Best Open-Source SEO Platforms and Tools (2026)

Every paid SEO platform is, underneath the dashboard, a combination of three things: a crawler, a database of other people's search data, and a scheduler that runs both on a loop. Open source covers the first and third completely. The second one, the part where someone already crawled the entire web and indexed billions of backlinks and keyword volumes, is the part no community project can replicate on a volunteer budget. That's not a knock on open source tools. It's just where the line actually sits, and most "free SEO tools" roundups blur it.

This list doesn't blur it. Every tool below is genuinely open source: you can read the code, fork it, and run it without paying anyone, under a license that says so (not a "free tier of a commercial product," which is a different thing covered in our free AI SEO tools roundup). We checked each project's GitHub repository directly this month for license, commit activity, and release dates, because an unmaintained open source tool is worse than no tool: it still runs, it just quietly falls behind as Google changes how it crawls and renders pages. A couple of projects that show up in older "open source SEO tools" lists didn't make the cut here because they've gone stale or the repo people keep linking to no longer exists in any current form. We'll say exactly which, and why.

If you're automating the parts of SEO that don't need third-party data (crawling, local Lighthouse runs, your own analytics, your own WordPress meta tags), this is a real toolkit. If you're hoping to replace Ahrefs or Semrush's keyword and backlink databases for free, skip to the section near the end that explains why that specific job doesn't have an open source answer yet.

Rank tracking

[SerpBear](https://github.com/towfiqi/serpbear) is a self-hosted keyword rank tracker. You run it as a Docker container or a Node app, point it at your domains and keywords, and it checks Google positions on a schedule, with email alerts when a keyword moves. License is MIT, the repo has 2,100+ GitHub stars, and it was last pushed in May 2026 with a tagged release (v3.1.0) from March 2026, so it's maintained, not abandoned.

The catch, and it's a real one: SerpBear doesn't scrape Google itself. It needs a third-party SERP API (ScrapingRobot, SerpApi, SearchApi, or similar) or your own proxy pool to actually fetch ranking data, and those cost money per lookup once you're past a small free tier. So "free rank tracker" is accurate for the software; it's not accurate for the SERP data feeding it once you're tracking more than a handful of keywords on a schedule. It suits someone comfortable running a Docker container who wants a private rank-tracking dashboard without a monthly SaaS seat, and who's already budgeted a few dollars a month for the scraping API.

Crawling and site audits

[SEOnaut](https://github.com/stjudewashere/seonaut) is a self-hosted site auditor, written in Go with a MySQL backend, that crawls a site and reports broken links, redirect chains, duplicate or missing meta tags, and heading order problems, sorted by severity. MIT licensed, 800+ stars, last commit in May 2026. There's also a hosted version at seonaut.org if you don't want to run it yourself, but the self-hosted path is the open source one. Limitation: the issue list is solid but narrower than a commercial crawler's. There's no built-in JavaScript rendering comparison or content-quality scoring, so it's an audit tool, not a replacement for Screaming Frog's full feature set.

That's where [LibreCrawl](https://github.com/PhialsBasement/LibreCrawl) picks up the slack. It's a web-based crawler built with Python Flask, explicitly positioned as a free alternative to Screaming Frog, Sitebulb, and DeepCrawl, with multi-tenant support so more than one person can run crawls against the same instance. MIT licensed, 1,000+ stars, actively pushed through September 2026. It's the newer and less battle-tested of the two, no tagged release history yet on GitHub, so treat it as promising rather than proven if you're putting it in front of a client. This replaces Greenflare in this list: Greenflare used to be the go-to open source desktop crawler, but its GitHub presence has effectively disappeared, the repos that still use that name are either abandoned, unrelated, or sitting at zero stars with no real activity. If you've bookmarked Greenflare from an older article, check before you install it.

[Unlighthouse](https://github.com/harlan-zw/unlighthouse) runs Google Lighthouse across every page of a site at once instead of one URL at a time, which is the actual gap in Lighthouse's own tooling. Point it at a domain or sitemap and it crawls, scores, and builds a dashboard of performance, accessibility, and SEO scores site-wide. MIT licensed, nearly 4,900 stars, pushed within the last week as of this writing, with a release (v0.19.1) from early October 2026. It's a command-line tool (npx unlighthouse --site yourdomain.com), so it suits developers more than marketers working alone, and a full crawl of a large site takes real time and CPU since it's launching a headless Chrome instance per page.

Performance testing

[Google Lighthouse](https://github.com/GoogleChrome/lighthouse) is the reference implementation Google itself ships inside Chrome DevTools, PageSpeed Insights, and the Chrome extension. Apache 2.0 licensed, nearly 31,000 stars, commits landing the same week we checked. You can run it from the command line, as a Node module, or through the GUI in Chrome, and it's the same scoring engine every other performance tool on this list either wraps or compares itself against. The limitation isn't the tool, it's the single-run nature of it: Lighthouse scores vary run to run on the same page depending on network and CPU conditions at that moment, so a single Lighthouse score is a snapshot, not a trend line.

[sitespeed.io](https://github.com/sitespeedio/sitespeed.io) is the trend line. It's a full performance-testing suite that runs real browsers (not just headless Chrome) against your pages on a schedule, stores historical results, and can alert when metrics regress. MIT licensed, over 5,000 stars, a release (v42.7.0) from September 2026. Run it via Docker, and it genuinely does continuous monitoring, which is the thing a single Lighthouse run can't do. It suits teams who want to track Core Web Vitals over time without paying for a SaaS monitoring dashboard; the setup for historical graphing (usually paired with Graphite or InfluxDB) takes more work than installing a single npm package.

Analytics

[Matomo](https://github.com/matomo-org/matomo) is the closest thing to a full Google Analytics replacement that's actually open source end to end, not just the client-side script. Self-host it on your own PHP and MySQL stack (there's also a paid Matomo Cloud option if you don't want to run it yourself), and you get full visitor-level data ownership, no sampling, and a permissive approach to what you track, since it's your server. GPL-3.0 licensed, nearly 22,000 stars, commits the same day we checked, with a release (5.14.1) from early October 2026. It suits anyone who wants analytics depth closer to GA4's reporting breadth, not just a privacy-friendly pageview counter, and who's willing to maintain a PHP app and database.

[Plausible](https://github.com/plausible/analytics), specifically the self-hosted Community Edition, trades Matomo's depth for simplicity: cookieless, lightweight tracking script, one clean dashboard, no cross-site or event-level drill-down the way Matomo offers. AGPL-3.0 licensed, over 29,000 stars, pushed within the last day as of this writing, latest tagged release (v3.2.1) from May 2026. Worth knowing before you self-host it: Plausible's paid Cloud product has some features (like funnels and certain integrations) that aren't in the free Community Edition, so "open source Plausible" and "Plausible the product" aren't fully identical feature sets. Fine for most small sites that just want "how many people visited and from where" without Google in the loop.

[Umami](https://github.com/umami-software/umami) sits between the two: simple like Plausible, but with more built-in detail (campaigns, revenue tracking, custom events) without Matomo's full configuration surface. MIT licensed, over 39,000 stars, the most-starred tool on this entire list, commits the same day we checked, release v3.4.0 from mid-September 2026. Deploy it with Docker or on Vercel in minutes. The limitation is less about the software and more about scope: it's an analytics tool, not an SEO crawler or rank tracker, so it answers "what happened on my site" and nothing about why you rank where you do in search.

Log analysis and bulk data work

[advertools](https://github.com/eliasdabbas/advertools) is a Python library, not an app with a dashboard, built for SEO and marketing data work at scale: crawling, robots.txt analysis, XML sitemap parsing, SERP data wrangling, and log file analysis, all returning pandas DataFrames you can filter and chart yourself. MIT licensed, nearly 1,500 stars, a PyPI release (0.18.0) from June 2026. It suits someone who already knows Python and wants to analyze a 2-million-line server log for crawl-budget waste or build a custom report pandas can't get from a GUI tool. The limitation is the obvious one: there's no interface. If you don't want to write Python, this isn't your tool, and the learning curve is real even for people who do.

All-in-one self-hosted suites

[SEO Panel](https://github.com/sendtogeo/Seo-Panel) bills itself as the original open source, self-hosted "SEO control panel," bundling rank tracking, backlink monitoring, and reporting for multiple sites into one PHP app you install on your own server. GPL-2.0 licensed, 138 stars. Here's the status check the brief for this piece specifically asked us to run: the GitHub mirror most people link to hasn't had a commit since September 2023, which is over three years stale as of October 2026. It still installs and technically still works, PHP/MySQL apps from that era don't just stop running, but there's no evidence of active maintenance keeping it current with how search engines serve results today. Treat it as a legacy option you might inherit on an old server, not a tool to newly adopt in 2026.

WordPress SEO plugins (GPL)

These four are properly open source in the sense that matters for a WordPress plugin: GPL-licensed, free to use indefinitely, with paid tiers that add features on top rather than gate the core plugin behind a paywall.

[Yoast SEO](https://github.com/Yoast/wordpress-seo) (free) is GPLv3 per its own readme, over 2,000 stars on its official GitHub mirror, with commits the same week we checked and a tagged release (28.6) from late September 2026. The free version handles meta tags, XML sitemaps, readability and SEO content analysis, and schema output. Yoast SEO Premium adds redirect management, internal linking suggestions, and multiple focus keyphrases per page, paid separately. The free plugin's limitation is well known among anyone who's used it: the content analysis is a useful checklist, not a ranking guarantee, and green bullets don't mean the page will rank.

[Rank Math](https://github.com/rankmath/seo-by-rank-math) (free) is GPLv3 or later per its readme, with an official GitHub mirror showing commits as recently as the same week we checked. The free tier is genuinely generous, schema types, Google Search Console integration, and redirect management are included that some competitors reserve for paid tiers, which is Rank Math's whole pitch. Rank Math Pro adds more advanced schema, a content AI assistant, and higher usage limits on things like the free tier's module count, paid separately. Note that Rank Math ships its actual releases through the WordPress.org plugin repository (SVN), not GitHub Releases, so the GitHub repo is better read as a mirror for code review than a release feed.

[The SEO Framework](https://github.com/sybrew/the-seo-framework) (free) is GPL-3.0, nearly 500 stars, commits the same day we checked, with a tagged release (5.1.4) from December 2025. Its whole design philosophy is the opposite of Yoast and Rank Math: no visual scorecards, no nagging upsell notices, it sets smart defaults and gets out of the way. Extensions (paid, sold separately through the developer's site) add things like local SEO and AMP support. It suits developers and agencies who find the other two plugins' admin screens noisy, and the limitation is the mirror image of that strength: if you want hand-holding and a visual checklist, this plugin deliberately doesn't give you one.

[Slim SEO](https://github.com/elightup/slim-seo) (free) is GPL v3, under 100 stars but actively pushed as recently as this month, positioned as the lightest-weight option of the four: automatic meta tags, sitemaps, and schema with close to zero settings screen. There's a Slim SEO Pro addon bundle sold separately for things like 404 monitoring and schema customization. It suits sites where every extra database query and settings screen is unwanted overhead, and the limitation is the same lightness that's the selling point: less manual control than Yoast or Rank Math for anyone who actually wants to hand-tune every page's output.

Comparison table

A sample open-source SEO stack for a small site

Say you're running one WordPress site and don't want a single paid SEO subscription. A reasonable stack looks like this: Rank Math free handles your on-page meta tags, sitemap, and schema since it's already running inside WordPress. Matomo or Umami, self-hosted on the same server or a small VPS, replaces Google Analytics for traffic and conversion tracking. SEOnaut or LibreCrawl, run occasionally (weekly or monthly, not continuously) against your own domain, catches broken links and redirect chains before Google's crawler does. Unlighthouse gives you a full-site Core Web Vitals pass before and after any theme or plugin change, instead of spot-checking one page at a time in PageSpeed Insights.

What that stack costs you isn't money, it's server resources and your own time running updates and reading dashboards nobody's paying to simplify for you. For a hobby site or a small agency testing whether the SEO automation habit is worth building at all, before committing to a paid platform, that's a fair trade. For an agency managing twenty client sites, the time cost of running and patching five separate self-hosted apps usually exceeds what a consolidated paid platform would have charged. Our SEO automation tools roundup covers that paid side of the equation, where the tools are priced specifically to remove that maintenance burden.

What open source can't replace

None of the tools above include a backlink index. Ahrefs and Semrush's core value isn't the dashboard, it's having already crawled a meaningful share of the web and stored who links to whom, which costs millions of dollars in crawling infrastructure to build and keep current. No open source project maintains that at comparable scale, because nobody's donating the server farm it would take.

The same gap applies to keyword search volume. Volume numbers come from licensed clickstream and search-engine data (Google's own Keyword Planner data, third-party clickstream panels), not from crawling websites, so there's no open source equivalent to scrape together. advertools can pull live SERP results if you feed it a scraping method, but "what ranks right now" and "how many people search this term per month" are different data problems, and only the second one is genuinely unavailable without paying someone who licensed it. Our guide on spying on competitor keywords goes through where that paid data actually comes from and which tools license it fairly.

If your plan is "run entirely on open source tools and never pay for SEO data," the honest limitation is that plan works for technical SEO, site monitoring, and your own analytics, and it does not work for competitive keyword or backlink research. Budget for one paid data source even if everything else in your stack is free.

Several of these tools now pair well with AI assistants through MCP. See our list of SEO MCP servers, including Google's open-source Analytics server.

If you outgrow free tools, the two paid suites most teams compare are covered in our Ahrefs review and Semrush review.

FAQ

Is open source SEO software actually free to use commercially?

Yes, for every tool listed here. MIT, Apache-2.0, GPL, and AGPL all permit commercial use. GPL and AGPL add conditions if you modify and redistribute the software itself (you have to share your changes under the same license), which matters if you're a developer shipping a derivative product, not if you're just running the tool for your own site.

What's the difference between "open source" and "source available"?

Open source means the license grants you the right to use, modify, and redistribute the code, under terms like MIT, GPL, or Apache. Source available sometimes just means you can read the code on GitHub with no such rights attached, or the license restricts commercial use or redistribution. Every tool in this piece carries a real OSI-style license (MIT, Apache-2.0, GPL, or AGPL), which is why we could verify it directly from the repository rather than taking a marketing page's word for it.

Can I run these tools without any coding experience?

Some, yes. Yoast SEO, Rank Math, The SEO Framework, and Slim SEO install like any WordPress plugin, no code required. Matomo, Plausible, and Umami are straightforward with Docker even if you've never written code, since the install instructions are copy-paste commands. SerpBear, SEOnaut, LibreCrawl, and especially advertools assume comfort with a terminal, Docker, or Python respectively.

Why isn't Greenflare on this list anymore?

Because we couldn't verify an actively maintained GitHub repository for it. The name still shows up in older open source SEO roundups, but the searches we ran turned up either abandoned forks, unrelated repos reusing the name, or projects with zero meaningful activity. LibreCrawl and SEOnaut cover the same crawling and auditing job with repos we could confirm are current.

Should I self-host analytics or just use a free tier of a paid tool?

Depends on what you're optimizing for. Self-hosting Matomo, Plausible, or Umami gets you full data ownership and no sampling, at the cost of running and patching a server yourself. A paid tool's free tier gets you zero maintenance, at the cost of usage caps and someone else holding your data. If GDPR-style data residency or avoiding third-party cookies matters to your business, self-hosted open source analytics is the more defensible choice, not just the cheaper one.