The Best Open Source SEO Tools in 2026
Five open source SEO projects worth running in 2026 — what each one actually does, who it suits, and what it costs to keep running.
Most open source SEO projects on GitHub are weekend demos. A few are tools you can put in front of a client. This is the short list for 2026, and we keep it current: what each one does, who it fits, and what it costs to keep running.
One thing to understand before you pick anything. Almost nothing in SEO is free to operate, and open source does not change that. Rank tracking needs SERP results from many locations. Keyword research needs volume estimates. Backlink data needs an index nobody can build alone. Somebody pays for those requests. The difference between the projects below and a legacy suite is not that the data becomes free. It is that you can see the code, run it yourself, and swap the data provider when you find a better price.
Disclosure first
We publish this guide, and Doverity is on it. That is not a neutral position, so here is the shape of it: Doverity is built on open-seo, an MIT-licensed SEO stack. The core is public, the license lets you run it without us, and nothing on this page requires buying anything from us.
We wrote the rest of this page to be useful even if you never open our app.
Doverity
Doverity is the widest tool here. Keyword research, rank tracking, backlink profiles, site audits, AI visibility, and a prompt explorer all live in one app. It fits a small team or an agency that wants one tool instead of four. SEONaut has a hosted version too. The difference is that Doverity's hosted app is the whole platform — research, tracking, and audits in one place — not only the crawler. The feature list is long, so start there rather than here.
Under the hood it pulls from DataForSEO, a pay-as-you-go data provider. That is the real cost of running it, and it is metered rather than bundled, so you pay for what you pull. The pricing page spells it out.
The part that separates it from the others: Doverity ships an MCP server and a set of agent skills. You can put Claude Code, Codex, or Cursor on the tedious work — a first pass of keyword research, an audit triage — and then read the results in the app. That is not the same as pointing an agent at a raw data API. The agent can save work back to your account, like keyword tags for clustering, and hand you a link to the same data it looked at, so you are not taking its summary on faith.
SerpBear
SerpBear does one thing: rank tracking. Unlimited keywords, unlimited domains, self-hosted in Docker.
You bring a SERP provider and pay that provider per request, so your bill tracks how many keywords you check and how often. It supports several providers with different price and coverage tradeoffs, which matters more than it sounds — this is the line item that decides whether the tool is cheap for you. It also reads Google Search Console for free and has a keyword research feature, both of which pair naturally with tracking.
One warning on that research feature. It leans on the Google Ads API, which requires an application and approval before you can call it. Budget an afternoon for the paperwork, not five minutes.
Pick SerpBear if rankings are the only number you actually check. It will not audit your site or profile your backlinks.
SEONaut
SEONaut is a self-hosted crawler for technical SEO. Point it at a domain and it walks the site, then reports what it found: broken links, missing or duplicated titles, redirect chains, thin pages, indexing problems.
It runs as a web app in Docker, which makes it a reasonable fit if more than one person needs to use it, and there is a hosted version if you would rather not run anything. It needs no paid data provider, so it is the cheapest category of tool on this page to operate — your costs are a server and your time.
Pick it when you need to audit sites repeatedly and want the results to live somewhere your team can reach.
LibreCrawl
LibreCrawl covers the same ground from a different angle. It is a crawler that runs on your desktop or in a browser, reports on links, SEO elements, and page performance, and exports the data. It handles JavaScript rendering and large crawls, and it is newer than SEONaut and still moving quickly.
Because it runs locally, there is no server to maintain and nothing leaves your machine. That makes it the easier first install if you just need to look at one site today, and the better option if the sites you audit belong to clients who would rather you not upload them anywhere.
Pick it over SEONaut if you work alone, or if local-only is a requirement. Pick SEONaut if you want a shared, always-on report.
SEOMachine
SEOMachine is not a data platform. It is a Claude Code workspace for producing long-form blog content — research, outline, draft, optimize — with commands for each step and a set of agents that handle the passes.
You run it by cloning the repo, installing its Python dependencies, adding context about your business, and opening it in Claude Code. It works with an Anthropic API account, and it can optionally read Google Analytics, Search Console, and DataForSEO so the writing is anchored to real numbers instead of guessed ones.
Pick it if your bottleneck is producing content, not pulling data. It solves a different problem from everything above it on this page, and it pairs well with one of the crawlers or a rank tracker rather than replacing them.
Smaller projects worth a look
These are narrower than the five above. Some are useful today, some are worth watching, and none of them will replace a full stack.
- openserp is a self-hosted SERP API in Go. It returns structured results from Google, Bing, Yandex, Baidu, DuckDuckGo, and Ecosia, and runs in Docker. If you are building your own rank tracker, or you want an agent to be able to pull live results without a paid API key, this is the plumbing.
- elmo tracks how often AI answer engines mention and cite a brand, self-hosted via Docker Compose. If you want to watch ChatGPT, Claude, and Gemini without buying an enterprise AEO platform, start here.
- SEO Panel has been around since 2010: a PHP control panel for rank tracking, audits, sitemaps, and backlink monitoring across many sites, with multi-user accounts. The interface shows its age. If you manage a lot of small sites and want one self-hosted place to keep an eye on them, it still does the job.
How to judge anything else
There are dozens more open source SEO projects on GitHub, most of them single-purpose crawlers and scripts. Before you invest an afternoon in one, check three things.
Is there a license? No LICENSE file means you have no right to use it, and no way to tell whether that changes.
When was the last commit? Google changes its results page and its APIs constantly. A crawler that has not been touched in two years is usually a crawler that returns broken data.
What does it cost to run? Read the README for the data source. If a project needs a SERP API or a backlink index and does not say so up front, you will find out from the bill.
A related note, because it comes up on other lists: check that a tool actually is open source. Some products publish SDKs for their API and call themselves open source. Look for the repository with the application in it, not just a client library.
Where to start, and what to send us
If you would rather not run four tools, start with Doverity — hosted, or self-hosted on Docker or Cloudflare Workers.
If you would rather assemble a stack from the projects above, that works too, and it is most of the point of open source.
This page gets revisited. If a project belongs on it and is not here, email support@doverity.com with what it does and why you use it, and we will put it through real work before adding it.