SEO/GEO audit fixes: titles, author E-E-A-T, schema, lastmod

README.md

@@ -181,7 +181,7 @@ The floating navbar is a mega-menu with multi-column dropdowns. The Technology a
 | **Icons** | React Icons | `post-body.tsx`, `more-stories.tsx`, `NotFoundPage.tsx`, `tagIcons.ts` |
 | **Date Formatting** | date-fns | `date.tsx` |
 | **SEO** | JSON-LD Structured Data | `lib/structured-data.ts` (Organization, WebSite, BlogPosting, TechArticle, BreadcrumbList schemas) |
-| **SEO** | Sitemap + robots.txt | `public/sitemap.xml`, `public/robots.txt` |
+| **SEO** | Sitemap (robots.txt lives at keploy.io root, keploy/landing) | `public/sitemap.xml` |
 | **Analytics** | Google Analytics, Microsoft Clarity, Apollo | `layout.tsx` (tracking scripts) |
 | **Deployment** | Vercel + CloudFront | Production at `keploy.io/blog` |
 | **CI/CD** | GitHub Actions | `build.yml`, `lighthouse_runner.yml`, `playwright.yml`, `indexnow.yml`, `submit-google-sitemap.yml`, `sync-sitemap.yml` |
@@ -254,7 +254,6 @@ blog-website/
 │   ├── images/             # Static images and illustrations
 │   ├── avatars/            # Testimonial avatars
 │   ├── favicon/            # Favicon set
-│   ├── robots.txt          # Crawl rules
 │   └── sitemap.xml         # Generated sitemap
 ├── styles/
 │   └── index.css           # Global styles + Tailwind directives

components/PostHeaderAuthors.tsx

@@ -186,12 +186,14 @@ const PostHeaderAuthors = ({
   blogreviewer,
   timetoRead,
   date,
+  modified = "",
   tags,
 }: {
   blogwriter: { name: string; ImageUrl: string; description: string }[];
   blogreviewer: { name: string; ImageUrl: string; description: string }[];
   timetoRead: number;
   date: string;
+  modified?: string;
   tags?: { edges: { node: { name: string } }[] };
 }) => {
   const sameAuthor =
@@ -209,6 +211,10 @@ const PostHeaderAuthors = ({
   writerDismissRef.current = writer.dismiss;
 
   const tagList = tags?.edges?.slice(0, 4) || [];
+  // Visible "Updated" date only when WP's modified date is a later calendar day
+  // than the publish date — matches the JSON-LD dateModified.
+  const showUpdated =
+    !!modified && !!date && modified.slice(0, 10) > date.slice(0, 10);
 
   return (
     <>
@@ -322,8 +328,13 @@ const PostHeaderAuthors = ({
           <div className="flex items-center gap-4 mt-0.5 flex-wrap">
             <span className="flex items-center gap-1.5 text-xs sm:text-sm text-gray-500">
               <CalendarIcon />
-              <span>{formatPostDate(date)}</span>
+              <time dateTime={date}>{formatPostDate(date)}</time>
             </span>
+            {showUpdated && (
+              <span className="flex items-center gap-1.5 text-xs sm:text-sm text-gray-500">
+                Updated <time dateTime={modified}>{formatPostDate(modified)}</time>
+              </span>
+            )}
             <span className="flex items-center gap-1.5 text-xs sm:text-sm text-gray-500">
               <ClockIcon />
               <span>{timetoRead} minutes</span>

components/post-header.tsx

@@ -9,6 +9,7 @@ export default function PostHeader({
   title,
   coverImage,
   date,
+  modified = "",
   author,
   categories,
   BlogWriter,
@@ -31,13 +32,17 @@ export default function PostHeader({
       {/* 3. Title */}
       <PostTitle>{title}</PostTitle>
 
-      {/* 4. Author card + social share icons */}
-      <div className="mb-8">
+      {/* 4. Author card + social share icons. An <aside> (not a <div>) so
+             readers/LLM extractors treat byline, date, tags and the
+             Summarize-with-AI CTAs as page chrome, and the article's first
+             paragraph is the first real content passage. Same box, same layout. */}
+      <aside aria-label="Post details" className="mb-8">
         <PostHeaderAuthors
           blogwriter={BlogWriter}
           blogreviewer={BlogReviewer}
           timetoRead={TimeToRead}
           date={date}
+          modified={modified}
           tags={tags}
         />
 
@@ -47,7 +52,7 @@ export default function PostHeader({
           <SummarizeWithAI />
           <PreferredSourceButton />
         </div>
-      </div>
+      </aside>
     </div>
   );
 }

components/post-title.tsx

@@ -1,10 +1,7 @@
+import { toHeadlineCase } from "../utils/titleCase";
+
 export default function PostTitle({ children }: { children: string }) {
-  function toTitleCase(str: string) {
-    return str
-      .toLowerCase()
-      .replace(/\b\w/g, (word) => word.toUpperCase());
-  }
-  const finalString = toTitleCase(children);
+  const finalString = toHeadlineCase(children);
   return (
     <h1
       style={{

lib/api.ts

@@ -239,7 +239,6 @@ export async function getReviewAuthorDetails(authorName) {
         edges {
           node {
             name
-            email
             avatar {
               url
             }

lib/author-box.ts

@@ -44,3 +44,55 @@ export function extractAuthorBox(html?: string): AuthorBoxMeta {
     bio: extractAuthorBio(html),
   };
 }
+
+// PublishPress injects its own author box into post content. On keploy.io it
+// renders empty (only an `#author-description-placeholder` div) and carries the
+// plugin's hardcoded `box-post-id-16` on every post, so crawlers saw an empty,
+// identical author block on all pages. The page renders its own per-post
+// AuthorCard from ppmaAuthorName/ppmaAuthorImage + the WP user bio, so the
+// plugin markup (and its scoped <style> block) is removed from the body.
+// Call extractAuthorBio on the RAW content first if you need its bio.
+// Match each complete begin/end block on its own (the body may not contain
+// another "begin code"), then drop only blocks that are PublishPress boxes, so
+// an earlier unrelated code block can never be swallowed.
+const CODE_BLOCK_RE = /<!--\s*begin code\s*-->(?:(?!<!--\s*begin code\s*-->)[\s\S])*?<!--\s*end code\s*-->/gi;
+const PP_BOX_STYLE_RE = /<style[^>]*>[^<]*pp-multiple-authors-boxes-wrapper[\s\S]*?<\/style>/gi;
+
+export function stripPublishPressAuthorBox(html?: string | null): string {
+  if (!html) return "";
+  return html
+    .replace(CODE_BLOCK_RE, (block) => (block.includes("pp-multiple-authors") ? "" : block))
+    .replace(PP_BOX_STYLE_RE, "");
+}
+
+type WpUserNode = { name?: string; avatar?: { url?: string }; description?: string };
+
+/**
+ * Pick the WP user whose display name matches the post's PublishPress author
+ * exactly (case-insensitive). `users(where:{search})` is a fuzzy match, so a
+ * loose first-hit would attach another person's bio to the post.
+ */
+export function pickWpUserForAuthor(
+  users: { edges?: { node?: WpUserNode }[] } | null | undefined,
+  authorName?: string | null,
+): WpUserNode | null {
+  const target = (authorName || "").trim().toLowerCase();
+  if (!target) return null;
+  return (
+    users?.edges
+      ?.map((e) => e?.node)
+      .find((n) => (n?.name || "").trim().toLowerCase() === target) ?? null
+  );
+}
+
+export type PublicAuthor = { name: string; avatar: { url: string } | null; description: string };
+
+/** Whitelist public author fields only — never email or other private data. */
+export function toPublicAuthor(node: WpUserNode | null | undefined): PublicAuthor | null {
+  if (!node?.name) return null;
+  return {
+    name: node.name,
+    avatar: node.avatar?.url ? { url: node.avatar.url } : null,
+    description: node.description || "",
+  };
+}

lib/structured-data.ts

@@ -60,6 +60,8 @@ export const SOCIAL_LINKS = [
   "https://www.g2.com/products/keploy/reviews",
   "https://www.capterra.in/software/1070466/Keploy",
   "https://aws.amazon.com/marketplace/reviews/reviews-list/prodview-xgwmdk4ivjjv4",
+  // Wikidata item for Keploy — the entity anchor LLMs use to resolve the brand.
+  "https://www.wikidata.org/wiki/Q137395272",
 ];
 
 type BreadcrumbItem = {
@@ -480,12 +482,16 @@ export const getBlogPostingSchema = ({
       "@type": "Organization",
       "@id": ORG_ID,
       name: ORG_NAME,
+      url: MAIN_SITE_URL,
       logo: {
         "@type": "ImageObject",
         url: ORG_LOGO_URL,
         width: 512,
         height: 512,
       },
+      // Repeated here (not only on the global Organization node) so a crawler
+      // that reads just this article still links the publisher to Wikidata.
+      sameAs: SOCIAL_LINKS,
     },
     isPartOf: blogReference(),
     // The blog article itself is free to read (no paywall/registration) — a
@@ -580,7 +586,12 @@ export const getBlogPostingSchema = ({
   // primary topic. This is the connective tissue that lets an AI engine attach
   // "Keploy does X" claims from any article to the one known tool entity.
   schema.mentions = [
-    { "@type": "SoftwareApplication", "@id": SOFTWARE_ID, name: ORG_NAME },
+    {
+      "@type": "SoftwareApplication",
+      "@id": SOFTWARE_ID,
+      name: ORG_NAME,
+      applicationCategory: "DeveloperApplication",
+    },
   ];
 
   return schema;

next.config.js

@@ -87,15 +87,8 @@ module.exports = {
           },
         ],
       },
-      {
-        source: '/robots.txt',
-        headers: [
-          {
-            key: 'Cache-Control',
-            value: 'public, s-maxage=86400, stale-while-revalidate=86400',
-          },
-        ],
-      },
+      // No /blog/robots.txt: crawlers only read robots.txt at the host root
+      // (RFC 9309), so all rules live in keploy.io/robots.txt (keploy/landing).
       {
         source: '/(.*)',
         headers: [

pages/community/[slug].tsx

@@ -16,6 +16,7 @@ import {
   getMoreStoriesForSlugs,
   getPostAndMorePosts,
   getReviewAuthors,
+  getReviewAuthorDetailsCached,
 } from "../../lib/api";
 import ContainerSlug from "../../components/containerSlug";
 import { useEffect, useMemo, useRef } from "react";
@@ -34,7 +35,12 @@ import {
   SITE_URL,
 } from "../../lib/structured-data";
 import { sanitizeTitle, getSafeDescription, buildPageTitle } from "../../utils/seo";
-import { extractAuthorBio } from "../../lib/author-box";
+import {
+  extractAuthorBio,
+  pickWpUserForAuthor,
+  toPublicAuthor,
+  stripPublishPressAuthorBox,
+} from "../../lib/author-box";
 import { getHowToSchema } from "../../lib/howToSchema";
 import { detectCodeLanguages, countWords, extractFaqs } from "../../utils/contentSchema";
 import { getTooltipsForSlug } from "../../config/keyword-tooltips";
@@ -63,7 +69,7 @@ const transformPostContent = (content: string, ppmaAuthorName: string) => {
     );
 };
 
-export default function Post({ post, posts, reviewAuthorDetails, preview }) {
+export default function Post({ post, posts, reviewAuthorDetails, writerDetails, preview }) {
   const router = useRouter();
   const { slug } = router.query;
   const morePosts = posts?.edges;
@@ -94,7 +100,13 @@ export default function Post({ post, posts, reviewAuthorDetails, preview }) {
   // "n/a", empty) that would otherwise 400 the next/image optimizer and render a
   // broken byline avatar; a real URL passes through unchanged (and it falls back
   // to the S3 AUTHOR_AVATAR_PLACEHOLDER when missing).
-  const writerAvatarUrl = resolveAuthorAvatar(post?.ppmaAuthorImage);
+  // Already reduced server-side to public fields (name, avatar, bio).
+  const wpWriter = writerDetails;
+  const ppmaAvatarUrl = resolveAuthorAvatar(post?.ppmaAuthorImage);
+  const writerAvatarUrl =
+    ppmaAvatarUrl === AUTHOR_AVATAR_PLACEHOLDER
+      ? resolveAuthorAvatar(wpWriter?.avatar?.url)
+      : ppmaAvatarUrl;
   // For JSON-LD only: a genuine author photo (absolute URL) or nothing — never
   // the placeholder or a junk value, either of which would be invalid in schema.
   const rawPpmaImage = (post?.ppmaAuthorImage ?? "").trim();
@@ -104,7 +116,10 @@ export default function Post({ post, posts, reviewAuthorDetails, preview }) {
   // content so the real bio is in the SSR HTML (no state, no effect).
   // realWriterBio is the actual author-box bio (or undefined); it feeds the
   // schema's "real bio only" guard so the generic UI fallback never enters JSON-LD.
-  const realWriterBio = extractAuthorBio(post?.content);
+  // Bio priority: PublishPress author-box bio in the content, then the
+  // author's WordPress user bio (Users > Profile > Biographical Info).
+  const realWriterBio =
+    extractAuthorBio(post?.content) || wpWriter?.description?.trim() || undefined;
   const blogWriterDescription = realWriterBio ?? "An author for Keploy's blog.";
 
   // Back-compat alias: other parts of this component previously used
@@ -279,6 +294,7 @@ export default function Post({ post, posts, reviewAuthorDetails, preview }) {
                 title={post?.title || "Loading..."}
                 coverImage={post?.featuredImage}
                 date={post?.date || ""}
+                modified={post?.modified || ""}
                 author={post?.ppmaAuthorName || ""}
                 categories={post?.categories || []}
                 BlogWriter={blogwriter}
@@ -298,7 +314,7 @@ export default function Post({ post, posts, reviewAuthorDetails, preview }) {
           <div ref={postBodyRef}>
             <PostBody
               content={transformPostContent(
-                post?.content,
+                stripPublishPressAuthorBox(post?.content),
                 post?.ppmaAuthorName,
               )}
               authorName={post?.ppmaAuthorName || ""}
@@ -396,13 +412,24 @@ export const getStaticProps: GetStaticProps = async ({
     // Same two records for every post — memoized in lib/api so a build makes
     // this request twice in total instead of twice per post.
     const authorDetails = await getReviewAuthors();
+    // Per-post writer profile (bio/avatar) for the author card. Memoized by
+    // name; a failure here must never 404 the post, so it degrades to null.
+    // Select the exact-name WP user here and keep only public fields, so
+    // nothing private (e.g. email) is serialized into __NEXT_DATA__.
+    const writerUsers = data.post?.ppmaAuthorName
+      ? await getReviewAuthorDetailsCached(data.post.ppmaAuthorName).catch(() => null)
+      : null;
+    const writerDetails = toPublicAuthor(
+      pickWpUserForAuthor(writerUsers, data.post?.ppmaAuthorName),
+    );
 
     return {
       props: {
         preview,
         post: data.post,
         posts: moreStories?.communityMoreStories || { edges: [] },
         reviewAuthorDetails: authorDetails,
+        writerDetails: writerDetails ?? null,
       },
       revalidate: REVALIDATE_CONTENT,
     };