Operator and contact

FupButik.dk is a Danish consumer information service operated by Kristian Mølholt Kristensen.

Technical contact: admin@fupbutik.dk. Please include the domain, affected URL, timestamp with time zone and any response reference.

Purpose and scope

A user submits a webshop URL for a consumer risk assessment. The crawler may fetch the submitted page and selected public product, business information, contact and terms pages. DNS and RDAP may provide additional technical domain information.

Visits are user initiated rather than a continuous bulk crawl. The scanner fetches server delivered content and does not execute JavaScript as a browser would. It does not bypass logins or attempt to solve access challenges.

Transparent identification

During normal analyses the crawler uses its own identity. It does not impersonate a browser, Googlebot or another service.

User-Agent: FupButikDetektiven/4.3 (+website risk analysis)

Signature-Agent on production: "https://fupbutik.dk"

HTTPS requests are signed using Web Bot Auth and Ed25519. Signature headers are Signature, Signature-Input and Signature-Agent. Each redirect request is signed separately. Outbound IP addresses have not been published as a verified IP list.

Public signing key

https://fupbutik.dk/.well-known/http-message-signatures-directory

{
    "keys": [
        {
            "crv": "Ed25519",
            "kty": "OKP",
            "x": "X340kqrM0LanD4wLhPZ6wFcw32_WuVx-0zJ24KsgkRI"
        }
    ]
}

Public key fingerprint: kSF3UXkStSIzecHQ9UuoKCRUnCZDRcE26iozYr89vHE

The key shown belongs to this environment. The private signing key is never published. A signed identity does not grant access to a webshop.

Robots.txt and access rules

Robots.txt is checked before content is fetched and at redirects. Allow and Disallow rules are matched against the URL using the most specific rule. The most specific matching User-agent group is used, falling back to wildcard rules. Policies are cached for up to one hour.

A robots.txt response of HTTP 404 or 410 permits fetching. HTTP 401 or 403 stops content fetching. Network failures, server errors and invalid or oversized policy files cause a pause. No new content pages are fetched while the policy cannot be verified.

Delays and request limits

Analyses in the same installation share a lock and request limit per destination hostname. The www and bare hostname share a limit across HTTP, HTTPS and ports. Different hostnames have separate limits. A maximum of 300 requests are sent per UTC day, including robots.txt and redirects.

At least two seconds pass after a response is received before the next request may begin. Concurrent requests to the same host are not allowed. Longer Crawl-delay, Request-rate and Retry-After values are respected. Waiting analyses retain their progress and use no AI call for the waiting step. Separate installations have separate request controls.

Opt out of visits

To opt out of visits by this crawler, publish the following group in robots.txt:

User-agent: FupButikDetektiven
Disallow: /

Individual paths may be excluded with Disallow. Changes are normally read within one hour. Contact admin@fupbutik.dk with questions or urgent requests.

Limited evidence

A robots.txt exclusion or a technical refusal is not by itself evidence of fraud and does not automatically reduce trust. When required content cannot be fetched, the investigation is marked Unavailable with the specific reason, without an overall risk assessment.

Excluding one subpage does not automatically exclude the entire webshop. Reports must distinguish retrieved information from missing evidence.

Read about our work on access to webshops (in Danish).

Data processing and retention

Relevant extracted public content is sent to Google’s Gemini API for risk assessment. Reports are stored for later reference. Data processing therefore also takes place outside FupButik’s own server.

Read about processing and your rights in our privacy policy.