fix(seo): correct canonical URLs, compress oversized images, add cache headers

[emir-karabeg]

Summary

• Fix wrong canonical URLs across all public pages — https://sim.ai → https://www.sim.ai
• Compress 6 oversized blog/landing images from 2.6MB to 300KB total
• Add Cache-Control headers for static assets
• Add SEO regression test to prevent future canonical URL violations

Context

A third-party SEO audit flagged three issues: wrong canonical URLs making pages unindexable, oversized images (up to 1.1MB), and poor Core Web Vitals. This PR addresses all three.

Changes

Canonical URLs

• Add SITE_URL constant (https://www.sim.ai) to lib/core/utils/urls.ts
• Replace all hardcoded https://sim.ai references across 30+ files (layouts, pages, blog frontmatter, JSON-LD, RSS feeds, redirects)
• Migrate models, integrations, and homepage metadata from getBaseUrl() to SITE_URL

Image Compression

• mothership/cover: 1,122KB → 99KB (converted PNG → JPEG for photographic content)
• v0-5/cover.png: 282KB → 66KB
• executor/cover.png: 184KB → 81KB
• series-a/cover.png: 123KB → 26KB
• workflow.png: 109KB → 31KB
• multiplayer-cover.png: 784KB → 18KB

Performance

• Add Cache-Control: public, max-age=86400, stale-while-revalidate=604800 for static assets (images, fonts, icons)

Regression Test

• New seo.test.ts scans all public-facing directories for:
  • Hardcoded https://sim.ai without www
  • getBaseUrl() used in page metadata (should use SITE_URL)
  • SITE_URL constant value correctness

Type of Change

• Bug fix
• Improvement

Testing

• All 285 test files, 5,075 tests passing
• Verified canonical URLs via curl on all flagged pages
• Verified all compressed images render correctly in dev

Checklist

• Code follows project style guidelines
• Self-reviewed my changes
• Tests added/updated and passing
• No new warnings introduced
• I confirm that I have read and agree to the terms outlined in the Contributor License Agreement (CLA)

[vercel]

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

1 Skipped Deployment

Project	Deployment	Actions	Updated (UTC)
docs	Skipped		Apr 14, 2026 11:14pm

[cursor]

PR Summary

Medium Risk
Touches SEO metadata/structured data across many public routes and content, which could affect indexing if any URL is miscomputed. Changes are mostly straightforward constant substitutions plus static-asset cache headers and should be easy to validate in production.

Overview
Standardizes all public-facing SEO URLs to use a single canonical base (SITE_URL = https://www.sim.ai) instead of hardcoded https://sim.ai or getBaseUrl() in metadata, updating Next.js metadataBase, canonical alternates, OpenGraph/Twitter URLs, JSON-LD, RSS, sitemap routes, and legacy redirects.

Adds a Vitest regression test (app/(landing)/seo.test.ts) that scans marketing/SEO surfaces to prevent reintroducing bare https://sim.ai and to ensure getBaseUrl() isn’t used for SEO metadata, and configures Cache-Control headers for static assets in next.config.ts.

^{Reviewed by Cursor Bugbot for commit 7d60d5e. Configure here.}

[greptile-apps]

Greptile Summary

This PR fixes canonical URL mismatches (https://sim.ai → https://www.sim.ai) across 30+ public-facing files by introducing a SITE_URL constant, compresses six oversized blog/landing images, adds Cache-Control headers for static assets, and adds an SEO regression test. The changes are thorough and well-scoped — all changed files import and use SITE_URL correctly, and the MDX frontmatter canonical fields are updated to match.

Confidence Score: 5/5

Safe to merge — all remaining findings are P2 style suggestions with no correctness or security impact.

The PR is a thorough, well-executed SEO fix. SITE_URL is correctly introduced and applied across every affected file. The only finding is a P2 precision nit on the Cache-Control source glob pattern (missing escaped dot), which has no practical security impact given that authenticated file-serving routes always set their own Cache-Control headers explicitly.

apps/sim/next.config.ts — minor glob pattern precision for the Cache-Control source rule.

Important Files Changed

Filename	Overview
apps/sim/lib/core/utils/urls.ts	Adds SITE_URL = 'https://www.sim.ai' constant; clean, well-documented addition alongside existing getBaseUrl().
apps/sim/next.config.ts	Adds Cache-Control: public, max-age=86400, stale-while-revalidate=604800 for static asset extensions; the source glob pattern lacks a literal . before the extension group (minor precision concern), and redirects updated to www.sim.ai.
apps/sim/app/(landing)/seo.test.ts	New regression test scanning SEO-relevant directories for bare https://sim.ai references and getBaseUrl() in metadata exports; covers all scan paths from the PR and correctly excludes the test file itself.
apps/sim/lib/blog/seo.ts	All JSON-LD url and breadcrumb item fields migrated from hardcoded https://sim.ai strings to SITE_URL template literals; complete and consistent.
apps/sim/app/(landing)/components/structured-data.tsx	All @id, url, installUrl, and breadcrumb fields switched from hardcoded https://sim.ai to SITE_URL; thorough sweep of the large JSON-LD graph.
apps/sim/app/(landing)/models/[provider]/[model]/page.tsx	Switches baseUrl from getBaseUrl() to SITE_URL; also replaces a hardcoded href='https://sim.ai' with href='/' so the "Build with this model" CTA stays on the current environment in staging/dev.
apps/sim/ee/whitelabeling/metadata.ts	Keeps getBaseUrl for dynamic whitelabel tenant metadata while replacing the hardcoded https://sim.ai creator URL with SITE_URL; correct separation of concerns.
apps/sim/content/blog/mothership/index.mdx	Canonical URL updated to https://www.sim.ai/blog/mothership; ogImage switched from .png to .jpg to match the new compressed JPEG cover.

Flowchart

%%{init: {'theme': 'neutral'}}%%
flowchart TD
    A[lib/core/utils/urls.ts\nSITE_URL = 'https://www.sim.ai'] --> B[Metadata / metadataBase]
    A --> C[JSON-LD structured data\n@id, url, breadcrumbs]
    A --> D[Open Graph / alternates\ncanonical, og:url]
    A --> E[RSS & Sitemap feeds\nchannel link, loc]
    A --> F[next.config.ts\nredirects to www.sim.ai]

    B --> G[Next.js pages\nlayout.tsx, page.tsx]
    C --> H[structured-data.tsx\nblog/seo.ts]
    D --> I[blog / integrations\nmodels / partners pages]
    E --> J[rss.xml/route.ts\nsitemap-images.xml]

    K[Cache-Control headers\nnext.config.ts] --> L[Static assets\n.svg .jpg .png .woff etc.\nmax-age=86400 / swr=604800]

    M[seo.test.ts\nregression test] --> N{Scans SEO dirs}
    N --> O[Assert: no bare https://sim.ai]
    N --> P[Assert: no getBaseUrl in metadata]
    N --> Q[Assert: SITE_URL = https://www.sim.ai]

Loading

_{Reviews (2): Last reviewed commit: "fix(seo): replace hardcoded URLs with SI..." | Re-trigger Greptile}

[cursor]

SEO test misses bare URLs in MDX links

Low Severity

The regression test scans .mdx files for hardcoded https://sim.ai but only checks for URLs enclosed in JS quotes, backticks, or YAML canonical: prefix. It misses Markdown link syntax like (https://sim.ai) and bare URLs in template-literal XML. Three existing blog posts contain (https://sim.ai) in body content, demonstrating the blind spot is already producing false negatives on the first run.

 

^{Reviewed by Cursor Bugbot for commit 67a91e4. Configure here.}

[emir-karabeg]

@cursor review

[emir-karabeg]

@greptile

[cursor]

✅ Bugbot reviewed your changes and found no new issues!

1 issue from previous review remains unresolved.

Comment @cursor review or bugbot run to trigger another review on this PR

^{Reviewed by Cursor Bugbot for commit 7d60d5e. Configure here.}