Menu
Analyze website
KNOW WHAT’S BEING CHECKED

A careful look
at your website.

LimeSeekBot checks public HTML pages and produces evidence you can act on.

Included in this release

Speed and image insights

Compare crawler response times and HTML sizes for the checked pages. We flag responses over 2 seconds and HTML over 256 KiB as prompts for review. Image insights inspect img, srcset and picture markup for WebP/AVIF references, dimensions and loading hints. Unknown formats stay unknown. Counts include repeated image occurrences across pages.

Response time includes redirects and HTML download, not browser rendering. Image files are not downloaded, so compression quality, actual image bytes and the format ultimately selected by the browser are not measured. Native lazy-loading guidance applies to offscreen images; important initially visible images should load eagerly. Missing dimensions may already be handled in CSS.

Mobile and desktop homepage lab tests

At the end of each report, Google PageSpeed Insights supplies separate Lighthouse performance scores for the homepage on simulated mobile and desktop devices. See First Contentful Paint, Largest Contentful Paint, Total Blocking Time, Cumulative Layout Shift, Speed Index and suggested improvements. Scores vary between runs and do not change your LimeSeek technical health score.

Google receives the public homepage URL and loads the page and its resources independently. These two homepage tests use no extra scan credits or page slots. A blocked homepage is skipped. Missing measurements and provider errors are shown without failing the SEO report. The measurements are laboratory results, not real-user Core Web Vitals.

Deeper SEO insights

Saved scans also support action plans, internal-link structure, heading and schema reviews, approximate content similarity and changes between checked pages. Coverage is always limited to the available sample; zero observed incoming links does not prove a page is orphaned.

Connected Search Console data can identify search opportunities and possible overlapping landing pages. Separate optional integrations add Google indexing evidence, CrUX real-user metrics and lab tests for sampled page types. External keyword, competitor, backlink and local-ranking data can be imported from CSV, or collected through an administrator-enabled paid provider after the website owner opts in. These figures identify their source and distinguish estimates from observed data.

What needs another kind of check

The HTML crawler does not execute JavaScript, inspect your Google index status, measure real-user Core Web Vitals or fetch backlink and keyword datasets. Homepage Lighthouse tests, when available, run separately through Google. JSON-LD syntax checks do not establish rich-result eligibility. External link destinations and image resources are not fetched independently.

How crawling works

We stay on the selected domain and its www counterpart, respect robots.txt and pace bounded requests using a shared limit per website across scans. Temporary errors receive a limited number of retries. Unavailable pages are recorded and skipped; access blocks, excessive cooldowns or repeated website failures stop the scan with progress saved. Only public HTTP and HTTPS destinations are accepted. Large responses, nonstandard ports and private-network destinations are rejected.

Sitemap discovery and internal links help find pages. Each scan checks sitemap.xml and declared sitemap locations, follows nested sitemap indexes and skips repeated sitemap URLs. After sitemap discovery, full scans check the homepage, then pages marked for re-check, then a fresh random sample of other allowed sitemap URLs within the plan limit. Different scans may select overlapping pages; results cover the selected sample. The scan then checks unique allowed pages up to your plan’s page limit; requested pages beyond that limit stay pending for a later scan. Sampling considers the readable documents within a safety limit of 100 sitemap documents and 2 MiB per response. Internal page links can fill any remaining page allowance. Sitemap files do not count as tested pages. Large or compressed sitemap files may need separate validation. Reports show when the page limit is reached, when a URL cannot be checked and whether sitemap discovery succeeded. Page depth is measured within the discovered link graph; sitemap entries have no inferred navigation depth.

Automatic robots.txt checks

You do not need to add a robots.txt record or change your website to start. LimeSeek reads the existing rules automatically, including applicable Allow and Disallow entries. We use LimeSeekBot rules when present, otherwise the general rules for all crawlers. Allowed pages are scanned; disallowed pages are skipped. If robots.txt is absent or no applicable rule blocks a page, it can be checked. Missing robots.txt and sitemap files appear as advisory findings with suggested improvements. If HTTPS cannot connect, we try HTTP and check its robots rules before continuing; the report records the HTTPS failure and the actual fetched URLs. Temporary failures while reading robots.txt receive bounded automatic retries. If its rules remain unavailable or crawler access is denied, the scan stops for administrator review and retry.

Only add websites you own or have permission to analyze. Scan results belong to the customer workspace and are visible to authorized site administrators.

Turn findings into improvements

Open any error, warning or notice to see evidence and suggested steps. Where useful, we include a code example and official guidance. Use the external-link icon to inspect the page in a new tab. After making changes, choose Re-check to prioritize that page in your next scan. Marking a page uses no scan credit; starting the next scan uses the normal allowance.

Interpret findings in context

A noindex directive may be intentional. Duplicate pages may serve a useful purpose. Length and heading suggestions are advisory. A passing technical check does not guarantee indexing or rankings.