Most SEO automation runs on paid software with monthly seat fees. Yet some of the most useful automation tools are free on GitHub. OpenSEO alone has more than 27,000 stars, and Google’s own Lighthouse CI checks pages on every code commit.
We reviewed 10 open-source SEO automation tools that you can run on your own server or laptop. They cover rank tracking, crawling, site-wide speed audits, Search Console dashboards and data for AI agents. For each one, we list the license, the language it’s built with, what it automates and what you need to run it.
- Open-Source SEO Automation Tools by Job
- 1. OpenSEO
- 2. SerpBear
- 3. Lighthouse CI
- 4. Unlighthouse
- 5. SiteOne Crawler
- 6. LibreCrawl
- 7. CrawlSEO
- 8. SEO for Agent-Native Teams
- 9. advertools
- 10. Scrapy
- What About n8n and Other Workflow Tools?
- Matching Each Open-Source Tool to Your SEO Job
- Costs to Plan For With Free SEO Software
- Before You Self-Host an SEO Tool
Open-Source SEO Automation Tools by Job
Star counts change daily, so treat them as a rough sign of community size.
| Tool | What it automates | License | GitHub stars |
|---|---|---|---|
| OpenSEO | Keyword research, rank tracking, backlinks, audits | MIT | 27,000+ |
| SerpBear | Rank tracking with email alerts | MIT | 2,100+ |
| Lighthouse CI | Lighthouse checks on every commit | Apache-2.0 | 7,100+ |
| Unlighthouse | Lighthouse scans across a whole site | MIT | 4,900+ |
| SiteOne Crawler | SEO, security and accessibility crawls | MIT | 990+ |
| LibreCrawl | Technical SEO crawls in a web app | MIT | 1,100+ |
| CrawlSEO | Search Console, crawls and Core Web Vitals | MIT | 690+ |
| SEO for Agent-Native Teams | Search Console data for AI agents | Apache-2.0 | 610+ |
| advertools | Crawls, sitemaps, robots.txt and logs in Python | MIT | 1,500+ |
| Scrapy | Custom crawlers and scrapers | BSD-3-Clause | 65,000+ |
1. OpenSEO
OpenSEO calls itself an open-source alternative to Semrush and Ahrefs. It covers keyword research, rank tracking, competitor insights, backlink analysis, site audits and AI visibility in one app.
The catch is data. OpenSEO pulls its numbers from the DataForSEO API, so you need your own DataForSEO key and pay for those requests yourself. If you’d rather not self-host, the hosted version costs $10 a month and adds a 28% markup on each DataForSEO request.
You can run it with Docker on your own server, or deploy it to your own Cloudflare account with one command. It also ships an MCP server, so AI agents like Claude Code can query your SEO data directly. Our OpenSEO review covers the hosted plan in more detail.
| Detail | Info |
|---|---|
| License | MIT |
| Built with | TypeScript |
| Use it for | A low-cost, all-in-one SEO platform |
| Repository | every-app/open-seo |
- Docker image and one-command Cloudflare deploy
- Email and password accounts with invite links for teams
- MCP server and reusable Agent Skills for AI agents
- Fork it to build your own SEO tool
2. SerpBear
SerpBear is a self-hosted rank tracker. You add unlimited domains and keywords, and it checks where you rank on Google and emails you when positions change.
It doesn’t scrape Google by itself. Instead, you connect a scraping service such as ScrapingAnt, ScrapingRobot, SearchApi, SerpApi or HasData, or use your own proxies. Many of these services have free tiers, which keeps small projects cheap. For paid options, see our list of rank tracker tools.
| Detail | Info |
|---|---|
| License | MIT |
| Built with | TypeScript |
| Use it for | Tracking keyword positions on a budget |
| Repository | towfiqi/serpbear |
- Unlimited domains and keywords
- Daily, weekly or monthly email updates
- Built-in API for reporting tools
- Google Search Console integration for real clicks and impressions
- Keyword ideas through a Google Ads test account
- Mobile app as a PWA
3. Lighthouse CI
Lighthouse CI comes from the Google Chrome team. It runs Lighthouse on every commit, shows what changed and can block a release when scores drop.
So developers catch a slow template or a missing meta tag before it reaches production. It works with common CI systems, and you can store reports on its own server to compare builds over time. Want other speed tools? Our list of Google Lighthouse alternatives covers them.
| Detail | Info |
|---|---|
| License | Apache-2.0 |
| Built with | JavaScript |
| Use it for | Stopping speed and SEO regressions before launch |
| Repository | GoogleChrome/lighthouse-ci |
- Runs Lighthouse automatically for each commit
- Assertions that fail a build when scores fall
- Optional server for storing and comparing reports
- Status checks on pull requests
4. Unlighthouse
Standard Lighthouse tests one URL at a time. Unlighthouse scans your entire site with Lighthouse and shows the results in a clean web interface.
It needs very little setup and uses smart sampling, so large sites with thousands of similar pages don’t take forever. As a result, you can spot the slowest templates in one run.
| Detail | Info |
|---|---|
| License | MIT |
| Built with | JavaScript |
| Use it for | Running Lighthouse on every page of a site |
| Repository | harlan-zw/unlighthouse |
- Site-wide Lighthouse scans from one command
- Web UI for browsing scores page by page
- Smart sampling for large sites
- CI mode for automated checks
5. SiteOne Crawler
SiteOne Crawler is a website analyzer, cloner and converter. It was rewritten in Rust, which made it 25% faster and cut memory use by 30%.
It runs as a single binary on Windows, macOS and Linux with no dependencies. You can launch an interactive wizard with 11 preset modes or use the command line in scripts. It also exports offline copies of a site, which helps with backups and audits.
| Detail | Info |
|---|---|
| License | MIT |
| Built with | Rust |
| Use it for | Fast crawls for SEO, security and accessibility |
| Repository | janreges/siteone-crawler |
- SEO, security, accessibility and performance checks
- Single native binary for Windows, macOS and Linux
- Interactive wizard with 11 preset modes
- Command-line interface for CI/CD pipelines
- Offline website export
6. LibreCrawl
LibreCrawl is a web-based, multi-user crawler for SEO analysis and site audits. Its creator describes it as an open-source alternative to Screaming Frog.
It renders JavaScript and exports crawl data to CSV, XLSX, JSON and XML. It’s built by one person, with no paid tier and no investors. Teams can share plugins through its Plugin Workshop.
| Detail | Info |
|---|---|
| License | MIT |
| Built with | Python |
| Use it for | A free, web-based alternative to desktop crawlers |
| Repository | PhialsBasement/LibreCrawl |
- Multi-user web interface
- JavaScript rendering
- Exports to CSV, XLSX, JSON and XML
- Plugin support
- API documentation
7. CrawlSEO
CrawlSEO combines Google Search Console, a site crawler, Core Web Vitals and an MCP server in one self-hosted dashboard. It targets founders who want SEO monitoring without learning a full SEO suite.
You run it with Docker Compose, and data stays in your own PostgreSQL database. Keyword and backlink data are optional through your own DataForSEO key, while Google Autocomplete suggestions work for free.
| Detail | Info |
|---|---|
| License | MIT |
| Built with | TypeScript |
| Use it for | Search Console and site health in one dashboard |
| Repository | crawlseo/crawlseo |
- Search Console integration
- Site crawler with a 2,000-page default limit per crawl
- Core Web Vitals monitoring
- Read-only MCP server for AI tools
- Optional keyword and backlink data with your own API key
8. SEO for Agent-Native Teams
This project is a Google Search Console MCP server and command-line tool. It doesn’t tell your AI agent what to do. Instead, it hands the agent facts.
Your agent gets your Search Console clicks, impressions and positions, plus visits from Google Analytics or Clicky. It also gets your Bing data, Google update dates and a crawl of your site. So the agent works from your own numbers rather than keyword-tool estimates.
| Detail | Info |
|---|---|
| License | Apache-2.0 |
| Built with | TypeScript |
| Use it for | Giving AI agents your real search data |
| Repository | iannuttall/seo |
- MCP server for AI agents
- Command-line interface
- Search Console, Google Analytics, Clicky and Bing data
- Google update dates for context
- Site crawl data
9. advertools
advertools is a Python library for online marketing and SEO analysis. It turns common SEO data into tables you can analyze with pandas.
For example, you can crawl a site, convert XML sitemaps and robots.txt files into tables, and parse server log files. Analysts who already work in Jupyter notebooks get the most from it. Log analysis is a big part of it, and our list of log file analysis tools covers the paid alternatives.
| Detail | Info |
|---|---|
| License | MIT |
| Built with | Python |
| Use it for | Scripted SEO analysis in Python and notebooks |
| Repository | eliasdabbas/advertools |
- Website crawler built on Scrapy
- XML sitemap and robots.txt parsing
- Server log file analysis
- URL and SERP data analysis
- Works with pandas DataFrames
10. Scrapy
Scrapy is a fast, high-level web crawling and scraping framework for Python. It isn’t an SEO tool out of the box, but many SEO crawlers, including advertools, are built on it.
Developers use it to check thousands of pages for missing tags, collect competitor data or monitor price and stock pages. It has one of the largest communities of any open-source crawler.
| Detail | Info |
|---|---|
| License | BSD-3-Clause |
| Built with | Python |
| Use it for | Building custom crawlers and scrapers |
| Repository | scrapy/scrapy |
- Fast, asynchronous crawling
- Custom spiders for any site structure
- Exports to JSON, CSV and XML
- Large plugin and middleware ecosystem
What About n8n and Other Workflow Tools?
Many SEO teams automate reports and alerts with n8n, which has more than 210,000 GitHub stars. However, n8n uses a fair-code license, not an OSI-approved open-source license. You can self-host it and read the code, but commercial use has limits. So check the license before you build a client service on it.
For more automation options, including paid tools, see our list of SEO automation tools and our content automation tools.
Matching Each Open-Source Tool to Your SEO Job
| If you need to… | Try |
|---|---|
| Replace a paid all-in-one SEO suite | OpenSEO |
| Track rankings cheaply | SerpBear |
| Stop speed regressions before launch | Lighthouse CI |
| Audit speed across a whole site | Unlighthouse |
| Crawl from the command line | SiteOne Crawler |
| Give a team a shared web crawler | LibreCrawl |
| Watch Search Console and site health together | CrawlSEO |
| Feed real data to AI agents | SEO for Agent-Native Teams |
| Analyze SEO data in Python | advertools |
| Build a custom crawler | Scrapy |
Costs to Plan For With Free SEO Software
Open source doesn’t always mean zero cost. Plan for these expenses before you switch.
- Hosting: A small VPS or cloud plan to run the app.
- Data APIs: OpenSEO and CrawlSEO need DataForSEO for keyword and backlink data, and SerpBear needs a scraping service.
- Maintenance: Someone has to apply updates and fix breaks.
- Setup time: Docker, databases and API keys take time to configure.
Even so, the trade-off often pays off. A team can replace several per-seat licenses with one server and a few API credits. Our SEO automation statistics show how widely teams already automate SEO work, and our open-source AI statistics show how fast open tools are growing.
Before You Self-Host an SEO Tool
What is the best open-source SEO tool?
OpenSEO is the closest open-source match to an all-in-one suite. For a single job, SerpBear handles rank tracking and Lighthouse CI handles speed checks.
Are open-source SEO tools really free?
The software is free, but you may pay for hosting and data APIs. For example, OpenSEO needs a DataForSEO key, and SerpBear needs a scraping service.
Can open-source SEO tools work with AI agents?
Yes. OpenSEO, CrawlSEO and SEO for Agent-Native Teams all include MCP servers, so agents like Claude Code can read your SEO data.
Is there an open-source alternative to Screaming Frog?
Yes. LibreCrawl and SiteOne Crawler are both free, MIT-licensed crawlers. advertools also crawls sites from Python.
Also See:
