National Cyber Warfare Foundation (NCWF)login

IPs for blocking malicious and abusive hosts at the network edge


0 user ratings
2026-11-12 13:24:25
milo
Red Team (CNA)
"IPs

IPs is a continuously updated threat feed of malicious, scanning, and tracking IP addresses harvested from distributed honeypots, built for defensive filtering with ipset, iptables, and fail2ban.








ToolShadowWhisperer/IPs — hourly-updated IP blocklists derived from multi-location honeypots
CategoryThreat intelligence feeds / IP blocklists
Primary UsePopulating ipset, iptables, and firewall deny lists to block known scanners, botnets, and active threats on assets you administer
Safe UsePurely defensive: intended for sysadmins and authorized defenders to harden their own networks, mail servers, and honeypots; it contains no offensive tooling
Telemetry NoteThe feed itself is telemetry — honeypot-derived evidence of probing and exploitation attempts; defenders can cross-reference hits in their own logs against the lists to detect exposure

IPs from ShadowWhisperer is not a tool in the traditional executable sense; it is a living threat intelligence repository, a collection of plain-text blocklists generated from honeypots the maintainer operates across multiple geographic locations. Every hour the lists refresh with addresses observed doing something abusive on the internet, and the repo is compressed periodically to keep clone sizes manageable. That hourly cadence matters for defenders: blocklists decay quickly, and a stale list is often worse than none because it breeds false confidence. The project sits squarely on the defensive side of the security tooling ecosystem, with topics like blocklist, fail2ban, ipset, iptables, and firewall making its intended deployment context unambiguous.


The architecture, as far as the README reveals, is a classic honeypot aggregation pipeline: sensors deployed in multiple locations attract and record hostile activity, and the resulting addresses are classified into thematic lists committed to Lists/. Notably, the README states that diffs, pull requests, and similar history artifacts are cleared, which means the repository intentionally discards its own provenance trail. For an operator consuming this feed, that is a tradeoff worth understanding: you get current data without historical context, so you cannot easily audit when a given address was first observed or why it landed in a category. Trust in the classification therefore rests entirely on trust in the pipeline rather than in reviewable evidence.


The taxonomy of the lists is where the analytical value lives. Lists/Threats is the high-signal core: addresses classified as active threats, including exploit attempts, compromised systems, and droppers. Lists/Threats_Unclassified is the staging area — addresses seen behaving suspiciously but not yet classified — which is useful for correlation and trend-watching but should be treated with more caution if you plan to hard-block, since classification lag means false positives are plausible. Lists/Probes captures partial-connection probing observed five or more times, a threshold that deliberately filters out one-off noise and single accidental connections.


The remaining categories segment internet background radiation into operationally meaningful buckets. Lists/Scanners lists internet-wide scanning companies, a category many defenders handle differently from hostile actors — some organizations tolerate census-style scanning for research reciprocity while blocking it at production edges. Lists/Tunnels covers proxies and VPNs, useful for identifying traffic that is anonymized rather than overtly malicious, a common signal in fraud and abuse prevention workflows. Lists/Trackers contains addresses used to track user activity, aligning with privacy hardening as much as security. Lists/Ads covers advertisement-serving infrastructure, and Lists/DNS enumerates common DNS resolvers — a list that is genuinely dual-edged, since public resolvers appear on it as identifiable infrastructure rather than as a threat.


Deployment is left to the consumer, but the repo's topics telegraph the canonical path on Linux. The natural pattern is to fetch a list, convert it into an ipset, and reference that set from an iptables rule, giving you O(1) matching across tens of thousands of entries instead of a linear chain walk. fail2ban integration is equally idiomatic: rather than only reacting to local log patterns, you can augment its ban logic with a pre-seeded deny set. Because the lists are plain text, one address per line by convention, they slot into essentially any consumer — edge firewalls, nginx/HAProxy deny lists, mail server ACLs, or SIEM watchlists — without format wrangling.


The maintainer's operational credibility is partially externalized: the README carries an AbuseIPDB contributor badge linking to the reporter profile, suggesting the same honeypot infrastructure feeds community abuse reporting. That is a meaningful signal, because it implies the observations are consistent enough to be submitted as abuse reports against a third-party standard, not just dumped raw. It also means there is a degree of overlap and cross-validation possible: an analyst who wants independent confirmation of a listed address can check it against AbuseIPDB before acting, which is a sensible workflow for anything you intend to block at a hard perimeter.


Licensing is Unlicense, effectively public domain, which removes any friction for commercial or automated consumption — you can mirror, diff, fork, and embed these lists into proprietary defensive products without attribution obligations. At roughly 207 stars, the project has modest but real traction, which for a blocklist repo is a reasonable proxy for a community of active consumers who would notice quality regressions like wholesale false positives. The absence of a programming language in the metadata reinforces that this is data, not code: there is nothing to build, nothing to execute, and no supply-chain surface beyond the lists themselves.


Operational cautions deserve explicit attention. Hard-blocking Threats_Unclassified at a strict perimeter risks collateral damage against shared infrastructure such as CGNAT ranges or cloud egress addresses that happened to probe a honeypot; the conservative pattern is to alert on that list while hard-blocking only Threats and Scanners per your tolerance. Consumers should also automate the hourly refresh rather than snapshotting once, since the value proposition is freshness — a feed that is a week stale misses rotating botnet infrastructure. And because the repo clears diffs, keep your own local history of what you ingested and when; that audit trail is what lets you later answer 'was this address on the list when we decided to block or allow it?'


From a detection-engineering standpoint, the lists are useful even if you never block anything. Correlating your own inbound connection logs against Scanners and Probes gives you a baseline of expected background noise so that genuinely novel sources stand out. Matching outbound traffic from your estate against Threats can flag compromised internal hosts beaconing to known bad infrastructure — a far more alarming signal than inbound scanning. In a purple-team or lab context, the lists provide realistic hostile address sets for testing detection rules without generating attack traffic yourself.


In sum, ShadowWhisperer/IPs is a low-friction, defensively-oriented intelligence feed that trades detailed provenance for hourly freshness and zero-cost licensing. It will not replace curated commercial feeds with rich context and confidence scoring, but as a free layer for ipset-based edge filtering, fail2ban augmentation, and log correlation on systems you are authorized to administer, it earns its place in a layered defense. Treat the unclassified list as triage data, keep your own ingestion logs, and the repo becomes a quiet, steady contributor to network hygiene.



Official project repository for ShadowWhisperer/IPs.

Download Tool

Educational analysis for authorized security professionals. Use only in controlled, authorized environments.






Source: OffensiveSec
Source Link: https://www.offsecblog.com/2026/11/ips-for-blocking-malicious-and-abusive.html


Comments
new comment
Nobody has commented yet. Will you be the first?
 
Forum
Red Team (CNA)



Copyright 2012 through 2026 - National Cyber Warfare Foundation - All rights reserved worldwide.