privacy
Privacy notice
Last updated: . Operator: NeoShield Security(個人事業主), a sole proprietorship operating from Japan. This notice explains what NeoShield collects, why, who processes it, how long it is kept, and the choices you have. It describes the system as built; it is not legal advice.
Quick summary
- Your e-mail address is your account identifier. The service also keeps operational records linked to it — listed in full below.
- Sign-in uses one-time e-mail codes. There are no passwords; codes are stored only as a keyed hash and expire after 5 minutes.
- AI features send the content you submit to Anthropic to produce a result. Most tools keep nothing afterwards; where the result becomes part of a saved record (for example a scan report), it is stored with that record.
- Card payments happen on Stripe-hosted pages; card numbers never reach this server.
- Analytics and advertising cookies are off until you opt in, and can be switched off again at any time in Cookie settings.
- You can download a copy of your data and request deletion from your account page, or by e-mail.
Who is responsible
For your account, purchases and general use of the site, the operator named above decides how data is used. For workspace features (the SOC Console, Security Center, monitored sites and device fleet), the workspace owner decides what is ingested; NeoShield stores and analyses that data on the owner's behalf, separated per workspace. Team customers can request a data processing agreement by e-mail.
What we store
- Account. E-mail address, role, whether the address has been verified, sign-in timestamps, sign-in code hash and expiry (cleared on use), e-mail preferences.
- Sessions. For each signed-in session: a random reference, start and last-seen times, the network your IP belongs to (/24 or /48 — not the full address) and a browser/OS label, so you can review and end sessions.
- Plans, purchases and credits. Plan grants with start and expiry, Team seats and invitations (invite links are stored only as hashes), add-on credit balances and their ledger, Stripe checkout and payment references, amount, currency, status and receipt e-mail. No card data.
- Website scanning. Targets you scan, the resulting reports (findings, evidence such as response headers and DNS records, scores and AI-written remediation text), finding status history, asset inventory, scheduled scans, notification addresses and webhook URLs you configure, and domain-verification records (method, time, who verified; the token itself is stored only as a hash).
- Workspace / SOC data (Team). Log events you ingest (which can contain IP addresses, host names and message text), alerts, incidents and actions, assets and device inventory, alert channels, and hashed API and agent keys.
- Academy. Progress, quiz results and certificate references.
- Messages and contributions. Contact-form name, e-mail, message and time; quote requests; blog submissions with your e-mail as author.
- Security and audit events. Sign-ins, failures, rate-limit and blocking decisions, purchases and grants, administrative actions — each with time, event type, your e-mail where known, IP address and a truncated browser identifier.
- Abuse prevention. Short-lived rate-limit counters (per e-mail, IP and network), a salted hash of a random device identifier used to stop free-tier abuse, and reputation data about e-mail domains and IP addresses.
- Usage metadata. Counts of which tools and features are used, per account and in aggregate, and AI usage metering (token counts and model name, never the prompt).
- Consent receipts. A random receipt id, the consent version, your analytics/advertising choices and the time — no IP or e-mail.
What we do not store
- Passwords — there are none.
- Card numbers — entered on Stripe's pages only.
- Input to offline tools — Quantum Vault, hashing, the secret and PII scanners, JWT/phishing/Sigma analyzers and similar tools process input in memory for that request and do not log or store it.
- Your password in the exposure check — only the first five characters of its SHA-1 hash are sent to Have I Been Pwned (k-anonymity).
- AI prompts in logs — AI calls are metered by token count; prompt text is not written to our logs.
Why we use it (and legal basis, where the GDPR applies)
- Provide the service you asked for: account, scans, reports, workspaces, purchases — performance of a contract.
- Keep the service and other users safe: security logging, rate limits, abuse and fraud prevention — legitimate interests.
- Accounting and tax records for purchases — legal obligation.
- Analytics and advertising cookies — consent, which you can withdraw at any time.
- Optional e-mails (daily briefing, plan reminders) — your choice; every briefing has a one-click unsubscribe.
Under Japan's Act on the Protection of Personal Information (APPI), we use personal information only for the purposes above, and we do not provide it to third parties except to the processors listed below, as required by law, or with your consent.
Cookies, analytics and advertising
A session cookie (NSS_SID: HttpOnly, Secure over HTTPS, SameSite=Lax) keeps you signed in. Google Analytics,
PostHog and Google AdSense are loaded only after you opt in, as separate choices, and never pre-selected. Advertising
is not shown while you are signed in, on dashboards, on scanner results, or on pages that are not indexed. The complete list
of cookies and browser storage, with lifespans, is in the Cookie Policy, where you can also
change your choice.
AI processing (Anthropic)
AI-assisted features send the content needed for that request to Anthropic's API: for example a pasted e-mail, code or log excerpt, a configuration file (with recognisable credentials removed first), your question, or the facts collected by a website scan. Your account e-mail and IP address are not included in the request. Tools that show a result once do not store it; features whose output is part of a saved record (scan reports, SOC incidents, Factory tool specifications) store the output with that record. Anthropic processes API data under its own commercial terms and privacy policy, which set its retention; we do not make promises on Anthropic's behalf. See Anthropic's privacy policy and commercial terms.
The NeoShield Security Agent (Pro and MSP / Team, when you choose to ask it) sends the selected finding's title, severity and classification, its scan evidence after cookies, authorization headers, tokens, keys, URL query strings and e-mail addresses are removed, NeoShield's remediation guidance, and any configuration text you paste. Reports are referred to by random identifiers; your e-mail and account ID are not sent. Prompts and answers are not stored. An audit record keeps the request type, those identifiers, the model, token counts and which data tools were allowed or refused, for the security-log retention period. The agent cannot change your systems, start a scan without your approval, or mark a finding resolved.
Processors and where data is processed
- Hostinger — web hosting, database and file storage, server e-mail (location depends on the hosting data centre; see the subprocessor list).
- Stripe — payments (United States and other countries).
- Anthropic — AI features (United States).
- Google — analytics and advertising, only with consent; public DNS-over-HTTPS lookups used by some tools.
- PostHog — product analytics, only with consent and only if enabled.
- Have I Been Pwned — password exposure check (k-anonymity range queries only).
- Configured e-mail provider — delivery of sign-in codes and notifications, if SMTP is configured.
The maintained list, with purposes and how we notify material changes, is published in the repository as SUBPROCESSORS.md and summarised on the Trust page. Some processors are outside Japan and your country; using the related feature involves transferring the relevant data to them under their terms.
How long we keep it
- Sign-in codes: unusable after 5 minutes or first use; the stored hash is cleared on use or on the next attempt.
- Sessions: end after 30 minutes idle or 12 hours after sign-in; registry entries are removed at the same time.
- Activity and security logs: 30 days (application log), with a size cap.
- Rate-limit counters: up to 24 hours.
- Scan reports: until you delete them or your account (the most recent 100 per account are kept in file storage).
- SOC events and incidents: until you clear the workspace, delete them, or delete the account.
- Purchases and payment references: as long as needed for tax and accounting records (Japanese law generally requires 7 years).
- Plan records after a pass lapses: kept as a lapsed record so renewals extend correctly; deleted with the account (except the accounting records above).
- Usage and analytics metadata: up to 400 days. AI token metering: 90 days. Consent receipts: 13 months.
- Backups: application backups rotate after 14 runs; the hosting provider's own backups rotate on its schedule. Deleted data can remain in backups until they rotate out.
The full schedule, including the deletion mechanism for each category, is in RETENTION_SCHEDULE.md.
Your rights and choices
- Download your data. Your account page provides a JSON export of the records linked to your e-mail.
- Delete your account. Request deletion from the account page. We confirm by e-mail and complete deletion of live data within 30 days, except records we must keep by law (purchase and tax records). Copies in backups expire as backups rotate. You can see the status of your request on the account page.
- Correction, access and other requests. E-mail security@neoshieldsecurity.com or use the contact form. Depending on where you live you may have rights under the GDPR, the APPI or other laws; we will honour applicable requests.
- Cookies. Change or withdraw consent at any time in Cookie settings.
- E-mail. Unsubscribe from the daily briefing with one click from any briefing.
Children
The service is intended for security professionals and learners and is not directed to children under 16. If you believe a child has provided data, contact us and we will remove it.
How we protect data
TLS with HSTS, prepared database statements, CSRF tokens on state-changing forms, outbound-request (SSRF) guards, hashed storage of every secret we issue, per-workspace data separation checked by automated tests, and a rule that AI-generated code is never executed on the server. Data at rest is not encrypted by the application itself. Details and the current assurance status are on the Security page.
Changes
We update this notice when the platform changes and show the date at the top. Material changes are also announced on the site.