SEO Analyzer (On-Page Audit + Score)

Run a full on-page SEO audit on any page. Enter a URL and the analyzer fetches it server-side (so cross-origin pages work), or paste raw HTML to keep everything in your browser. It runs 40+ checks across indexing, meta tags, social cards, headings, links, images, structured data and technical signals, then gives you a 0–100 score with the exact problems to fix first and a link to the Codeswap tool that fixes each one.

How to use the SEO Analyzer (On-Page Audit + Score)

There are two ways to run an audit:

• Analyze a URL — type any address and the analyzer fetches the live HTML through our server (browsers cannot read cross-origin pages directly, so a server fetch is required). This unlocks the technical checks too: HTTP status, HTTPS, redirect chains, compression and mixed content. Optionally add a target keyword to test whether it appears in the title, meta description and H1, and to measure keyword density.
• Paste HTML — copy a page's source (right-click → View Page Source, or any saved HTML) and paste it. This mode runs entirely in your browser and is perfect for auditing a draft from your CMS before it goes live, or any HTML you do not want to send anywhere.

Either way you get a weighted score out of 100, a letter grade, and every check grouped into Indexing & Meta, Social Sharing, Content & Headings, Links, Images, Structured Data and Technical. Critical problems (missing title, noindex, no H1, missing alt text) cost the most points; warnings cost half. Each issue links to the dedicated Codeswap tool that fixes it, and you can copy the report as text or download it as JSON.

What an on-page SEO analyzer actually checks

On-page SEO is everything inside a page's own HTML that helps search engines understand and rank it. An analyzer parses that HTML the way a crawler does and surfaces the signals that matter. The most heavily weighted are the ones that decide whether a page can rank at all:

• Title tag — the single strongest on-page signal and the clickable headline in search results. It should be unique, 30–60 characters, and describe the page.
• Meta description — not a ranking factor directly, but it is usually the snippet under your title, so it drives click-through rate. Aim for 120–160 characters.
• Indexability — a stray noindex in a meta tag or the X-Robots-Tag response header silently keeps a page out of Google entirely. This is the first thing to rule out when a page will not rank.
• Canonical URL — tells search engines which version of a URL is authoritative, preventing duplicate-content dilution.
• Headings — exactly one H1 stating the topic, with H2–H6 forming a logical, non-skipping outline, signals a well-structured page to both crawlers and screen readers.
• Open Graph & Twitter Cards — control how the page looks when shared on social platforms and in chat apps. Missing tags mean ugly or blank previews and fewer clicks.
• Structured data (JSON-LD) — makes a page eligible for rich results (stars, FAQs, breadcrumbs). Invalid JSON-LD is worse than none, so validity is checked too.
• Images — missing alt text hurts accessibility and image search; missing width/height causes layout shift (a Core Web Vitals problem).

The technical group adds what only a real request can reveal: the HTTP status code, whether the final URL is HTTPS, how many redirects sit in the way, whether the server compresses its response, and whether an HTTPS page loads insecure http:// resources (mixed content, which browsers block). Together these checks approximate what a search-console crawl plus a Lighthouse SEO audit would tell you, in one pass.

Common use cases

• Pre-publish QA — paste a draft from your CMS and catch a missing title, thin content or an accidental noindex before it ships.
• Auditing a competitor — enter a competitor's URL to see exactly which on-page signals they get right, and where the gaps are.
• Debugging "why won't this page rank?" — the indexability and canonical checks instantly reveal the silent killers that keep a page out of search.
• Fixing broken social previews — the share-card preview shows what Slack, LinkedIn and X will display, so you can fix Open Graph tags before a launch.
• Client & handoff reports — copy the text report or download the JSON to attach to a ticket or a deliverable.
• Migration checks — after a replatform, batch-check key URLs for lost meta tags, broken canonicals or new redirect chains.

How the 0–100 score is calculated

Every check carries a weight that reflects its real ranking impact. The score is the percentage of available weight you earn: a pass earns the full weight, a warning earns half, and an error earns zero. Purely informational rows (link counts, readability, microdata detection) carry no weight and never move the score. The grade bands are:

• A (90–100) — excellent; only minor polish left.
• B (80–89) — good; a few warnings worth clearing.
• C (70–79) — average; real issues are costing you.
• D (55–69) — weak; several important signals are missing.
• F (under 55) — critical problems such as a missing title, no H1, or a noindex directive.

Because the heaviest weights sit on the make-or-break signals, you cannot score well while a critical issue is open — which is exactly how a search engine would treat the page.

Privacy and how the URL fetch works

In Paste HTML mode nothing leaves your browser at all — the parsing and scoring run locally in JavaScript, consistent with every other Codeswap tool.

In Analyze a URL mode the page is fetched once by our server, because browser security (CORS) prevents a web page from reading another site's HTML directly. That fetch is deliberately locked down: only http and https URLs on ports 80/443, with the target host re-validated at every redirect and blocked from resolving to private, localhost, or cloud-metadata addresses. Responses are capped at 2 MB with short timeouts, and the request is rate-limited. We do not store the URL, the fetched HTML, or the report — the HTML is analyzed in your browser and discarded when you close the tab.