Tool profilesBased on official documentation, not tested by us

CrowdSec: a free, open source engine that reads your logs, spots attackers and blocks them, with a shared community blocklist

CrowdSec is an open source security engine that analyses logs and web requests for attack patterns and blocks the offending IPs, and can use a blocklist built from what other users detect. Based on its official README, documentation and pricing page, not tested by us: what it does, what it needs, its limits and what we could not confirm.

Type

Intrusion detection and blocking engine (reads logs, blocks abusive IPs)

License and cost

MIT licence. The README and documentation present the engine as open source and the pricing page mentions a free Community plan without listing its limits. Paid products exist (blocklists, IP reputation API, premium console) starting at $49 a month for the API and in the thousands per month for the others, which we did not evaluate

Reviewed

11 Nov 2026

Version

Not shown on the pages we read (the documentation selector showed v1.8)

Best for

Anyone who runs their own server, with a public web service, SSH or similar, and wants automatic blocking of brute force and scanning without writing their own rules.

Watch out for

It acts on a machine you administer and blocks traffic, so a wrong rule or a missing allowlist could block people you want in. No page we read covers allowlists or lockouts. Start in a test machine and read what it would block first.

← All tool profilesNo. 11 / 11

In one sentence

CrowdSec is a security engine that reads your server's logs and web requests, looks for attack behaviour (it calls the patterns "scenarios") and asks a separate component to block the source. Optionally it also shares what it sees with a community and downloads a blocklist in return.

Who it is for

People who run their own Linux, Windows or container hosts and expose something to the internet: a web app, SSH, an API behind a reverse proxy. It is not for a founder whose whole product sits on a hosted platform where the provider owns the servers. We cover it as analysis only.

What the documentation says

Everything below comes from the README, the documentation introduction and the pricing page, read on 11 November 2026.

  • What it is: the repository describes it as open source intrusion detection and prevention, a web application firewallFirewall A filter that decides which network traffic may reach a device or service and which is blocked. Tap again to open the glossary (WAF) and bot detection for Linux, Windows, Docker and Kubernetes, with a crowdsourced blocklist of malicious IPs. The licence is MIT.
  • How it works: the engine analyses logs and HTTP requests with behaviour-based scenarios. It ships by default with scenarios such as brute force, port scan and web scan, and more can be taken from the Hub.
  • The pieces: a Log Processor does the detection; the Local API sits between it and the Remediation Components (also called bouncers), which do the blocking at places such as firewalls, web servers or reverse proxies. There is a command line tool (cscli), a web Console and a Central API used to share alerts and receive blocklists.
  • Web application firewall: its AppSec component filters HTTP requests from compatible web servers. The README says it is built on Coraza and that ModSecurity-style rules and the OWASP Core Rule Set load as they are.
  • The community side: the documentation says that when you participate you share the attacks you detect and block, and in return the engine downloads a curated list of validated attackers.
  • Where to install: the README says it can be installed on Linux, Windows, Docker, OPNsense, Kubernetes and more. For centralised logs the documentation advises running it next to your log pipeline, not on production workloads.
  • Free and paid: the pricing page mentions a free Community plan without giving limits. It lists paid products, among them an IP reputation API from $49 a month for 5,000 queries and blocklist and tracker products starting at $1,900 to $2,000 a month for small companies. The page says commercial use of CrowdSec data needs a partnership programme.

The good

  • Open source under MIT, with one engine for logs and for web requests.
  • Blocking is separate from detection, so you choose where to enforce it.
  • Ready-made scenarios for the common abuse (brute force, scans) from the start.
  • It can run in containers, on Kubernetes and on more than one operating system.

The limits it shows

  • Sharing is the design. The documentation describes participation as sharing the attacks you detect. The pages we read do not say whether it is on or off by default, so check before you enrol a server.
  • It needs logs to read. It only sees what your services write or what your web server passes it.
  • The free plan's limits are not listed on the pricing page.
  • The pricing page carries vendor testimonials with figures such as "zero false positives". These are the vendor's claims; we did not verify them.

When not to use it

  • On a machine you do not control or have no permission to change.
  • If you cannot say what data about your users may leave the machine. Decide that first.
  • If you expect it to replace patching, strong authentication or backupsBackup A copy of your data kept apart from the original, so you can recover it if something is lost, broken or locked. Tap again to open the glossary. It blocks bad traffic; it does not fix a vulnerable app.
  • If you cannot test a rule before it blocks real people.

What it costs you

A server you administer, time to install the engine and at least one remediation component, and time to watch what it blocks and to tune it. We found no figures for setup time or resource use. It is not a one-off: scenarios and blocklists change over time.

Typical setup mistakes

  • Installing it on a production host without a way back in. Keep a second access route and an allowlist for your own addresses. This is our advice; the pages we read do not cover it.
  • Enrolling a server without reading what is shared. Check what leaves the machine.
  • Detecting but never enforcing. Without a remediation component nothing is blocked.
  • Trusting the blocklist blindly. Review what it blocks for the first days.
  • Downloading it from an unofficial place. Use the official repository or package source and check what you get. This is our advice.

How to check that it works

Official documentation and our approach, not something we ran: in a throwaway virtual machine you own, with nothing real on it, install the engine and one remediation component, then make repeated failed logins from another test machine you own and see whether the attacker address appears in the decisions list and gets blocked.

Where to start

We did not test the tool, so we offer no tutorial for now. Start with the documentation: httpsHTTPS The secure version of the web address protocol: it encrypts what travels between a visitor and your site and checks that the site is the real one. Tap again to open the glossary://docs.crowdsec.net/docs/intro

What we could not confirm

We did not install or run CrowdSec. We read the README, the documentation introduction and the pricing page, not the getting started guide, the supported platforms list or the data-sharing details. No page we read showed a release version or date, whether sharing is on by default, how allowlists work, or the free plan's limits. We do not know how often it blocks legitimate traffic or how it compares with other tools.

Glossary: WAF, brute force, firewall, exposure scanner. Runbook 01, security check.

More free tools

Other profiles in this section.

Sources

Follow us

New profiles and security news: YouTube (@safet2ship) and Instagram (@safet2ship).

YouTube channel · Instagram

Get the free runbooks

One-page security checklists for founders who build with AI. You get a confirmation email first, then the runbooks.

Send me the runbooks