docs: make Email SDK agent-ready

[leoisadev1]

Automated PR - Greptile review workflow

[vercel]

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

Project	Deployment	Actions	Updated (UTC)
email-sdk-fumadocs	Ready	Preview, Comment	May 29, 2026 9:32pm

[leoisadev1]

@greptile review

[greptile-apps]

Greptile Summary

This PR makes the Email SDK docs site agent-ready by adding llms.txt/llms-full.txt content-type headers, new robots.txt and sitemap.xml routes, Open Graph/Twitter meta tags, and two new documentation pages (provider-readiness, ai-resources). The agents/skill.mdx page is relocated to getting-started/agent-skill.mdx and the top-level agents/ category is dissolved.

• New infrastructure routes (robots.txt, sitemap.xml) are cleanly implemented: robots.txt uses a single wildcard rule with a Sitemap: pointer; sitemap.xml resolves all page URLs to absolute URLs via new URL(url, siteUrl) with a complete five-entity XML escape function.
• __root.tsx meta tags are updated to add og:type, og:title, og:description, and twitter:card: summary using shared constants; no og:url or canonical tag is emitted (intentionally left to per-page routes to avoid a static root value on every page).
• provider-readiness.mdx and ai-resources.mdx are new well-structured pages covering production adapter selection, fallback safety, test commands, and how to feed the docs to coding assistants.

Confidence Score: 5/5

This PR is safe to merge — all changes are documentation content and supporting infrastructure with no logic that touches the email-sending core.

The new route handlers (robots.txt, sitemap.xml) are server-only and straightforward; the sitemap correctly resolves relative paths to absolute URLs and escapes all five required XML entities. The meta-tag additions in the root route use shared constants and avoid introducing incorrect canonical or og:url values. The two new docs pages add content only.

No files require special attention.

Important Files Changed

Filename	Overview
apps/fumadocs/src/lib/shared.ts	Extracts siteDescription and siteUrl as shared constants; siteUrl reads from SITE_URL / VITE_SITE_URL env vars with a hardcoded fallback, used only in server-side route handlers so the process.env access is appropriate.
apps/fumadocs/src/routes/__root.tsx	Adds og:type, og:title, og:description, twitter:card (summary), twitter:title, and twitter:description meta tags using the new shared constants; no og:url or canonical are added, which is intentional.
apps/fumadocs/src/routes/sitemap[.]xml.ts	New sitemap route generating absolute URLs via new URL(url, siteUrl); escapeXml covers all five XML entities (&, <, >, ", '); Content-Type correctly set to application/xml.
apps/fumadocs/src/routes/robots[.]txt.ts	New robots.txt route; simple and correct — one wildcard User-agent with Allow: / and a Sitemap pointer using siteUrl.
apps/fumadocs/content/docs/getting-started/provider-readiness.mdx	New page covering production checklist, recommended adapter pairs, fallback safety, retry/fallback ordering, hook usage, and test coverage table.
apps/fumadocs/content/docs/getting-started/ai-resources.mdx	New page documenting llms.txt, llms-full.txt, and per-page Markdown endpoints with grounding examples for assistants.
apps/fumadocs/vite.config.ts	Adds robots.txt and sitemap.xml to the pre-render path list so they are included in the static build output.

_{Reviews (4): Last reviewed commit: "fix: address Greptile review feedback" | Re-trigger Greptile}

[greptile-apps]

Greptile Summary

This PR makes the Email SDK docs site agent-ready by adding llms.txt, llms-full.txt, robots.txt, and sitemap.xml endpoints, two new docs pages (provider readiness and AI resources), and moving the agent skill page into the getting-started section. It also enriches the root route with OG/Twitter meta tags and a canonical link.

• New routes: robots.txt (explicit AI crawler allowlist), sitemap.xml (full URL inventory), and Content-Type headers on both llms.txt endpoints — all clean additions.
• New docs pages: provider-readiness.mdx and ai-resources.mdx cover production adapter selection, fallback safety rules, and machine-readable resource URLs; content is accurate and consistent with the SDK.
• Meta tag additions in __root.tsx: The canonical <link> and og:url are both hardcoded to the root domain (siteUrl), so every docs page will emit the wrong canonical URL pointing to the homepage — this needs to be made per-page before merging.

Confidence Score: 3/5

The docs content and new endpoints are solid, but the root route emits a static canonical URL on every page, which will tell search engines that all docs pages are duplicates of the homepage.

Almost all of the changes — new routes, docs pages, nav restructuring, Content-Type fixes — are clean and correct. The one issue that needs attention before merging is in __root.tsx: the canonical link and og:url are both set to the root domain for every page on the site, so every individual docs page will incorrectly declare itself a duplicate of the homepage. This will actively harm search indexing of the doc pages this PR is designed to surface.

apps/fumadocs/src/routes/__root.tsx needs per-page canonical and og:url values instead of the static siteUrl root.

Important Files Changed

Filename	Overview
apps/fumadocs/src/routes/__root.tsx	Adds OG/Twitter meta tags and a canonical link — but the canonical href and og:url are both hardcoded to the root domain, so every docs page will emit the wrong canonical URL. Also uses twitter:card summary_large_image without a twitter:image tag.
apps/fumadocs/src/lib/shared.ts	Adds siteDescription and siteUrl constants; siteUrl falls back to the hardcoded domain when env vars are absent. Clean change.
apps/fumadocs/src/routes/sitemap[.]xml.ts	New sitemap route that enumerates all docs pages and their Markdown equivalents. Includes /llms.txt and /llms-full.txt which are non-HTML entries; escapeXml only handles &, <, > which is sufficient for URLs.
apps/fumadocs/src/routes/robots[.]txt.ts	New robots.txt route explicitly allows all AI crawlers; per-bot Allow entries are redundant given the wildcard Allow: / rule, but functionally correct.
apps/fumadocs/content/docs/getting-started/provider-readiness.mdx	New production checklist page covering adapter selection, fallback safety, retry order, hooks, and test coverage. Content is accurate and consistent with the codebase.
apps/fumadocs/content/docs/getting-started/ai-resources.mdx	New page documenting machine-readable endpoints (llms.txt, per-page .md, llms-full.txt) for AI assistant use. Clean and accurate.

Sequence Diagram

sequenceDiagram
    participant Bot as AI Crawler / Search Tool
    participant Robots as /robots.txt
    participant Sitemap as /sitemap.xml
    participant LLMs as /llms.txt
    participant LLMsFull as /llms-full.txt
    participant Docs as /docs/{page}.md

    Bot->>Robots: GET /robots.txt
    Robots-->>Bot: Allow: / + Sitemap pointer

    Bot->>Sitemap: GET /sitemap.xml
    Sitemap-->>Bot: All HTML + Markdown URLs

    Bot->>LLMs: GET /llms.txt
    LLMs-->>Bot: Compact docs index (text/plain)

    Bot->>LLMsFull: GET /llms-full.txt
    LLMsFull-->>Bot: Full docs bundle (text/plain)

    Bot->>Docs: GET /docs/getting-started/quickstart.md
    Docs-->>Bot: Per-page Markdown

Loading

_{Reviews (2): Last reviewed commit: "docs: make Email SDK agent-ready" | Re-trigger Greptile}

[leoisadev1]

@greptile review