Feed live·rebuilt ·every 6 hours·28/29 active sources OK·1 needs review: sblam·1 dormant: feodo_tracker (acknowledged)
xfeedspublic threat intelligence
Operator console · primary tier

Corroborated threat intelligence, refreshed every six hours.

8,828Safe to blockHigh-confidence, corroborated evidence
3,608Worth challengingRate-limit, tarpit, or elevate for review
12,436Published today12,345 IPv4 · 91 IPv6
78%Rejected before publish44,480 observations held back this run

Corpus growth and turnover · last 40 refreshes

Safe to blockWorth challengingAddedRemoved
02,0004,0006,0008,00010,00011-0611-1311-20

Last 40 refreshes. Hover or focus a column for that refresh's numbers.

Source health per runAll active sources OK1 missing2+ missing

Look up a specific address below, download a feed, or review deployment examples. Full aggregate evidence in analysis.

Local check

Check an address

Does it appear in this feed right now?

Runs entirely in your browser. The production page loads lookup.json only when you check; no query is logged.
Local lookup
Ready when you are

Enter an address to compare it with the current published feed.

Downloads

Download a feed format.

Primary / commercial-safe tier

Non-commercial equivalents retain these filenames under /noncommercial/.

Deploy

Deployment examples

Reference commands for common environments. Use the feed format above and schedule refreshes on the same six-hour cadence.

$ curl -sS https://neilweitzel.github.io/xfeeds/iptables.ipset | sudo ipset restore -! $ sudo iptables -I INPUT -m set --match-set xfeeds src -j DROP

Loads the IPv4 high-confidence feed, then drops matches. Add the first line to cron every six hours.

About the project

What xfeeds is, and why it exists.

xfeeds is a public, corroborated malicious-IP feed. Every published record is backed by at least one high-precision source or by two independent classes of evidence — not by the number of source files that happen to repeat it. Widely-scoped prefixes and known cloud, CDN, and resolver space are held back rather than published as blocks.

The corpus draws on 30 sources across 9 independent evidence classes that can publish a record, plus 8 further classes that corroborate without ever admitting one — because their licence forbids redistribution, or the upstream behind them has stopped publishing, refreshed every six hours.

Why it matters

An open, auditable alternative to opaque blocklists.

Commercial blocklists rarely publish the evidence behind a record. xfeeds does: the analysis surface shows the publication funnel, the corroboration histogram, per-source health, and the IPv4 address-space distribution, so a security team can decide for themselves what to trust.

Everything is generated from committed artifacts, hosted on GitHub Pages, and reproducible from source.

Cite and follow

Citeable research output.

Each release is archived on Zenodo with a persistent DOI, so xfeeds can be cited in papers, reports, and other tools. Cite the concept DOI 10.5281/zenodo.22045733 — it always resolves to the newest published version.

Maintained by Neil Weitzel
https://orcid.org/0009-0007-2546-2331

Machine-readable citation metadata is in CITATION.cff, and archival policy is in docs/CITABILITY.md.