Free tools · no signup

Free SEO tools

Three free SEO tools backed by the open-source pseolint engine — SpamBrain checker, thin-content scanner, doorway-page detector. $0, no signup, runs in a 60-second median. Each tool now produces a per-template verdict, not a per-URL list. Pick a tool below; methodology and comparison table follow.

Credibility layer:the engine has been empirically calibrated against a curated corpus of reputable, in-production pSEO sites. It audits by template — it detects your site's URL templates, samples pages stratified across them, and produces one verdict per template cluster rather than a flat per-URL list. Verdicts are reproducible at a fixed sample-seed; findings cluster instead of dumping per-pair noise; severity demotions are auditable via summary.appliedSeverityDemotions. Dated snapshot results, the open-source corpus, and the trade-offs we accepted are documented at /methodology.

What these tools are for

Programmatic-SEO sites (template-driven content at scale) and AI Overview readiness — SpamBrain triggers from the March 27, 2026 core update that tightened scaled-content signals on date-stacked corpora (the most recent classifier shift to demote pSEO sites), the May 7, 2024 site-reputation-abuse policy (now detected by links/host-section-divergence), the March 5, 2024 scaled-content-abuse update, plus the AEO patterns that determine ChatGPT, Perplexity, and Google AI Overview citations.

What they aren't

A general SEO audit. For Core Web Vitals use PageSpeed Insights. For broken links + general crawl use Sitebulb ($35/mo) or Screaming Frog ($259/yr). For competitor / backlink data use Ahrefs ($129/mo) or Semrush ($139.95/mo).

How rules feed into per-template verdicts

The engine audits by template, not by URL. After template detection (Phase 1), the engine samples pages stratified across templates and runs 32 rules (Phase 2). Rules aggregate differently depending on their type:

  • Per-page → template uniformity score. spam/thin-contentfires on each sampled page and feeds the template's uniformity score (0–1). A score ≥0.8 means ≥8 of 10 sampled pages are thin — that template gets a critical verdict regardless of the other templates' health.
  • Corpus-wide (not template-scoped). spam/near-duplicate is computed across the full sampled corpus — it compares pages across templates to surface cross-template duplication, not just within a single template.
  • Per-page → template-level signal. aeo/citable-facts fires per sampled page; an 80% fire rate on a template (8/10 pages lack citable facts) becomes a template-level finding reported in the template card, not a list of 8 individual URL findings.
  • Site verdict. The site verdict is the worst-performing template that covers ≥5% of URLs (spec §15.1 siteVerdictFromTemplates). A healthy landing page template cannot mask a broken product-listing template if that listing template holds most of your URLs.

These three tools each foreground a different slice of that rule surface — but the underlying per-template aggregation runs in every one.

What these tools actually detect

Programmatic SEO sites get demoted for a small, well-defined set of reasons, and almost all of them trace back to Google's SpamBrain system — the machine-learning spam classifier Google rolled out in 2018 and re-architected to score pages, not just links. The same classifier drives scaled-content-abuse demotions and powers site-reputation-abuse enforcement against parasite-SEO landing pages. If your site is templated, your audit needs to mirror how SpamBrain reasons about templates — and that is exactly what pseolint does: it audits your templates, not your URLs.

The pseolint v0.7.4 engine ships 47rules across spam, content, aeo, links, schema, tech, data, and cannibalization categories. Each rule's findings aggregate per template — you get one verdict per template cluster, not a flat per-URL list. The spam rules map directly to documented SpamBrain signals — thin content, doorway patterns, near-duplicate clusters, boilerplate ratio, template diversity, scaled-content density, and link-spam detectors. The aeo rules cover answer-engine readiness for AI Overviews, Perplexity, and ChatGPT search.

The whole engine is MIT-licensed on GitHub, so anything you run in the browser here you can also run locally against pre-deploy builds, in CI, or via the Model Context Protocol server. The JSON report contains every finding, every rule reference, and a deterministic severity score between 0 and 100.

How it compares to hosted SEO crawlers

Most pSEO teams already pay for at least one general-purpose crawler. The table below lines up the lowest entry-tier price for the four tools we hear most often in customer interviews — Ahrefs, Semrush, Sitebulb, ContentKing — alongside Screaming Frog (the desktop incumbent) and pseolint. Numbers are list price as published on each vendor's pricing page; per-seat surcharges and annual-billing discounts are excluded for parity.

ToolEntry priceAudit limit (entry tier)SpamBrain-aware?
pseolint (this site)$0/monthup to 200 pages, stratified across templates, 3 audits/day per browserYes — template-aware engine; per-template verdicts
Ahrefs$129/month (Lite plan)10,000 crawl credits/month, 1 projectNo — generic site-audit ruleset
Semrush$139.95/month (Pro plan)100,000 pages/month across 5 projectsNo — generic site-audit ruleset
Sitebulb$35/month (Lite plan)1 project, 10,000 URLs per auditPartial — thin-content + duplicate detection
Screaming Frog$259/year (paid license)Free up to 500 URLs, then unlimited per licenseNo — desktop crawler, raw signals only
ContentKing$44/month (Basic plan)500 pages monitored, 1 userPartial — change-tracking + on-page audit

The honest read: pseolint is not a replacement for Ahrefs or Semrush if you need backlink data, keyword volumes, or rank tracking. It is a replacement for the "site audit" module those suites bolt on, specifically for templated programmatic content. The engine shipped its v0.4.0 redesign on April 29, 2026, added change-driven monitoring in v0.5.0 (May 1, 2026), shipped the links/host-section-divergence site-reputation-abuse detector in v0.5.1 (May 3, 2026), the v0.5.2 credibility layer (May 3, 2026), and the template-aware engine — auditing by template rather than by URL, producing one verdict per template cluster via siteVerdictFromTemplates. Runs as a Cloudflare R2 + Inngest pipeline on Vercel, MIT-licensed end-to-end so anything you see in the browser audit is reproducible from the CLI.

All three tools share the same backend audit — the difference is which subset of rules each one foregrounds. If you want the full report (every rule, every page) just use the homepage audit instead.

Behind the scenes, each invocation runs an asynchronous Inngest job that performs sitemap discovery, stratified URL sampling, parallelized HTTP fetching with backpressure-controlled concurrency, headless Playwright rendering for SPA-shaped origins, semantic boilerplate detection, SimHash 64-bit fingerprinting for near-duplicate clustering, and severity-graded finding enrichment. The MIT-licensed CLI mirrors the same pipeline locally — pipe its JSON output into your CI gate or your editor's diagnostics panel via the Model Context Protocol adapter.

The three tools cover three distinct SpamBrain-risk surfaces: the SpamBrain checker targets scaled-content-abuse and thin-content signals across your full template set, the thin-content scanner isolates the spam/thin-content and content/boilerplate-ratio rules per template, and the doorway-page detector maps to the site-reputation-abuse policy enforced since May 7, 2024. Each tool samples pages stratified across detected templates and returns a per-template verdict — not a flat per-URL list — so triage goes straight to the template responsible rather than to hundreds of individual URLs.

A worked example: Halverwood Cabinetry pointed all three tools at its 2,100-page /cabinet/{wood}-{finish} catalog. The SpamBrain checker returned a domain risk of 57; the thin-content scanner isolated 612 pages under the 300-word floor with a median body of 188 words; the doorway-page detector clustered 140 URLs at a 0.90 SimHash and flagged the shared quote-request funnel they all pointed to. Operator Renske Adelaar rebuilt the worst template with per-finish drying times, a measured hardness rating, and a price band, and a re-run two weeks later dropped the domain risk to 24.

Sources