Documentation

Shumway SEO Toolkit v0.1.9 docs

Use this page to find the right workflow, understand what each screen exports, and check the browser differences that affect Shumway SEO Toolkit v0.1.9.

For release history, see the changelog.

Common Workflows

1. Open the page you want to inspect.
2. Click the Shumway SEO Toolkit browser action.
3. Open the workspace from the extension action or context menu for deeper review.
4. Copy individual fields or export datasets when you need to share or archive the review.

The extension reads the active tab on demand. It is designed for normal HTTP and HTTPS pages and does not run a background crawler.

Feature Reference

The sections below match the main workspace navigation and the major data surfaces available for export.

Export Reference

Exports are generated locally in the browser from the current page audit. Quick Exports focus on one screen, while the Exports screen can combine selected reports.

• Datasets include issues, metadata, technical, discovery, HTTP trace, Web Vitals, headings, links, images, schema, robots, social, style, SERP, tools, accessibility, and support data where available.
• Formats include CSV, TSV, JSON, Markdown, and page-copy document formats where supported.
• Special downloads include page copy TXT/Markdown/HTML/DOCX/ODT, image ZIP files, SERP preview PNG/WebP images, accessibility tree JSON, schema JSON, robots.txt files, and robots test results.
• Browser download behavior may vary, but exports stay local and are generated from the active tab.

Limits, Truncation, And Skipped Data

The toolkit favors complete machine-readable exports and readable human reports. The limits below are intentional guardrails so very large pages, huge images, and inaccessible cross-origin resources do not freeze the browser.

• CSV, TSV, and JSON keep full cell values for tabular exports. CSV and TSV quote cells with delimiters, quotes, or line breaks.
• Markdown table exports shorten very long cells to 180 characters so tables stay readable. The shortened value ends with three dots.
• TXT exports preserve full values, normalize whitespace, and print records as labeled field blocks separated by blank lines.
• JavaScript links are shown as "JavaScript link" and mailto share links are shown as "Email share link" in readable exports to avoid dumping noisy script or share URLs.
• Rendered text fields used by headings, links, schema snippets, accessibility names, descriptions, and text previews are capped at 500 characters per field.
• Issue previews show five example rows by default, then provide an in-product expand action when more matching rows exist.
• The rendered accessibility tree shows the first 100 semantic nodes before "Show all"; exports and JSON include the captured accessibility node list.
• The style inventory samples computed styles from the document element, body, and first 500 rendered DOM elements under body.
• The style inventory reports up to 30 font families and 48 colors, sorted by frequency and then value.
• Schema field notes show the first 12 notes until expanded.
• Robots.txt fetches are capped at 512 KiB and cached for five minutes per origin inside the active extension session.
• CrUX responses are cached locally for 24 hours. PageSpeed Insights responses are cached locally for one hour. Use Refresh in Web Vitals to bypass the cached value.
• Source HTML fetches are capped at 2 MiB. SERP favicon fetches are capped at 1 MiB.
• Image ZIP downloads include up to 200 unique image URLs. Each image member is capped at 10 MiB and the full ZIP image payload is capped at 100 MiB.
• Image ZIP manifests include skipped image rows when the unique URL limit, per-image limit, total byte limit, fetch errors, or browser download restrictions prevent an image from being included.
• Cross-origin stylesheets are skipped when the browser blocks CSS rule access; linked stylesheet URLs are still listed when available.
• The accessibility tree skips script, style, template, noscript, aria-hidden, hidden, and display:none elements because they are not useful rendered semantic nodes.

Browser Differences

The core audit and export workflow is shared across supported browsers. The practical differences are browser API limits and permission behavior.

• Chrome and Edge include the current-site JavaScript toggle.
• Firefox and Safari hide the JavaScript toggle because those browsers do not expose the same extension API.
• Safari supports HTTP status badges for normal page loads and search-result page loads once the extension has site permission.
• Safari may require enabling the extension and granting site permission before the toolkit can read the active page.
• Context-menu and download behavior can differ slightly by browser, but the same local-first privacy model applies.

Privacy In Plain English

No accounts, analytics, tracking pixels, background crawling, or server-side page uploads. The toolkit reads the active tab when you use it, generates exports locally, and only opens third-party lookups when you click those buttons.

Screens

Overview Section

Overview is the triage screen. It summarizes whether the page appears indexable and lists the highest-priority local review items.

• Indexability: current URL, canonical URL, robots meta, X-Robots-Tag, and robots.txt result.
• Technical fields: HTTP status, redirect counts, viewport, language, HSTS, charset, word count, and character count.
• Metadata: title, H1, meta description, keywords, and text-length signals for preview use.
• Source vs rendered: compares source title, H1, description, canonical, and robots meta against rendered values.
• Discovery: robots.txt URL, sitemap URLs, hreflang tags, noscript count, resource hints, URL parameters, and likely faceted parameters.
• Review items: local warnings and informational checks grouped from the audit engine.

Headings

The Headings section shows the rendered H1-H6 outline in page order. It counts each heading level, supports copying individual heading text, and exports heading rows.

Heading text is normalized from the rendered DOM and capped at 500 characters per heading. JSON exports preserve the heading objects, while CSV, TSV, Markdown, and TXT use the simplified level/text rows shown in the UI.

Review items may flag missing H1, multiple H1s, title/H1 duplication, empty headings, duplicate heading text, UI-label headings, and skipped heading levels.

Links

The Links section inventories anchors from the rendered page. It records visible anchor text, resolved URL, raw href, whether href exists, rel tokens, target, and link kind.

• Link kinds include internal, external, anchor, mailto, tel, and other.
• Sort by page order, kind, text, URL, or rel tokens in the full workspace.
• Copy and export link data as CSV, TSV, JSON, or Markdown.
• Use the context menu or Tools section to highlight nofollow, sponsored, and UGC links on the page where the browser supports that context.
• Rendered anchor text is normalized and capped at 500 characters. The resolved URL stays available separately from the raw href.
• Filtered link exports export only the matching subset, such as internal, external, follow, nofollow, empty-anchor, duplicate-anchor, sponsored, UGC, or unsafe target=_blank links.

Review items may flag JavaScript hrefs, empty hrefs, hash-only links, links without visible anchor text, unsafe target=_blank links, duplicate anchor text pointing to different URLs, and one URL with multiple anchor texts.

Images

The Images section shows image previews, alt text, source URLs, dimensions, loading mode, and width/height attribute coverage.

• Filter images with and without alt text.
• Sort by page order, filename, alt text, source URL, size, or loading mode in the workspace.
• Copy image URLs, export image rows, or download page image files where browser permissions allow it.
• Review image CSV/TSV/JSON/Markdown exports or image ZIP downloads locally.
• Image exports include the rendered source URL, alt text, dimensions, loading mode, width and height attributes, and duplicate-source signals where available.
• Image ZIP downloads are limited to 200 unique image URLs, 10 MiB per image, and 100 MiB total image bytes. Skipped images are listed in the ZIP manifest with a reason.

Review items may flag missing alt text, priority images without alt text, missing width/height attributes, tiny images that may be decorative, and very large image dimensions.

Schema

The Schema section detects JSON-LD, Microdata, and RDFa. JSON-LD is parsed locally and shown with type, display name, parse status, source, notes, and Google rich-result style signals where available.

• Sort by page order, format, schema type, JSON name, or validity.
• Export schema as CSV, TSV, JSON, or Markdown.
• Open the current page in Schema Validator or Rich Results Test.
• Review recommended-field notes for common Article, Product, FAQPage, BreadcrumbList, Organization, WebSite, and LocalBusiness types.
• JSON-LD exports preserve parsed structured data. CSV, TSV, Markdown, and TXT summarize format, type, name, status, source, warnings, errors, recommended-field notes, and duplicate IDs.
• Large schema detail panels show the first 12 field notes until expanded.

Review items may flag parse errors, Google rich-result warnings or errors, missing recommended fields, duplicate schema @id values, schema URL and canonical mismatches, and breadcrumb structured-data notes.

Robots

The Robots section fetches robots.txt for the current origin, reads sitemap declarations, evaluates the current URL, and compares crawler access for search and AI-related crawlers.

• Core search crawlers include Googlebot and Bingbot.
• AI search crawlers include OAI SearchBot, Claude SearchBot, PerplexityBot, and detected site-specific bots.
• AI training crawlers include GPTBot, ClaudeBot, Google Extended, Applebot Extended, CCBot, and detected agents.
• User-retrieval crawlers include ChatGPT User, Claude User, Perplexity User, and detected agents.
• Use the URL tester to test a path or URL against one crawler or multiple crawler categories.
• View or download the live robots.txt content when available.
• The side navigation count is the number of user-agents detected in robots.txt, not the total known crawler catalog.
• Robots exports can include selected crawler rows only. Overlapping groups are de-duped by user-agent so the same crawler appears once in the exported rows.
• Robots test-result exports include tested crawler, user-agent, status, tested URL, matched rule, matched line, decision source, also-matched rules, and ignored-group rules.
• Robots.txt fetches stop at 512 KiB and are cached for five minutes per origin during the active session.

Review items may flag robots meta noindex, X-Robots-Tag noindex, robots meta/header conflicts, conflicting robots directives, and robots.txt rules that disallow the current URL.

HTTP Trace

The HTTP Trace section shows the active tab's captured navigation chain: status codes, response headers, server redirects, final response details, JavaScript redirects, and meta-refresh redirects.

• Copy the full trace, just the URLs, or the final URL.
• Download the full trace or individual hops as CSV, TSV, JSON, or Markdown where supported.
• Reload a page when no trace has been captured yet.
• Server redirect rows and the final response can be exported individually. When multiple HTTP entries exist, the HTTP status section also offers combined status exports.
• JavaScript and meta-refresh redirects are detected from the page context and reported separately from server redirects.

Web Vitals

The Web Vitals section combines local browser metrics, Google CrUX field data, CrUX history, and PageSpeed Insights lab diagnostics for the current URL.

• Local Core Web Vitals are captured from the active page load in the current browser viewport.
• CrUX current field data reports real Chrome user experience data for mobile and desktop when Google has enough data.
• CrUX history shows weekly 28-day rolling windows, with mobile and desktop charts paired by metric.
• CrUX rows are labeled URL-level when exact page data is available and sitewide when Google falls back to origin-level data.
• PageSpeed Insights is synthetic Lighthouse lab data from Google servers. It can be run manually or automatically from the Web Vitals setting.
• Open PageSpeed, Lighthouse, DebugBear, and GTmetrix buttons help users compare external performance views when needed.
• Copy or download Web Vitals data as TSV, CSV, JSON, Markdown, or TXT.

Social

The Social section inventories Open Graph and X/Twitter metadata. The full workspace includes an editable preview tester for Facebook/Open Graph, LinkedIn/Open Graph, and X/Twitter cards.

• Edit title, description, and image URL fields to preview alternate social snippets.
• Export social metadata as CSV, TSV, JSON, or Markdown.
• Review title and description length signals while editing preview copy.

Review items may flag OG title differences, OG description differences, and OG URL/canonical mismatches.

Style And Visual Review

The Style section extracts computed fonts, color usage, CSS variables, stylesheet URLs, style tag counts, inline style counts, and accessible CSS rule counts.

• Color swatches show the detected value, usage count, and usage category.
• The local color tool converts selected colors to HEX, RGB, and HSL.
• Where supported, the eyedropper can pick colors from the current page.
• Auto-copy can copy a preferred color format when selecting a swatch.
• The accessible color palette setting improves visual severity states for colorblind users.
• Computed style sampling uses the document element, body, and first 500 rendered DOM elements under body.
• The displayed inventory is capped at 30 font families and 48 colors. CSS variables, stylesheet URLs, style tag counts, inline style counts, and accessible CSS rule counts are reported separately.
• Cross-origin stylesheets are skipped when browser security prevents reading cssRules; the stylesheet URL still appears in the stylesheet list when the link is visible in the DOM.

SERP Preview

The SERP section builds editable desktop and mobile Google-style snippets from the current title, URL path, favicon, and meta description.

• Edit URL/breadcrumb, title, and description text.
• Toggle rich elements such as rating, review count, price, availability, delivery, returns, store nearby, read more, forum/Q&A, related rows, sitelinks, and pill links.
• Download desktop or mobile previews as PNG or WebP.
• Text is truncated by measured pixel width for desktop and mobile preview limits: title width is 580 px on desktop and 805 px on mobile; description width is 920 px on desktop and 680 px on mobile.
• The exported image uses the same truncated title and description as the preview canvas.

Webmaster Tools

The Webmaster Tools section provides direct shortcuts and local helpers for Google Search Console and Bing Webmaster Tools.

• Open Search Console Domain Property or URL Prefix Property views.
• Open Search Console Performance reports filtered to the current page for either property type.
• Open Search Console indexing reports for either property type.
• Open Bing Domain Site or URL Prefix Site views.
• Inspect the current page in Bing Webmaster Tools.
• Add percent-change chips to supported Search Console and Bing comparison rows and scorecards.

Tools

The Tools section collects fast local actions and external lookups for the current page.

• Highlight nofollow, sponsored, and UGC links on the current page.
• Toggle JavaScript for the current site in Chrome and Edge, then reload. Firefox and Safari do not expose this Chromium-only extension API.
• Run Google site search, HTTP-version search, page search, page image search, or site image search.
• Export visible Google search results as CSV and copy the current Google query on supported result pages.
• Search the current Search Console property when using Search Console.
• Open Ahrefs, Archive.org, Moz Link Explorer, Semrush, Similarweb, Schema Validator, or Rich Results Test.
• Review the rendered accessibility tree with landmark, heading, control, image, issue, role, name, relationship, path, and fix summaries.
• The rendered accessibility tree is derived from the current DOM. It is not Chrome DevTools' private internal accessibility tree.
• Accessibility details include roles, names, descriptions, text, URL fields, src/srcset candidates, ARIA attributes, states, relationships, issue labels, fix guidance, and locator paths. Inline elements without their own asset URL show the page URL plus DOM path.
• The tree skips script, style, template, noscript, aria-hidden, hidden, and display:none elements. The panel initially shows the first 100 semantic nodes and can expand to the full captured list.

Third-party lookup buttons open new tabs only when clicked. Shumway SEO Toolkit does not send background requests to those services.

Exports

Exports are generated locally in the browser from the current page audit. The full workspace lets you choose datasets and formats.

• Datasets: issues, metadata, technical, discovery, HTTP trace, Web Vitals, headings, links, images, schema, robots, social, style, SERP, tools, accessibility, and support data where available.
• Formats: CSV, TSV, JSON, Markdown, and page-copy document formats where supported.
• Additional outputs: page copy TXT/Markdown/HTML/DOCX/ODT, image ZIP downloads, SERP PNG/WebP previews, accessibility tree JSON, schema JSON, robots.txt files, and robots test result downloads.
• Full export bundles the selected audit data into a single selected format where supported.
• CSV, TSV, and JSON are the best formats when you need complete machine-readable values. Markdown is optimized for readable tables and truncates cells longer than 180 characters.
• TXT exports are optimized for plain reading: they print only non-empty fields, keep full values, normalize whitespace, and separate records with blank lines.
• Page Copy exports are different from audit exports. They extract readable rendered content from article, main, or body content; preserve headings, paragraphs, blockquotes, preformatted text, list items, table rows, links, and image references; and output TXT, Markdown, HTML, DOCX, or ODT.
• Page Copy skips navigation-like wrapper noise when a cleaner article or main content candidate has more text. It does not crawl linked pages or download embedded images.

Context Menu Actions

Right-click page, link, or selected content where supported to access Shumway SEO Toolkit actions.

• Open SEO Workspace.
• Highlight Nofollow Links.
• Toggle JavaScript For This Site in supported browsers.
• Search Indexed Pages For This Domain.
• Image Search This URL or Image Search This Domain.
• Download Page Copy, Image Files, Overview TSV, Headings TSV, Links TSV, Schema JSON, robots.txt files, and sitemap URL files where available.
• Open the current page or link in Schema Validator, Rich Results Test, Ahrefs, Archive.org, Moz, Semrush, or Similarweb.
• Export visible Google results and copy Google query on supported Google search pages.
• Show percent change in supported Search Console and Bing Webmaster Tools comparison views from the main right-click menu.
• Search the current Search Console property on supported Search Console pages.

Settings And Accessibility

Settings are stored in browser extension storage and stay local to the browser unless the user changes them, resets the extension, or uninstalls it.

• Accessible color palette: uses a colorblind-friendlier severity system. Blue indicates passed or indexable, yellow indicates review, plum indicates critical, and neutral states remain subdued.
• PageSpeed Auto-run: controls whether PageSpeed Insights runs automatically when the Web Vitals tab opens.
• Theme toggle: the full workspace stores a light/dark preference separately from page content.

Privacy Model

Shumway SEO Toolkit reads the active tab only when used. Core page analysis and exports run locally in the user's browser. The extension does not use analytics, tracking pixels, remote logging, accounts, or background crawling.

Web Vitals can request Google CrUX and PageSpeed Insights data for the current URL. Those responses are cached in browser storage so repeat views avoid unnecessary API calls, and the Web Vitals Refresh action requests fresh remote performance data.

See the privacy policy for the full policy text. For support, privacy, or store-review questions, email support@shumway.io.