Skip to content

Method

How TrapCheck works

Check the address, recognize listed organizations, and review the warning signs and checks that completed.

You paste a text, email, DM, or link. We extract http(s) URLs, run local structure and language checks, and — only when available for your plan and the lookup succeeds — consult connected reputation services. We never open the suspect URL in a browser.

Thin evidence produces an inconclusive result, not a finding that a website is dangerous. A green result means no significant warning signs appeared in the checks performed; it does not guarantee safety.

Normal website behavior is not a scam finding

Long addresses, registration age and redirects within the same registered website do not by themselves establish danger. A short address can lead to a longer sign-in address. TrapCheck distinguishes the submitted link from destinations it observes, and checks those destinations for warning signs.

Reviewed service routes can recognize Gmail’s normal sign-in flow when connection and redirect checks finish. This does not approve arbitrary Google links or messages asking for passwords or money. Connection timeouts and inspection limits are missing information; a verified certificate error remains a warning.

Free registration checks use IANA’s official RDAP registry directory to find the relevant HTTPS registry. A registry may omit dates, restrict access or fail to respond. Registration facts identify when a domain was registered; they do not certify its owner’s honesty.

Organization recognition is included in Free

TrapCheck matches listed websites against U.S. and UK government domain registers, FDIC bank records, NCES education records, National Cancer Institute profiles, HRSA community-health-center records, Care Quality Commission care-directory records, and reviewed publisher references. Some sources identify an exact department website path rather than an entire domain. Coverage is incomplete. Education uses the 2024 institution directory; a retrieval date does not make an older reporting period current.

Results show the organization, evidence source and review dates separately from known-threat lookups. Matching a domain does not approve every page, subdomain, redirect, message or payment request. Missing or expired identity evidence does not mean a site is illegitimate.

For newly recognized organizations, only the exact listed HTTPS homepage can qualify for green through recognition, after connection and redirect checks complete without warning signs. A listed department path does not approve the parent website or neighboring departments. Threat findings and scam-message warnings take precedence. Directory matching uses a stored public-data snapshot and does not send your submitted link to these directories.

Contains public sector information licensed under the Open Government Licence v3.0. This service uses Care Quality Commission information and UK government domain-registration data. These organizations do not endorse TrapCheck. CQC warns that some updates, including registration cancellations, can be delayed; recognition is not a current registration, care-quality or licensing check.

What the statuses mean

  • CONNECTED — this check ran and succeeded (always-on local checks, or a scan lookup that completed).
  • CONFIGURED — enabled here. Not a successful scan. Catalog pages never use CONNECTED for Web Risk, Safe Browsing, VirusTotal, urlscan, or RDAP.
  • UNAVAILABLE — missing key, disabled, or the lookup failed (including RDAP HTTP 403). Never scored.
  • PLANNED — Phase 2. Screenshot and QR are not live.

Adapters on this deployment

  • URL structure

    CONNECTED

    This local check runs on every submit.

    Always on. Checks scheme, host, ports, shorteners, and hidden redirects.

  • Typosquat hints

    CONNECTED

    This local check runs on every submit.

    Always on. Compares hosts to well-known consumer brands. Not a full trademark scan.

  • Scam-language heuristics

    CONNECTED

    This local check runs on every submit.

    Always on. Looks for urgency, credential asks, impersonation, payments, and OTP harvest.

  • Domain parse

    CONNECTED

    This local check runs on every submit.

    Always on. Splits host, registrable domain, and TLD. Never visits the site.

  • Organization identity

    CONNECTED

    This local check runs on every submit.

    Matches exact listed hosts against a dated public-directory snapshot. No match does not imply danger. Identity is separate from threat checks and does not verify pages, messages or licensing status.

  • Google Web Risk

    CONFIGURED

    Enabled on this deployment. A result is CONNECTED only if that scan’s lookup succeeded. Missing keys and runtime failures (including RDAP HTTP 403) are UNAVAILABLE and never scored.

    Credential present. A result is CONNECTED only if that scan's lookup succeeds.

  • Google Safe Browsing

    UNAVAILABLE

    No key, disabled, or not usable here — never scored.

    No GOOGLE_SAFE_BROWSING_API_KEY. Adapter never scores.

  • VirusTotal (commercial)

    UNAVAILABLE

    No key, disabled, or not usable here — never scored.

    Commercial license confirmation is required; the public API is not used in this product.

  • urlscan.io

    UNAVAILABLE

    No key, disabled, or not usable here — never scored.

    No URLSCAN_API_KEY. Adapter never scores.

  • RDAP domain registration

    CONFIGURED

    Enabled on this deployment. A result is CONNECTED only if that scan’s lookup succeeded. Missing keys and runtime failures (including RDAP HTTP 403) are UNAVAILABLE and never scored.

    Enabled for scans. CONNECTED only if that scan's RDAP or WHOIS lookup succeeds. HTTP 403 is UNAVAILABLE and never scored.

  • Redirect path

    CONFIGURED

    Enabled on this deployment. A result is CONNECTED only if that scan’s lookup succeeded. Missing keys and runtime failures (including RDAP HTTP 403) are UNAVAILABLE and never scored.

    Manual server-side redirects with local/private destination blocking.

  • TLS certificate

    CONFIGURED

    Enabled on this deployment. A result is CONNECTED only if that scan’s lookup succeeded. Missing keys and runtime failures (including RDAP HTTP 403) are UNAVAILABLE and never scored.

    No-cost server-side TLS handshake enabled.

  • Screenshot analysis

    PLANNED

    Phase 2 — not built. We will not fake it.

    Phase 2. Upload is disabled in this MVP — we will not pretend to read images.

  • QR code scan

    PLANNED

    Phase 2 — not built. We will not fake it.

    Phase 2. In-app QR capture is disabled — we will not invent a decode.

What we will not do

We do not invent AI confidence percentages. We do not pretend screenshot or QR analysis works. We do not follow your link as if we were you.