- From code and AI to the gadgets on your desk — pick a topic and start exploring.
+ From code and AI to the gadgets on your desk, pick a topic and start exploring.
+
+
+
{post.readingTime}
@@ -148,10 +148,10 @@ function HorizontalCard({ post, className, index }: { post: BlogPostSummary; cla
{/* Number stamp or image */}
-
diff --git a/morya-techcoder/components/ui/MagicBorderCard.tsx b/morya-techcoder/components/ui/MagicBorderCard.tsx
index 7cd93af..2932c88 100644
--- a/morya-techcoder/components/ui/MagicBorderCard.tsx
+++ b/morya-techcoder/components/ui/MagicBorderCard.tsx
@@ -110,16 +110,16 @@ export default function MagicBorderCard({ post, size = "default", className }: P
)}
-
-
+
+
{!isCompact && (
<>
-
-
+
+
{post.readingTime}
-
+
Read
diff --git a/morya-techcoder/content/posts/your-localhost-and-127-0-0-1-aren-t-the-same.mdx b/morya-techcoder/content/posts/your-localhost-and-127-0-0-1-aren-t-the-same.mdx
new file mode 100644
index 0000000..d5ea1b3
--- /dev/null
+++ b/morya-techcoder/content/posts/your-localhost-and-127-0-0-1-aren-t-the-same.mdx
@@ -0,0 +1,462 @@
+---
+title: Your localhost and 127.0.0.1 aren't the same
+excerpt: We generally believe localhost and 127.0.0.1 are the same, but they are not!
+category: Technology
+tags:
+ - Networking
+ - Container
+thumbnail: >-
+ /content/blog/your-localhost-and-127-0-0-1-aren-t-the-same/localhost-post-thumbnail.png
+author: team-techcoder
+date: 2026-09-22
+draft: false
+featured: true
+priority: 2
+difficulty: Intermediate
+prerequisites:
+ - Basic know how of containerization
+review: {}
+seo:
+ title: Why `localhost` and `127.0.0.1` Aren’t the Same Thing
+ ogImage: >-
+ /content/blog/your-localhost-and-127-0-0-1-aren-t-the-same/localhost-post-thumbnail.png
+ noindex: false
+---
+Why `localhost` and `127.0.0.1` Aren’t the Same Thing
+
+You have an application running inside a Docker container.
+
+You open your browser on your laptop and type:
+
+```
+http://localhost:3000
+```
+
+The application opens.
+
+Now stop for a second.
+
+**Where did `localhost` point to?**
+
+Your laptop?
+
+Or the container?
+
+The application is running inside the container, after all. So if `localhost` means “this machine”, shouldn't it refer to the container?
+
+And then there is another address you've probably used hundreds of times:
+
+```
+http://127.0.0.1:3000
+```
+
+That works too.
+
+At this point, it's easy to think:
+
+> `localhost` and `127.0.0.1` are just two ways of writing the same thing.
+
+That assumption works often enough that most developers never question it.
+
+Until you hit a problem where:
+
+```
+localhost:3000 → works
+127.0.0.1:3000 → doesn't
+```
+
+Or the other way around.
+
+So what's actually happening?
+
+The answer starts with something simple:
+
+**`localhost` is a name. `127.0.0.1` is an address.**
+
+But before getting there, we need to answer the container question.
+
+## So, whose `localhost` is it?
+
+If an application is running inside a container and it makes a request to:
+
+```
+localhost:3000
+```
+
+`localhost` refers to **the container itself**.
+
+Not your laptop.
+
+Not the Docker host.
+
+The container has its own network environment, including its own loopback interface.
+
+So if your application inside the container does this:
+
+```
+GET http://localhost:3000
+```
+
+the request stays inside that container.
+
+Think of it like this:
+
+```
+Docker container
+
+┌─────────────────────────────┐
+│ │
+│ Application │
+│ │ │
+│ ▼ │
+│ localhost:3000 │
+│ │ │
+│ ▼ │
+│ same container │
+│ │
+└─────────────────────────────┘
+
+```
+
+This is why `localhost` can be confusing when Docker enters the picture.
+
+When you type `localhost` in your browser, **your browser is the one making the connection**.
+
+So its `localhost` refers to the machine where the browser is running.
+
+When an application inside a container uses `localhost`, **the application is making the connection**.
+
+So its `localhost` refers to the container.
+
+Same word.
+
+Different network environment.
+
+That distinction alone explains a lot of Docker networking confusion.
+
+But it still doesn't explain why `localhost` and `127.0.0.1` aren't technically the same thing.
+
+For that, we need to look at what `127.0.0.1` actually is.
+
+## `127.0.0.1` is an IP address
+
+`127.0.0.1` is an IPv4 address.
+
+More specifically, it's a **loopback address**.
+
+Loopback addresses are used when a machine needs to communicate with itself.
+
+So when you connect to:
+
+```
+127.0.0.1:3000
+```
+
+you're saying:
+
+> Connect to port 3000 on this machine's loopback interface.
+
+The traffic doesn't need to go out to your router or across the network.
+
+It stays on the machine.
+
+That's why developers use it so often for local development:
+
+```
+127.0.0.1:3000
+127.0.0.1:8080
+127.0.0.1:5000
+
+```
+
+But `localhost` works differently.
+
+## `localhost` is a hostname
+
+Unlike `127.0.0.1`, `localhost` isn't an IP address.
+
+It's a **hostname**.
+
+Your operating system needs to resolve that name to an IP address before it can make a connection.
+
+For example:
+
+```
+localhost
+ ↓
+IP address
+ ↓
+connect
+```
+
+On many systems, `localhost` is configured to point to `127.0.0.1`
+
+Which is why the two appear interchangeable most of the time.
+
+But `localhost` isn't permanently tied to `127.0.0.1`.
+
+It can also resolve to the IPv6 loopback address `::1`
+
+So you can have:
+
+```
+localhost
+ │
+ ├── 127.0.0.1 IPv4
+ │
+ └── ::1 IPv6
+```
+
+This is the first reason `localhost` and `127.0.0.1` can behave differently.
+
+## Where `::1` comes into the picture
+
+If you've never seen `::1` before, It's simply the IPv6 equivalent of the IPv4 loopback address.
+
+You can think of the pair like this:
+
+```
+IPv4 → 127.0.0.1
+IPv6 → ::1
+```
+
+Both mean:
+
+> this machine
+
+but they belong to different versions of the Internet Protocol.
+
+So when your browser connects to `localhost:3000`
+
+the connection might ultimately use `127.0.0.1:3000` or `[::1]:3000` depending on how your system resolves the hostname and how the application is listening.
+
+And that's where things can go wrong.
+
+## Your server decides where it listens
+
+Suppose your development server is listening only on:
+
+```
+127.0.0.1:3000
+```
+
+Now your browser tries:
+
+```
+localhost:3000
+```
+
+If `localhost` resolves to `127.0.0.1`, everything is fine.
+
+But if the connection uses IPv6 and ends up at:
+
+```
+[::1]:3000
+```
+
+your server may not be listening there.
+
+You now have:
+
+```
+localhost:3000
+ ↓
+ ::1:3000
+ ↓
+server isn't listening here
+ ↓
+connection fails
+```
+
+while:
+
+```
+127.0.0.1:3000
+ ↓
+127.0.0.1:3000
+ ↓
+server is listening here
+ ↓
+ works
+```
+
+The application didn't disappear.
+
+The port didn't change.
+
+You're still connecting to your own machine.
+
+You simply connected using a different address.
+
+## This is why checking the port isn't always enough
+
+When something running locally refuses to connect, the usual debugging process is:
+
+> Is the server running?
+
+Then:
+
+> Is the port correct?
+
+Those are good questions.
+
+But there's another one that's easy to miss:
+
+> **What address is the server listening on?**
+
+For example, these aren't necessarily equivalent `127.0.0.1:3000` and `[::1]:3000`
+
+The first is IPv4.
+
+The second is IPv6.
+
+Your application can be listening on one and not the other.
+
+That's enough to make `localhost` and `127.0.0.1` behave differently.
+
+## And Docker makes the original question even more important
+
+Go back to our container.
+
+Suppose your application is running inside Docker and listening on `localhost:3000`
+
+Inside the container, that means:
+
+```
+container → localhost:3000
+```
+
+It does **not** automatically mean:
+
+```
+container → host machine:3000
+```
+
+And from your laptop, this:
+
+```
+localhost:3000
+```
+
+means something different again.
+
+It refers to your laptop's network environment.
+
+Docker then provides mechanisms such as port publishing/forwarding to connect the host and container:
+
+```
+Browser
+ │
+ │ localhost:3000
+ ▼
+Host machine
+ │
+ │ published port
+ ▼
+Container
+ │
+ │ application
+ ▼
+server
+
+```
+
+This is why an application can be perfectly healthy inside a container but still be unreachable from your browser if the container's port isn't exposed correctly.
+
+And if you exec into the container and run:
+
+```
+curl localhost:3000
+
+```
+
+you are testing something different from running the same command on your host.
+
+Same command.
+
+Same word.
+
+Different `localhost`.
+
+## The useful distinction
+
+At this point, there are three things worth keeping separate:
+
+```
+localhost
+ ↓
+hostname
+
+
+127.0.0.1
+ ↓
+IPv4 loopback address
+
+
+::1
+ ↓
+IPv6 loopback address
+
+```
+
+And there's another distinction when containers are involved:
+
+```
+Host's localhost
+ ≠
+Container's localhost
+
+```
+
+They can refer to completely different network environments.
+
+So when debugging a local connection, don't just ask:
+
+> “Is it running on localhost?”
+
+Ask:
+
+> **“Whose localhost?”**
+
+Then ask:
+
+> **“What IP address does that name resolve to?”**
+
+And finally:
+
+> **“What address is the server actually listening on?”**
+
+Those questions usually get you much closer to the actual problem than simply restarting the application and trying the URL again.
+
+## So, are `localhost` and `127.0.0.1` the same?
+
+Not technically.
+
+`localhost` is a hostname.
+
+`127.0.0.1` is an IPv4 address.
+
+They commonly lead to the same machine, which is why developers often treat them as interchangeable.
+
+But the distinction matters when:
+
+* your application listens only on IPv4 or IPv6
+* you're working with Docker
+* you're using WSL
+* you're dealing with proxies or VPNs
+* a development server binds to a specific interface
+* the client and server are running in different network environments
+
+The next time you see:
+
+```
+localhost:3000 ❌
+127.0.0.1:3000 ✅
+
+```
+
+don't immediately assume the application is broken.
+
+First find out **where those two connections are actually going**.
+
+That small difference in understanding can save a lot of time when debugging local networking.
diff --git a/morya-techcoder/docs/manual-qa-checklist.md b/morya-techcoder/docs/manual-qa-checklist.md
new file mode 100644
index 0000000..48d4447
--- /dev/null
+++ b/morya-techcoder/docs/manual-qa-checklist.md
@@ -0,0 +1,64 @@
+# TechCoder Manual QA Regression Checklist
+
+Reusable release checklist for the public TechCoder website. Run P0 cases on every release; run P1 for every release candidate or major change; run P2 for broader visual, compatibility, and operational confidence. Record the build/commit, environment, browser/device, tester, date, and pass/fail for each run. File failures with the test ID, route, viewport, reproduction steps, and screenshot or console/network evidence.
+
+## Test Setup
+
+- Test the production build (`npm run build` then `npm run start`) for release validation, not only `next dev`.
+- Use a clean/private browser session for first-visit theme, storage, and cross-browser checks. Also repeat relevant persistence checks in the same browser profile.
+- Use a representative published article with a thumbnail, headings, links, and article actions. Use an article with a review/rating or rich MDX blocks when those content types changed.
+- Newsletter success/duplicate checks need working Resend credentials and a controlled test address. Without configuration, verify the graceful service-error state instead; do not use a real visitor address.
+- Keystatic is available in development. To test its local production gate, follow the repository's `docs/keystatic.md` guidance and use an isolated local build only; never enable its unauthenticated local-storage API on a real deployment.
+- Viewports: narrow phone (320-375 CSS px), typical phone (390-430 px), tablet (768-1024 px), and desktop (1280 px or wider). Test touch and keyboard input where relevant.
+- Respect reduced-motion settings in applicable visual checks. Avoid treating intended lazy/deferred content as missing before scrolling it into view.
+
+## P0: Critical Release Checks
+
+| Test ID | Area | Test Scenario | Steps | Expected Result | Priority |
+|---|---|---|---|---|---|
+| P0-01 | Availability / routes | Public routes render successfully | Open `/`, `/blog`, `/about`, `/contact`, `/privacy`, and `/terms`; open one published `/blog/[slug]` and one populated `/topics/[category]`. | Each route returns usable page content, correct page heading, and shared public-site navigation/footer; no crash, blank page, hydration error, or blocking console error. | P0 |
+| P0-02 | Homepage discovery | Primary calls to action and topic discovery work | From `/`, activate “Start exploring,” “Browse topics,” each available topic tile, featured article cards, and “View all” links. | Links reach the intended article index, topic anchor/page, category listing, or article. The `#topics` target is visible below the fixed header. Topics marked coming soon communicate that state and still follow their configured route behavior. | P0 |
+| P0-03 | Navigation | Desktop and mobile navigation remain usable | At desktop, follow Home, Articles, Programming, AI, Technology, Reviews, Join Newsletter, footer links, and logo. At phone width, open/close the menu, follow a route and newsletter link, then reopen and dismiss with Escape or backdrop. | Every link reaches the intended location; active desktop route is indicated where applicable. Mobile menu exposes the same primary destinations, closes after navigation/dismissal, restores body scrolling, and does not cover or trap the page after closing. | P0 |
+| P0-04 | Article reading | Published article renders and in-page navigation works | Open a published article directly; inspect title, byline, date, reading time, cover, category, body, and related/adjacent links. Use desktop table of contents or mobile contents controls to jump between headings. | Article content and assets render; heading links target the correct sections; active/completed TOC state updates while scrolling; related/back/adjacent links navigate to valid content. | P0 |
+| P0-05 | Newsletter | Newsletter submission gives a truthful outcome | At home or blog page, submit a controlled valid address. Also submit an already-subscribed address, an invalid address, and (without provider configuration or with simulated network failure) a failure case. | New subscription shows success and clears the field; duplicate offers “Use a different email”; invalid/service failures show an understandable alert with retry path; submit control indicates/blocks duplicate in-flight submission. No success is shown on provider failure. | P0 |
+| P0-06 | Responsive layout | Critical content and controls fit small screens | Check `/`, `/blog`, a long article, a populated and empty topic, `/contact`, and footer at 320-375 px; scroll top to bottom and interact with visible controls. | No horizontal page overflow, clipped text, overlapping controls, inaccessible menu/action targets, or obstructed article content; images/cards adapt to a single-column layout and long email address wraps. | P0 |
+| P0-07 | Critical interaction errors | No broken internal links or dead primary controls | Exercise every visible primary CTA, article card, topic card, back link, nav/footer link, search/filter control, and article action on representative pages. | Internal links resolve to the intended page; buttons have a visible effect or state; no dead control, accidental external destination, or unintended 404. | P0 |
+
+## P1: Important Regression Checks
+
+| Test ID | Area | Test Scenario | Steps | Expected Result | Priority |
+|---|---|---|---|---|---|
+| P1-01 | Article index | Search and category filters combine correctly | On `/blog`, search for a known title word, try a partial/case-varied query, choose a category, clear/change the query, select All Posts, and load `/blog?category=`. | Results match the search and selected category; the URL-seeded filter is selected on initial render; All Posts restores the full set. | P1 |
+| P1-02 | Search empty state | No-result search recovers cleanly | Search for a unique nonsense string, then edit or clear the query. | “No posts found” guidance appears for zero results; changing the query returns matching cards without a stale empty/error state. | P1 |
+| P1-03 | Topic and category routes | Populated, empty, and invalid topic paths behave correctly | Open a populated `/topics/[category]`, a coming-soon/empty topic, a populated `/blog/category/[category]`, and an unknown category URL. Test “All articles,” “Browse all articles,” and other-topic links. | Populated routes show correct articles and counts; empty topics show the appropriate empty/coming-soon copy and browse path; category alias shows only that category; unknown paths resolve to a proper not-found response. | P1 |
+| P1-04 | Newsletter validation | Email validation and retry interaction work | Try empty submission, malformed email (for example `name@`), whitespace-surrounded valid email, then activate retry/use-different-email after error/duplicate. | Browser/API validation prevents malformed input from subscribing; usable addresses can be submitted; reset returns to an editable idle form. | P1 |
+| P1-05 | Theme | Theme switching and persistence work | Toggle light/dark from desktop and mobile navigation; reload; open a new tab or revisit; with cleared preference, test system light and dark preference. | Theme applies across the public site, control label/pressed state matches it, saved choice persists, and first visit follows system preference without a distracting wrong-theme flash. | P1 |
+| P1-06 | Article actions | Reading tools work at desktop | On an article at desktop width, scroll and check reading percentage/time, use Copy, Save, Focus, Print, and Share where supported; reload after saving. | Progress increases and reaches finished state near the end; Copy confirms clipboard success; Save toggles pressed state and persists for the article; Focus toggles reading presentation; Print opens a usable print view; Share invokes native share or fallback without a crash. | P1 |
+| P1-07 | Mobile reading chrome | Article toolbar and contents sheet work on phone | Open an article at phone width, scroll past the hero, test back-to-articles, title, Open contents, floating reading button, heading selection, close button, backdrop, and Escape. Tab through the open sheet and close it. | Scrolled toolbar appears with the current article title; back returns to `/blog`; sheet opens with correct headings/progress, closes through supported methods, heading selection jumps and closes, focus enters/is constrained/restored, and body scrolling is locked only while open. | P1 |
+| P1-08 | Loading states | Route transitions have stable loading presentation | Navigate to `/blog` and between articles on a network/device where loading is observable; inspect skeleton before content resolves. | Blog/article skeleton resembles final layout, does not expose broken controls, and content replaces it without a major layout jump. | P1 |
+| P1-09 | Content / MDX | Representative content blocks render and remain operable | Inspect posts containing code blocks, tables, tabs, callouts, images, embeds, quotes, recommendations, or other changed MDX components; operate tabs/copy controls if present. | Blocks format correctly, images have appropriate alt/captions, code is readable and copy works where provided, embeds load or fail gracefully, and no block overflows the article column. | P1 |
+| P1-10 | Accessibility | Keyboard and semantic basics work | Navigate major pages with Tab/Shift+Tab/Enter/Escape; check visible focus, form labels/errors, menu/dialog names and expanded state, button names/pressed states, heading order, and image alternatives. | Interactive controls are keyboard reachable/operable with a visible focus indicator; state is conveyed semantically; form feedback is announced; no focus disappears behind overlays or fixed chrome. | P1 |
+| P1-11 | Motion preferences | Reduced-motion mode remains comfortable | Enable OS/browser reduced motion; reload and test navigation, theme change, article sheet/TOC, carousel, and page reveals. | Essential content and actions remain available; nonessential motion is reduced or absent and no animation blocks interaction. | P1 |
+| P1-12 | SEO / metadata | Route metadata and crawl endpoints are correct | Inspect page title, description, canonical, Open Graph/Twitter tags on home, blog, topic, and article pages. Open `/robots.txt` and `/sitemap.xml`; inspect article JSON-LD. | Titles/descriptions and canonicals identify the current route; article social metadata uses the article content/image where available; structured data is valid JSON; robots allows public pages and disallows `/api/` and `/keystatic/`; sitemap contains indexable routes and published posts, not excluded content. | P1 |
+| P1-13 | Errors / not found | Unknown content and unavailable services fail gracefully | Open a nonexistent article, topic, and category path; submit newsletter while backend configuration is unavailable or network is blocked. | Unknown content shows the app's not-found page and is not rendered as a valid article; newsletter shows an error/retry state; remaining site navigation still works. | P1 |
+| P1-14 | Visual consistency | Shared shell and content styling remain coherent | Compare navigation, page widths, heading hierarchy, links, badges, cards, dividers, footer, and light/dark palette across home, list, topic, article, and legal pages. | Shared elements align and use consistent states; no missing styles, unintended unthemed surface, layout shift, clipped focus ring, or contrast regression. | P1 |
+| P1-15 | Cross-browser | Core journeys pass in supported browser engines | Repeat P0-02, P0-03, P0-04, P0-05 (configured or graceful failure), and P0-06 in current Chrome, Safari, and Firefox; include iOS Safari/Android Chrome when mobile releases matter. | Routing, controls, typography, theme, images, forms, sticky/fixed reading UI, and responsive layouts work consistently; unsupported share/view-transition features degrade without errors. | P1 |
+
+## P2: Broader Confidence Checks
+
+| Test ID | Area | Test Scenario | Steps | Expected Result | Priority |
+|---|---|---|---|---|---|
+| P2-01 | Homepage sections | Lower-page sections and expandable content behave | Scroll through trusted-by, topic grid, article rails/carousel, capabilities, newsletter, enabled FAQ, final CTA, and footer. Operate carousel navigation and each FAQ question if the FAQ is enabled. | Deferred sections appear when reached; carousel cards can be browsed without clipping; each FAQ expands/collapses and reports its state; disabled/flagged sections do not leave awkward gaps. | P2 |
+| P2-02 | Carousel / touch | Article carousel remains usable by pointer and touch | At phone/tablet width, swipe/drag or use exposed carousel controls through multiple cards; move away and return to the section. | Cards remain readable and navigable with no page-level horizontal overflow; carousel pauses when offscreen/backgrounded and resumes without a broken position. | P2 |
+| P2-03 | Scroll utilities | Scroll-to-top control appears and works | On a long page, scroll beyond its reveal threshold, activate Scroll to Top, and repeat in light/dark theme. | Control appears at the intended position, scrolls to top, is not covered by reading controls, and remains legible in both themes. | P2 |
+| P2-04 | Legal and contact content | Informational pages remain complete and actionable | Review About, Contact, Privacy, and Terms; activate team social links and email links; check privacy/terms footer links. | Content is readable and complete; mail links open a mail handler; social links point to the listed profiles; there is no expectation of a contact form or account flow. | P2 |
+| P2-05 | Performance | Production pages meet agreed performance budget | On a production build, run Lighthouse or equivalent on home, blog, topic, and article at mobile and desktop profiles; inspect LCP/CLS/INP and network failures. | No material regression against the team's baseline/budget; above-the-fold content and images load promptly, layout remains stable, and no unexpected heavy request or repeated failing request appears. | P2 |
+| P2-06 | Asset quality | Images and icons survive responsive/theme changes | Check homepage hero, article thumbnails/covers, author avatars, and in-article images at phone and desktop sizes, in both themes. | Correct asset and crop appear, images retain intended aspect ratio without broken placeholders, and informative images have meaningful alternative text. | P2 |
+| P2-07 | Admin gate (conditional) | Keystatic availability matches the environment | In development, open `/keystatic` and a representative content editor; for a local production build with the documented opt-in, verify the gate only in an isolated environment. In normal production, request `/keystatic` and `/api/keystatic/*`. | Editor is available only in intended development/local test conditions; normal production returns not found and does not expose an unauthenticated content-write API. | P2 |
+| P2-08 | Motion / visual polish | Hover and focus states remain stable | On fine-pointer desktop, hover navigation, cards, links, buttons, and theme control; resize while on blog/article; inspect long titles and narrow widths. | Hover effects do not shift surrounding layout or hide content; resize recalculates navigation states; long text wraps/truncates intentionally without overlapping actions. | P2 |
+
+## Run Record
+
+| Release / change | Commit or build | Environment | Browser / device | Tester | Date | P0 result | P1 result | P2 result | Follow-up issues |
+|---|---|---|---|---|---|---|---|---|---|
+| | | | | | | | | | |
diff --git a/morya-techcoder/docs/seo.md b/morya-techcoder/docs/seo.md
index 7c91888..eb7f01c 100644
--- a/morya-techcoder/docs/seo.md
+++ b/morya-techcoder/docs/seo.md
@@ -134,6 +134,10 @@ different purposes and should remain consistent.
## Internal discovery
+`public/llms.txt` provides a concise, curated entry point for AI agents. Keep
+its links aligned with the canonical public routes and populated topic hubs;
+it supplements, but does not replace, the sitemap or ordinary navigation.
+
Sitemaps supplement, but do not replace, links. Public routes connect through
the site navigation, blog/topic listings, and footer. Article pages link their
category badge to `/topics/{category}`, show same-category recommendations in
diff --git a/morya-techcoder/public/content/blog/your-localhost-and-127-0-0-1-aren-t-the-same/localhost-post-thumbnail.png b/morya-techcoder/public/content/blog/your-localhost-and-127-0-0-1-aren-t-the-same/localhost-post-thumbnail.png
new file mode 100644
index 0000000..b3fabe3
Binary files /dev/null and b/morya-techcoder/public/content/blog/your-localhost-and-127-0-0-1-aren-t-the-same/localhost-post-thumbnail.png differ
diff --git a/morya-techcoder/public/llms.txt b/morya-techcoder/public/llms.txt
new file mode 100644
index 0000000..2f7715c
--- /dev/null
+++ b/morya-techcoder/public/llms.txt
@@ -0,0 +1,29 @@
+# TechCoder
+
+> TechCoder is an independent technology publication with practical programming guides, AI coverage, technology explainers, and product reviews for curious readers and developers.
+
+Use published article dates and update dates as context. Treat TechCoder articles as editorial coverage, not as official vendor documentation; follow cited primary sources when exact product behavior or current technical details matter. Draft posts and private CMS/API routes are not public resources.
+
+Article pages are served as regular HTML; separate Markdown versions are not currently published. Start from the blog or a topic page, then fetch relevant article pages as needed.
+
+## Start here
+
+- [Home](https://techcoder.tech/): Overview of TechCoder and links to current coverage.
+- [Articles](https://techcoder.tech/blog): Published article index.
+- [About TechCoder](https://techcoder.tech/about): Publication and editorial information.
+
+## Topics
+
+- [Programming](https://techcoder.tech/topics/programming): Programming languages, frameworks, developer tools, and software engineering.
+- [AI](https://techcoder.tech/topics/ai): AI models, tools, and developer workflows.
+- [Technology](https://techcoder.tech/topics/technology): Technology trends, products, and explainers.
+
+## Site information
+
+- [Contact](https://techcoder.tech/contact): Contact the publication.
+- [Privacy policy](https://techcoder.tech/privacy): Privacy information.
+- [Terms of service](https://techcoder.tech/terms): Site terms.
+
+## Optional
+
+- [XML sitemap](https://techcoder.tech/sitemap.xml): Current indexable public routes, including published articles.
\ No newline at end of file