.
Never send these
Never post a password, API token, session link, API key, vault key, one-time code, or a full data export. Support does not need them and will never ask for them.
Other kinds of help
- Security problem: do not open a public issue. Use the responsible disclosure contact on the Trust & security page.
- Privacy question: read the privacy policy first; it describes the stored data, processors, export and deletion.
- A finding is wrong: open the finding in your Privacy Center and use its “Dispute or correct this” guidance; if it is not yours, use “Not me”.
- A removal is stuck: open the case in your removal queue — it shows the current status and the next step LeakGuard can honestly give.