Loading…
Free SEO tool
Score, priorities, pages, evidence, and private exports.
The score appears after the crawl finishes.
Core SEO
Titles, descriptions, H1, canonical
Technical
Robots, sitemap, indexability, HTTPS
Content
Thin pages, image alt text, internal links
AI readiness
Crawlable answers, entities, sources, freshness
How the report works
Assetzaar checks the parts that usually block discovery first: titles, descriptions, headings, canonical URLs, robots rules, sitemap access, structured data, social previews, internal paths, crawl depth, image text, security headers, and crawler access. It also reports whether optional discovery files are present without treating them as ranking requirements. The free scan is deterministic, so it runs without spending AI tokens.
The paid mode is for teams that want a clearer plan after the crawl. The AI review reads the findings, groups the highest leverage work, and explains what a developer or marketer should change first. That keeps the free version useful while reserving token spend for deeper recommendations.
The report is built for sharing with founders, clients, and engineers. PDF provides a branded deliverable, Markdown is ready for tickets or documentation, CSV provides a page inventory, and JSON supports automations. All formats include the score, page list, priority buckets, AI-search readiness, and discovery evidence.
Status codes, redirect chains, index signals, canonical targets, robots rules, sitemap consistency, strict public-port handling, and bounded response evidence.
Readable headings, unique metadata, useful word count, internal paths, image descriptions, and social preview tags.
Crawlable answers, useful schema, entity signals, freshness cues, summaries, citations, and internal discovery paths.
Branded PDF plus CSV, Markdown and JSON exports with crawl topology, observed evidence, owner/effort labels, implementation recipes, and optional Pro AI recommendations.
Published methodology
Deterministic checks are scored from observable HTML, response headers, crawl paths and public discovery files. Failures with high impact carry more weight than warnings or informational findings. AI does not change the underlying crawl score; Pro uses the same evidence to explain priorities and implementation order.
Specialized checks activate only when the crawler detects relevant evidence. A site is not penalized for lacking ecommerce schema or framework signatures when those features do not apply.
| Surface | What is checked | Activation |
|---|---|---|
| Next.js / Vercel | Server-rendered metadata, canonical, links, platform signals and App Router fix recipes | When framework signals are detected |
| Ecommerce / Merchant | Product, Offer, image, price, currency, availability, policies and feed | When Product schema is detected |
| AI / answer engines | Direct answers, question headings, lists, tables, schema, trust and discovery files | Every public content page |
| Core search | Status, indexability, redirects, robots, sitemap consistency, canonicals, hreflang, headings and link topology | Every crawl |
Agency workflow
Use the branded sample to review the deliverable, or discuss an agency workflow for recurring audits and implementation support.
What makes this audit different
Open a finding to see the observed title length, canonical state, response headers, redirect chain, graph depth, schema types, or resource response behind it.
A compact baseline stays in the current browser. The next audit identifies new and resolved issues, changed titles or canonicals, indexability shifts, and material word-count changes.
When Next.js is detected, affected checks expose App Router metadata, sitemap, robots, rendering, and indexability implementation patterns rather than generic advice alone.
Product and Offer data, merchant policy evidence, crawler access and answer extractability activate conservatively without treating optional files or FAQ markup as ranking requirements.
Assetzaar will not publish a “state of SEO” benchmark until there are at least 100 eligible audits in a cohort. Published aggregates must exclude domains, URLs, page text and user identity; suppress any segment with fewer than 20 observations; and state the sample period and selection limits. Until those gates are met, no benchmark number is shown.
July 16, 2026
Added sampled crawl topology, per-check evidence, canonical and hreflang validation, sitemap consistency, redirect-chain and truncation diagnostics, private browser-local comparisons, Next.js fix recipes, and spreadsheet-safe CSV export. Empty decorative alt attributes and intentional secondary noindex pages are no longer treated as automatic failures.
July 15, 2026
Added an on-page synthetic sample, private Markdown and JSON exports, lifecycle analytics, clearer free-tool guarantees, and a reusable indexable tool-page shell. Removed llms.txt from the scored AI-readiness model because it is not a confirmed ranking or citation requirement.
July 14, 2026
Added Merchant product/feed checks, Next.js and Vercel detection, X-Robots-Tag handling, a downloadable sample report, published methodology, limitations and benchmark privacy gates.
Run it before publishing a new landing page, after a redesign, or when traffic drops and you need a fast baseline. It is also useful before sending a page to paid traffic because it catches missing titles, weak previews, blocked crawlers, and pages that are hard for answer engines to summarize.
The report is intentionally practical. It does not try to replace Search Console, analytics, or a full content strategy. It gives you a clean first pass, shows the highest-risk gaps, and turns the findings into a list that a builder can act on without reading a long export.
Use the free mode for routine checks and quick QA. Use the complete mode when you need context, plain-language reasoning, and a sharper order of operations for a team or client. Both paths keep private network targets blocked and avoid crawling unsafe URLs.
FAQ
An AI SEO audit combines a deterministic technical crawl (titles, canonicals, robots, structured data, internal links) with an AI review that reads the findings and explains what to fix first in plain language. On Assetzaar the crawl and report are free; the AI analysis is the optional Pro layer.
Yes. Every audit exports as a branded, client-ready PDF plus CSV, Markdown and JSON — with the score, prioritized fixes, per-URL evidence and implementation recipes — so agencies can hand the report to founders, clients or engineers as-is.
Yes. Free mode crawls up to 3 pages and runs deterministic SEO, technical, structured data, content, and AI-search readiness checks without spending AI tokens.
Only the complete AI audit calls the model. The free audit uses code-only checks, while the AI audit adds an executive summary, prioritized fixes, developer notes, and a deeper branded report.
It checks whether public pages are crawlable and easy to interpret: direct answers, question headings, useful schema, visible ownership, freshness, internal links, summaries, and citation paths. Optional llms.txt files are reported separately and do not affect the score.
Yes. The report downloads as a generated PDF with Assetzaar branding, score, priorities, discovery files, page details, AI-search readiness, and AI recommendations when available.
Yes. The audit rejects localhost, private IP ranges, credentialed URLs, non-HTTP protocols, excessive redirects, oversized responses, and slow requests.
Yes, when Product schema is detected. It checks product images, Offer data, price, currency, availability, policy data, and whether a conventional Merchant feed endpoint is available. Google policy and account eligibility still require separate review.
Yes. The audit form includes a downloadable synthetic sample PDF clearly labelled as an example, so teams can inspect the deliverable without submitting a URL.
Yes. After the first live audit, Assetzaar stores a compact baseline in that browser only. A later audit highlights score changes, new and resolved issues, added or missing sampled URLs, title and canonical changes, indexability shifts, and material word-count changes. The baseline can be cleared from the report.
No. The map covers only the bounded crawl sample. A page with no observed inlink is labelled as a candidate for review, not a proven orphan. A full-site crawl, sitemap comparison, Search Console, and server logs are needed for broader conclusions.