Privacy
Addresses in FoxTrust's data
FoxTrust rates IP addresses so that sites can protect their sign-in and sign-up pages against brute force, credential stuffing and automated abuse. An IP address can be personal data, so this section explains what happens to the addresses FoxTrust lists.
- Who is responsible: RedFox Software, which runs FoxTrust. Contact: disputes@foxtrust.dev.
- The data: IP addresses and ranges, the network category of each (for example hosting, VPN, Tor exit), observed behaviour (for example SSH brute-force attempts) with the time it was last seen, a risk score, the network's AS number, name and registration country, and the country, region and city where the address is located. No names, emails, hostnames, coordinates, postal codes, locations finer than a city, or request contents.
- Where it comes from: public lists of network ranges (hosting, cloud, VPN, Tor, unallocated space), public lists of addresses reported for attacks or seen probing decoy systems, attacks that servers running FoxAuth reported (below), internet registry data, location data published by network operators, and DB-IP's free country and city databases. Not from the people the addresses belong to.
- Why, and on what basis: network and information security and fraud prevention, under legitimate interest (GDPR Art. 6(1)(f)). FoxTrust weighed this interest against the interests of the people behind listed addresses: categories describe networks, behaviour fades and leaves the record, sources and matched ranges are never published, and every listing can be disputed.
- Who receives it: anyone who downloads the signed snapshot files (risk, categories, observed
behaviour and the country) or the signed attacker list (addresses seen guessing passwords in the
last 7 days, read by FoxAuth servers); the operators who run FoxTrust's
/verifyin front of their own sites. The network's AS number, name and registration country, and the region and city, are given only through the API and this site: to holders of API keys, for the addresses they ask about, and to visitors of My IP for their own address. Sites decide themselves whether to allow, challenge or block a request. - How long:
- observed behaviour: the raw reports for 90 days after the address was last seen, daily summaries for up to 455 days; a reason leaves verdicts earlier, once it adds less than half a risk point;
- network categories: the history of when a range was listed, to explain past verdicts;
- published snapshot files: one year after they were replaced;
- dispute records: one year after the dispute is closed.
- Processor: Cloudflare carries the traffic of this site, the API and the snapshot files, so it sees the address of each request. FoxTrust sends the data to no one else.
- Your rights: you can ask what FoxTrust holds about an address you use, have a wrong listing corrected, ask for a reason to be hidden while it is checked, object to a listing, or ask for erasure. How, and what proof of control is needed, is on the dispute page. You can also complain to a data protection supervisory authority.
Downloads of the published files
FoxTrust counts downloads of its published files (the snapshots, the attacker list and the manifest), so that the people who build FoxTrust and FoxAuth know how many installations use them.
- What is kept: per day, per kind of file and per kind of client, how many files were sent, how many requests got "not modified" and how many failed, and an estimate of how many different readers there were. The kind of client is read from the
User-Agentheader: FoxAuth, FoxTrust's/verify, FoxTrust's own site and API, or other. For FoxAuth and FoxTrust the version is kept too. - What is never kept: addresses, the header itself, or anything else about a single request. The reader estimate is made in memory with a hash of the address and a random value that changes every day; neither is stored, and both are dropped when the day ends.
- Who sees it: FoxTrust's operators only. It is never shown to customers or on this site, and never used to rate addresses.
- How long: 13 months.
Your account
You sign in to FoxTrust with a foxauth account. FoxTrust stores only what it needs to run your keys:
- the identity foxauth gives us for you (the issuer and an opaque subject id), and when you last signed in;
- your API keys, as hashes only, with their labels and limits;
- how many lookups each key made per day, by outcome.
Your name and email stay in foxauth. FoxTrust holds them only for your sign-in session, at most 12 hours, to show them, and keeps neither after it. To change them, or to delete them and your foxauth account, use your foxauth account settings.
Before you delete your foxauth account, revoke your API keys on your account page. FoxTrust is not told when a foxauth account is deleted, so keys you leave keep working until they are revoked; what remains of the account holds no name or email.
Your lookups
FoxTrust does not store or log the addresses you look up, one at a time or in a batch. It keeps only per-key daily counts. For those addresses you are the controller and RedFox Software, which runs FoxTrust, is your processor: it uses them only to answer you, under the processing terms in the terms of use.
/verify reports
A /verify that runs with one of your keys sends FoxTrust counts of its decisions once a minute, so that your account page can show them. Without a key it sends nothing, and you can turn the reports off while keeping the key.
- What a report holds: how many requests were allowed, challenged or blocked, by policy rule, decision reason, verdict level, reason code, country and network (AS number), and how many challenges were shown, passed, refused or blocked.
- What it never holds: a visitor's address or network, the page asked for, headers, cookies or tokens.
- Who sees them: only the account whose key sent them.
- How long: counts per minute for 8 days, counts per hour for 32 days.
/verify itself writes visitors' addresses to its own log only as their /24 or /48 network, unless its operator turns full addresses on.
FoxAuth usage reports
FoxAuth servers, unless their operator turns it off, send FoxTrust one anonymous usage report a day, so that the people who build FoxAuth know how it is used. A FoxAuth operator can turn the reports off in the FoxAuth console.
- What a report holds: a random identifier of the FoxAuth installation, its version, runtime, operating system and storage, which features are on, and sizes as ranges (for example "11-100" applications), with the previous day's token and server-error counts, also as ranges. Newer FoxAuth versions add the public half of the installation's key pair, which it later proves to receive a key of its own (below).
- What it never holds: addresses, hostnames, issuers, names, emails, user, client or application identifiers, or any text an operator or user typed.
- The connection's address: used only in memory to limit how many reports one sender can send, as a hash with a seed drawn at start. FoxTrust does not store or log it.
- Who sees them: FoxTrust's operators only. They are never shown to customers or on this site, and never used to rate addresses.
- How long: each report for 13 months; after that only daily totals without any installation identifier.
Lookups by FoxAuth servers
FoxAuth servers, unless their operator turns it off in the FoxAuth console, look up the network (AS), region and city of the addresses their own users sign in from, so that their administrators can see where a sign-in came from. They do this with an API key of their own, which a server gets by proving the key pair it has sent in its usage reports for at least 7 days.
- Who is responsible: the operator of each FoxAuth server is the controller for its users' addresses. RedFox Software, which runs FoxTrust, processes them on that operator's behalf, as a processor, under the processing terms in the terms of use.
- What FoxTrust does with them: answers the lookup and nothing else. The addresses are not stored, not logged, and never used to rate addresses, in feeds or in snapshots. Only the number of lookups per key and day is kept.
- The server's own address: when it registers for its key, used only in memory to limit how often one sender can try, as a hash with a seed drawn at start. FoxTrust does not store or log it.
- What FoxTrust keeps about the server: its random installation identifier, the public half of its key pair, and its key as a hash, with that key's daily counts.
FoxAuth attack reports
FoxAuth servers, unless their operator turns it off, send FoxTrust once an hour the public addresses that attacked them in the previous hour, for example by guessing passwords or scanning for configuration files. FoxAuth's attack reports page says how an operator turns them off.
- Who is responsible: RedFox Software, as for all addresses in FoxTrust's data.
- What a report holds: a random identifier of the FoxAuth installation, its version, how the
operator says the server is reached, the hour, and per attacking address (IPv4, or an IPv6
/64) how many failed sign-ins, failed client or administrator credentials, mail requests, request floods and scanner requests it saw, with the kind of path scanned (for example "git"). - What it never holds: usernames, emails, passwords, the paths or pages asked for, user agents, request contents, hostnames, issuers or names.
- Why, and on what basis: network and information security, under legitimate interest (GDPR Art. 6(1)(f), Recital 49). Reports can be forged, so an address is listed only when several unrelated FoxAuth servers report it, never on one server's word.
- The sending server's address: used only in memory to limit how many reports one sender can send, as a hash with a seed drawn at start. FoxTrust does not store or log it.
- How long: raw reports for 30 days. An address that several servers report becomes observed behaviour in FoxTrust's data, kept as described above (90 days, then daily summaries); its reasons never name FoxAuth or any server.
- Your rights: a listing from these reports can be disputed like any other, on the dispute page.
Visitors of this site
This site sets no tracking cookies and runs no scripts. A session cookie exists only while you are signed in. FoxTrust does not log page views with your address.
My IP, /ip and /ip.json use the address of your request only to answer you: /my-ip looks it up in FoxTrust's published data and shows you the result, /ip and /ip.json echo it. The address is not stored or logged, and the per-address request limit keeps only a hash of it in memory, salted afresh each time the site starts, never the address itself.
Contact
Questions about this page, or about a listing: disputes@foxtrust.dev.