LibreCrawl

New

The free alternative toScreaming Frog SEO Spider

🟡 Intermediate⏱️ ~15 min to installYou save €22/mo

LibreCrawl is Screaming Frog without the licence. You give it your site's address, it visits every page and returns what an SEO audit needs: missing or duplicate titles and meta descriptions, headings, broken links, redirect chains, images without alt text, canonicals, sitemap. It renders JavaScript (React, Vue, Angular sites) with a real browser, measures Core Web Vitals through PageSpeed Insights, and exports everything as CSV, Excel, JSON or XML, with no cap. Screaming Frog stops at 500 URLs on the free tier; here there is no limit, the project reports crawls of over a million URLs. It is Python with a web interface in your browser, local or on a server, MIT licence. The repository is one year old, close to 1,000 stars, with fixes every couple of weeks: young, but alive. A plugin workshop lets you add your own analyses.

935View on GitHub ↗

Versus Screaming Frog SEO Spider: the match

Price
LibreCrawl
€0, on your machine or your server
Screaming Frog SEO Spider
279 dollars a year, free tier capped at 500 URLs
URL limit
LibreCrawl
None, crawls of over a million URLs
Screaming Frog SEO Spider
500 on free, unlimited with the licence
JavaScript rendering
LibreCrawl
Included, with a real Chromium
Screaming Frog SEO Spider
Licence only

Screaming Frog SEO Spider ↗

VS	LibreCrawl	Screaming Frog SEO Spider ↗
Price	€0, on your machine or your server	279 dollars a year, free tier capped at 500 URLs
URL limit	None, crawls of over a million URLs	500 on free, unlimited with the licence
JavaScript rendering	Included, with a real Chromium	Licence only
Core Web Vitals	PageSpeed Insights built in	With the licence, through the PageSpeed API
Exports	CSV, Excel, JSON, XML, no limit	Capped on free
Where it runs	Web interface, local or on a shared server	Desktop app, one licence per person

How to install it (even from zero)

Two ways: Docker in three commands, or Python locally. The prompt picks according to your machine, installs what is missing and opens the interface on port 5000.

Paste this prompt into Claude (Code or Desktop)

It checks what you have, installs LibreCrawl from the official repository and runs a first test crawl:

Installe LibreCrawl sur ma machine (crawler SEO open source, dépôt GitHub PhialsBasement/LibreCrawl, site librecrawl.com) en suivant son README. Vérifie d'abord ce que j'ai. Si Docker est installé : clone le dépôt, copie le fichier .env.example en .env et lance docker compose up -d. Sinon : installe Python 3 si besoin, crée un environnement virtuel, installe les dépendances de requirements.txt, lance playwright install chromium puis python main.py --local. Ouvre ensuite http://localhost:5000 et vérifie que la page répond. Fais un premier crawl de test sur une petite page pour confirmer que tout marche, avec le rendu JavaScript activé. Si c'est un serveur exposé sur internet, protège l'accès (mot de passe ou reverse proxy avec authentification) avant de me donner l'adresse.

Review the GitHub repository and check whether it is safe to install given my machine and my infrastructure. Ask me as many questions as possible, using widgets, to clarify my request and be as precise as you can.

2
Run your first real crawl
Paste your site's address, keep JavaScript rendering if your site relies on it, and read the tabs: pages, links, detected issues, Core Web Vitals.
3
Export and fix in order
Export the issues as Excel or CSV, sort by severity, and start with missing titles, broken links and redirect chains: those are the quick wins.

What you actually do with it

Audit your site before a redesign

A full crawl gives you the list of pages, their titles, their inbound links: you know what to keep, redirect or delete.

Find a client site's broken links

404s, redirect chains, images without alt text: a clean export to send over, without buying a licence for a one-off audit.

Follow a site after a migration

You rerun the crawl every week and compare the exports: vanished pages and missing redirects stand out.

Project health, at a glance

Before installing an open-source tool, one question matters: is the project alive? These scores are computed from public GitHub data. They are not made up.

45/100

Fair

Decent project, keep an eye on it over time.

Activity: 15
Maintenance: 85
Community: 52
Maturity: 55
Momentum: 20

The repo in numbers

GitHub stars: 935
Forks: 201
Contributors: 11
Commits: 103
Project age: 1 yrs
Last commit: 9 d ago
Size: 0.4 MB
License: MIT

Built with

Python 49.4%
JavaScript 29.3%
HTML 16.5%
CSS 3.9%
Shell 0.5%

Scores computed automatically from the GitHub API (last-30-days activity, update freshness, community size, project age). Refreshed daily.

Loucash’s take

7/10

If you pay for Screaming Frog to crawl more than 500 pages, LibreCrawl does the job without the licence: same basic checks, JavaScript rendering, Core Web Vitals, unlimited exports. For a site audit, a migration or a monthly follow-up, you have what you need. What I like: the web interface you can share on a server for the whole team, the absence of a cap, and a project that fixes fast, with close to 1,000 stars in a year. What may bite: it is young, carried by a small number of people, with no numbered release; Screaming Frog's advanced reports (fine custom extraction, crawl comparison) are less polished; and the PageSpeed API has its quotas. For regular professional use, keep an eye on the repository's activity.

It’s for you if: You do SEO for yourself or for clients and you do not want to pay a yearly licence to crawl more than 500 pages.