seo: structured data, indexable provider/FAQ content, metadata overhaul

[leoisadev1]

Summary

We rank behind Cloudflare's "Sandbox SDK" docs for the query sandbox sdk. The site gave crawlers very little to work with: the homepage was an animated landing page whose only text was "One sandbox API. Every provider." (the phrase "Sandbox SDK" appeared nowhere in the body), there was no structured data at all, and /providers was in the sitemap while returning 404.

This adds the on-page substance and machine-readable signals, without changing any product claims.

New src/lib/seo.ts centralizes the SEO surface — canonical URLs, tagline, short meta description vs. long schema description, keywords, the FAQ corpus, and schema builders:

export const siteDescription    // ~155 chars, for <meta name="description">
export const siteDescriptionLong // full detail, for JSON-LD
export const faqs: readonly Faq[]  // 7 Q&As, rendered AND emitted as FAQPage
export const organizationSchema, websiteSchema, softwareSchema, faqSchema
export function breadcrumbSchema(trail), techArticleSchema(article)

Structured data via a small <JsonLd data={...} /> component (escapes < before injecting). Root layout emits Organization + WebSite + SoftwareSourceCode; the homepage emits FAQPage; every docs page emits TechArticle + a BreadcrumbList built from its slug, resolving each segment's real title through source.getPage() and falling back to title-cased slug.

Homepage gains two indexable sections plus a visible Sandbox SDK kicker above the H1 and a keyword-bearing intro paragraph:

• a provider grid (7 cards) deep-linking into /docs/providers/*
• an FAQ accordion rendering the same faqs that back the FAQPage schema
• footer links to Providers / Integrations / Compatibility / Security / GitHub / npm

Previously the homepage linked out to almost nothing; these are the internal links that let PageRank reach the docs tree.

Metadata: metadataBase, a title template (%s · Sandbox SDK), authors/publisher/category, explicit robots + googleBot (max-snippet: -1, max-image-preview: large), and fuller OG/Twitter. Docs pages now use their own generated OG image (getPageImage(page)) instead of the one shared og-image.png.

Fixes: dropped the dead /providers entry from the sitemap (curl → 404; the real route is /docs/providers), and sitemap.ts now imports siteUrl instead of hardcoding the origin.

Off-site: README (root and published package) now open with a sentence that actually names and defines the product, and packages/sdk/package.json gains a descriptive description + keywords for npm search.

Verified lint, typecheck, and build pass, and checked the static export: canonical URLs are per-page (not all /), the emitted @types are Organization/WebSite/SoftwareSourceCode/FAQPage, docs pages carry BreadcrumbList and per-page og:image, and /providers is gone from sitemap.xml.

Link to Devin session: https://app.devin.ai/sessions/95f7fbaac03841f3ae947ea53480fd32
Requested by: @leoisadev1

[devin-ai-integration]

🤖 Devin AI Engineer

I'll be helping with this pull request! Here's what you should know:

✅ I will automatically:

• Address comments on this PR. Add '(aside)' to your comment to have me ignore it.
• Look at CI failures and help fix them

Note: I can only respond to comments from users who have write access to this repository.

⚙️ Control Options:

• Disable automatic comment, CI, and merge conflict monitoring

[vercel]

The latest updates on your projects. Learn more about Vercel for GitHub.

Project	Deployment	Actions	Updated (UTC)
sandbox-sdk	Ready	Preview	Aug 24, 2026 3:19pm

[greptile-apps]

Greptile Summary

This change improves the Sandbox SDK documentation site’s search presentation and discoverability by centralizing metadata, adding structured data, expanding provider and FAQ content, generating page-specific social metadata, and removing an obsolete sitemap entry.

The production static export built successfully. Browser checks confirmed that all seven homepage provider links resolve, homepage and documentation structured data is parseable, documentation canonical and social-image URLs resolve, and the sitemap includes current provider documentation routes while excluding the retired /providers URL.

No product-facing defects were found. The change is safe to merge.

Confidence Score: 5/5

Safe to merge: the changed documentation site builds and its newly added navigation, metadata, structured data, social image, and sitemap behavior were exercised successfully.

The final review contains no independent findings. Production-export browser checks covered the primary user-visible paths introduced by the change.

Files Needing Attention: No files require follow-up based on the completed checks.

T-Rex Logs

What T-Rex did

• The documentation site was built with bun run --cwd apps/fumadocs build and the Next.js static export completed successfully.
• The generated export was served and browser validation was performed against the internal URL, verifying seven provider cards and destinations, JSON-LD types for Organization, WebSite, SoftwareSourceCode, FAQPage, TechArticle, and BreadcrumbList, canonical and breadcrumb URLs, the og image path, and sitemap inclusion/exclusion behavior.
• The runtime validation script was executed against the served export with the specified environment, and it exited successfully with all validations passing for the seven provider cards and destinations, JSON-LD types, canonical and breadcrumb URLs, the og image, and sitemap behavior.

_{Ran code and verified through T-Rex}

_{Reviews (1): Last reviewed commit: "seo: structured data, richer metadata, p..." | Re-trigger Greptile}