What to do next on this site

You are working on ONE specific website: austincoffeeshowdown.com (live at https://austincoffeeshowdown.com/), a Coffee site in the WholeTech network, served from /var/www/austincoffeeshowdown.com/ on the droplet (ssh root@143.198.182.180). Every instruction below is for austincoffeeshowdown.com and no other site — make all changes on austincoffeeshowdown.com. Act like a senior editor and web engineer who knows the Coffee niche cold.

GOAL (on austincoffeeshowdown.com)
Lock down sensitive data — nothing private is publicly reachable — on the site austincoffeeshowdown.com.

WHY IT MATTERS
Make sure a logged-out visitor or bot cannot reach anything sensitive on this site: /.git/, /.env, backups (.bak/.zip/.sql/.tgz), config files, .htpasswd, and any private or admin path must return 403 or 404, never 200. Then scan the served HTML, JS, and JSON for leaked keys, tokens, internal emails, or personal data and remove them.

HOW TO APPROACH IT
1. Confirm nothing sensitive is publicly reachable: as a logged-out visitor, /.git/, /.env, backups (.bak/.zip/.sql/.tgz), config files, .htpasswd, and any private or admin path must return 403 or 404 — never 200.
2. Scan the served HTML, JS, and JSON for leaked secrets, API keys, tokens, internal emails, or personal data, and remove anything that should not be public.
3. Enforce HTTPS and confirm the standard security headers are present (HSTS, X-Frame-Options, X-Content-Type-Options, Referrer-Policy, and a sensible Content-Security-Policy).
4. Harden anything interactive against a determined automated or AI-driven attacker: validate and rate-limit forms, eliminate open redirects and reflected input, and make sure every agent/API endpoint and machine file (AGENTS.md, llms.txt, *.json) returns ONLY public data.

CONSTRAINTS
- Match the site's existing design, fonts, layout and voice — never a generic AI-template look.
- Keep the HTML valid; do not break nav, internal links, the promo bar, analytics, or ad units.
- Use only real, current, accurately-attributed information — no fabricated facts, numbers, or sources.
- Reuse the site's existing Google Analytics and AdSense.

WHAT "DONE" LOOKS LIKE
- A logged-out request to every sensitive path returns 403/404 — verified, not assumed.
- No secret, key, token, internal email, or personal datum appears anywhere in the served HTML, JS, or JSON.
- HTTPS is forced and security headers are confirmed live; the Agents-First surface exposes only public information — the site is mythos-attack-proof.
- Deployed to the droplet and verified on the live URL (screenshot from a Windows machine — never run a browser on the droplet), with a short summary of exactly what changed.

Think through the best approach first, then complete the whole task autonomously before stopping.

You are working on ONE specific website: austincoffeeshowdown.com (live at https://austincoffeeshowdown.com/), a Coffee site in the WholeTech network, served from /var/www/austincoffeeshowdown.com/ on the droplet (ssh root@143.198.182.180). Every instruction below is for austincoffeeshowdown.com and no other site — make all changes on austincoffeeshowdown.com. Act like a senior editor and web engineer who knows the Coffee niche cold.

GOAL (on austincoffeeshowdown.com)
Verify HTTPS, security headers & a safe Agents-First surface — on the site austincoffeeshowdown.com.

WHY IT MATTERS
Confirm the certificate is valid and HTTPS is forced, the standard security headers are present (HSTS, X-Frame-Options, X-Content-Type-Options, Referrer-Policy, CSP), and the Agents-First endpoints expose only public, non-sensitive information — trustworthy to humans and safe for the agents reading it.

HOW TO APPROACH IT
1. Confirm nothing sensitive is publicly reachable: as a logged-out visitor, /.git/, /.env, backups (.bak/.zip/.sql/.tgz), config files, .htpasswd, and any private or admin path must return 403 or 404 — never 200.
2. Scan the served HTML, JS, and JSON for leaked secrets, API keys, tokens, internal emails, or personal data, and remove anything that should not be public.
3. Enforce HTTPS and confirm the standard security headers are present (HSTS, X-Frame-Options, X-Content-Type-Options, Referrer-Policy, and a sensible Content-Security-Policy).
4. Harden anything interactive against a determined automated or AI-driven attacker: validate and rate-limit forms, eliminate open redirects and reflected input, and make sure every agent/API endpoint and machine file (AGENTS.md, llms.txt, *.json) returns ONLY public data.

CONSTRAINTS
- Match the site's existing design, fonts, layout and voice — never a generic AI-template look.
- Keep the HTML valid; do not break nav, internal links, the promo bar, analytics, or ad units.
- Use only real, current, accurately-attributed information — no fabricated facts, numbers, or sources.
- Reuse the site's existing Google Analytics and AdSense.

WHAT "DONE" LOOKS LIKE
- A logged-out request to every sensitive path returns 403/404 — verified, not assumed.
- No secret, key, token, internal email, or personal datum appears anywhere in the served HTML, JS, or JSON.
- HTTPS is forced and security headers are confirmed live; the Agents-First surface exposes only public information — the site is mythos-attack-proof.
- Deployed to the droplet and verified on the live URL (screenshot from a Windows machine — never run a browser on the droplet), with a short summary of exactly what changed.

Think through the best approach first, then complete the whole task autonomously before stopping.

You are working on ONE specific website: austincoffeeshowdown.com (live at https://austincoffeeshowdown.com/), a Coffee site in the WholeTech network, served from /var/www/austincoffeeshowdown.com/ on the droplet (ssh root@143.198.182.180). Every instruction below is for austincoffeeshowdown.com and no other site — make all changes on austincoffeeshowdown.com. Act like a senior editor and web engineer who knows the Coffee niche cold.

GOAL (on austincoffeeshowdown.com)
Build the definitive Coffee cornerstone guide — on the site austincoffeeshowdown.com.

WHY IT MATTERS
Create the single most useful, authoritative page on Coffee anywhere — the resource the whole site points to. Deep, current, expertly structured, and link-worthy. This anchors the site's authority and is the page most likely to earn rankings and citations.

HOW TO APPROACH IT
1. Study the strongest existing pages on this topic anywhere and decide what would make THIS the single most useful page in Coffee — then build exactly that.
2. Write substantial, original content with a clear structure: a strong thesis up top, well-organized sections, concrete real examples and current numbers, and a decisive takeaway.
3. Ground every claim in current, authoritative sources and link out to them inline; never invent facts or statistics.
4. Wire it into the site: link from the homepage and nav, cross-link /sources/ and related pages, and add Article + FAQ schema.

CONSTRAINTS
- Match the site's existing design, fonts, layout and voice — never a generic AI-template look.
- Keep the HTML valid; do not break nav, internal links, the promo bar, analytics, or ad units.
- Use only real, current, accurately-attributed information — no fabricated facts, numbers, or sources.
- Reuse the site's existing Google Analytics and AdSense.

WHAT "DONE" LOOKS LIKE
- The page is genuinely the strongest Coffee resource on the site — deep, accurate, and a pleasure to read, not a thin template.
- Every factual claim is current and traceable to a real, linked source.
- It's discoverable: linked from the homepage, in the sitemap, and submitted via IndexNow.
- Deployed to the droplet and verified on the live URL (screenshot from a Windows machine — never run a browser on the droplet), with a short summary of exactly what changed.

Think through the best approach first, then complete the whole task autonomously before stopping.

You are working on ONE specific website: austincoffeeshowdown.com (live at https://austincoffeeshowdown.com/), a Coffee site in the WholeTech network, served from /var/www/austincoffeeshowdown.com/ on the droplet (ssh root@143.198.182.180). Every instruction below is for austincoffeeshowdown.com and no other site — make all changes on austincoffeeshowdown.com. Act like a senior editor and web engineer who knows the Coffee niche cold.

GOAL (on austincoffeeshowdown.com)
Make the site mythos-attack-proof — harden against automated & AI-agent abuse — on the site austincoffeeshowdown.com.

WHY IT MATTERS
Assume a determined automated, AI-driven attacker. Enforce HTTPS, add a sensible Content-Security-Policy plus the standard security headers, eliminate open redirects and reflected user input, validate and rate-limit any form, and ensure every agent/API endpoint and machine file (AGENTS.md, llms.txt, *.json) returns ONLY public data — never internal or personal information.

HOW TO APPROACH IT
1. Confirm nothing sensitive is publicly reachable: as a logged-out visitor, /.git/, /.env, backups (.bak/.zip/.sql/.tgz), config files, .htpasswd, and any private or admin path must return 403 or 404 — never 200.
2. Scan the served HTML, JS, and JSON for leaked secrets, API keys, tokens, internal emails, or personal data, and remove anything that should not be public.
3. Enforce HTTPS and confirm the standard security headers are present (HSTS, X-Frame-Options, X-Content-Type-Options, Referrer-Policy, and a sensible Content-Security-Policy).
4. Harden anything interactive against a determined automated or AI-driven attacker: validate and rate-limit forms, eliminate open redirects and reflected input, and make sure every agent/API endpoint and machine file (AGENTS.md, llms.txt, *.json) returns ONLY public data.

CONSTRAINTS
- Match the site's existing design, fonts, layout and voice — never a generic AI-template look.
- Keep the HTML valid; do not break nav, internal links, the promo bar, analytics, or ad units.
- Use only real, current, accurately-attributed information — no fabricated facts, numbers, or sources.
- Reuse the site's existing Google Analytics and AdSense.

WHAT "DONE" LOOKS LIKE
- A logged-out request to every sensitive path returns 403/404 — verified, not assumed.
- No secret, key, token, internal email, or personal datum appears anywhere in the served HTML, JS, or JSON.
- HTTPS is forced and security headers are confirmed live; the Agents-First surface exposes only public information — the site is mythos-attack-proof.
- Deployed to the droplet and verified on the live URL (screenshot from a Windows machine — never run a browser on the droplet), with a short summary of exactly what changed.

Think through the best approach first, then complete the whole task autonomously before stopping.

You are working on ONE specific website: austincoffeeshowdown.com (live at https://austincoffeeshowdown.com/), a Coffee site in the WholeTech network, served from /var/www/austincoffeeshowdown.com/ on the droplet (ssh root@143.198.182.180). Every instruction below is for austincoffeeshowdown.com and no other site — make all changes on austincoffeeshowdown.com. Act like a senior editor and web engineer who knows the Coffee niche cold.

GOAL (on austincoffeeshowdown.com)
Make the homepage genuinely distinctive — kill any generic-template feel — on the site austincoffeeshowdown.com.

WHY IT MATTERS
Give the homepage a clear point of view, a memorable hero, and real utility above the fold. A distinctive, intentional design lifts trust, time-on-page, and the Agents-First score out of generic-template territory.

HOW TO APPROACH IT
1. Study the strongest existing pages on this topic anywhere and decide what would make THIS the single most useful page in Coffee — then build exactly that.
2. Write substantial, original content with a clear structure: a strong thesis up top, well-organized sections, concrete real examples and current numbers, and a decisive takeaway.
3. Ground every claim in current, authoritative sources and link out to them inline; never invent facts or statistics.
4. Wire it into the site: link from the homepage and nav, cross-link /sources/ and related pages, and add Article + FAQ schema.

CONSTRAINTS
- Match the site's existing design, fonts, layout and voice — never a generic AI-template look.
- Keep the HTML valid; do not break nav, internal links, the promo bar, analytics, or ad units.
- Use only real, current, accurately-attributed information — no fabricated facts, numbers, or sources.
- Reuse the site's existing Google Analytics and AdSense.

WHAT "DONE" LOOKS LIKE
- The page is genuinely the strongest Coffee resource on the site — deep, accurate, and a pleasure to read, not a thin template.
- Every factual claim is current and traceable to a real, linked source.
- It's discoverable: linked from the homepage, in the sitemap, and submitted via IndexNow.
- Deployed to the droplet and verified on the live URL (screenshot from a Windows machine — never run a browser on the droplet), with a short summary of exactly what changed.

Think through the best approach first, then complete the whole task autonomously before stopping.

You are working on ONE specific website: austincoffeeshowdown.com (live at https://austincoffeeshowdown.com/), a Coffee site in the WholeTech network, served from /var/www/austincoffeeshowdown.com/ on the droplet (ssh root@143.198.182.180). Every instruction below is for austincoffeeshowdown.com and no other site — make all changes on austincoffeeshowdown.com. Act like a senior editor and web engineer who knows the Coffee niche cold.

GOAL (on austincoffeeshowdown.com)
Add an interactive tool and an email capture for Coffee — on the site austincoffeeshowdown.com.

WHY IT MATTERS
Build one genuinely useful interactive element for Coffee (a calculator, checklist, finder, or quiz) plus a tasteful email capture. Tools earn repeat visits, backlinks, and a first-party audience the site actually owns.

HOW TO APPROACH IT
1. Study the strongest existing pages on this topic anywhere and decide what would make THIS the single most useful page in Coffee — then build exactly that.
2. Write substantial, original content with a clear structure: a strong thesis up top, well-organized sections, concrete real examples and current numbers, and a decisive takeaway.
3. Ground every claim in current, authoritative sources and link out to them inline; never invent facts or statistics.
4. Wire it into the site: link from the homepage and nav, cross-link /sources/ and related pages, and add Article + FAQ schema.

CONSTRAINTS
- Match the site's existing design, fonts, layout and voice — never a generic AI-template look.
- Keep the HTML valid; do not break nav, internal links, the promo bar, analytics, or ad units.
- Use only real, current, accurately-attributed information — no fabricated facts, numbers, or sources.
- Reuse the site's existing Google Analytics and AdSense.

WHAT "DONE" LOOKS LIKE
- The page is genuinely the strongest Coffee resource on the site — deep, accurate, and a pleasure to read, not a thin template.
- Every factual claim is current and traceable to a real, linked source.
- It's discoverable: linked from the homepage, in the sitemap, and submitted via IndexNow.
- Deployed to the droplet and verified on the live URL (screenshot from a Windows machine — never run a browser on the droplet), with a short summary of exactly what changed.

Think through the best approach first, then complete the whole task autonomously before stopping.

You are working on ONE specific website: austincoffeeshowdown.com (live at https://austincoffeeshowdown.com/), a Coffee site in the WholeTech network, served from /var/www/austincoffeeshowdown.com/ on the droplet (ssh root@143.198.182.180). Every instruction below is for austincoffeeshowdown.com and no other site — make all changes on austincoffeeshowdown.com. Act like a senior editor and web engineer who knows the Coffee niche cold.

GOAL (on austincoffeeshowdown.com)
Neighborhood coffee crawl guide pages — on the site austincoffeeshowdown.com.

WHY IT MATTERS
Build one crawl page per neighborhood, East Austin, South Congress, Mueller, Domain, ordering three to five real shops into a walkable route with what-to-order notes. Neighborhood plus coffee queries are steady, and shops link to guides that flatter their block. (Fable enhanced action — also on the global worksheet.)

HOW TO APPROACH IT
1. Plan the cleanest way to accomplish this, then execute it well.
2. Use real, current information and match the existing quality of the site.

CONSTRAINTS
- Match the site's existing design, fonts, layout and voice — never a generic AI-template look.
- Keep the HTML valid; do not break nav, internal links, the promo bar, analytics, or ad units.
- Use only real, current, accurately-attributed information — no fabricated facts, numbers, or sources.
- Reuse the site's existing Google Analytics and AdSense.

WHAT "DONE" LOOKS LIKE
- The task is fully done, accurate, and live.
- Deployed to the droplet and verified on the live URL (screenshot from a Windows machine — never run a browser on the droplet), with a short summary of exactly what changed.

Think through the best approach first, then complete the whole task autonomously before stopping.

You are working on ONE specific website: austincoffeeshowdown.com (live at https://austincoffeeshowdown.com/), a Coffee site in the WholeTech network, served from /var/www/austincoffeeshowdown.com/ on the droplet (ssh root@143.198.182.180). Every instruction below is for austincoffeeshowdown.com and no other site — make all changes on austincoffeeshowdown.com. Act like a senior editor and web engineer who knows the Coffee niche cold.

GOAL (on austincoffeeshowdown.com)
Seasonal bracket tournament of Austin coffee shops — on the site austincoffeeshowdown.com.

WHY IT MATTERS
Make the showdown literal: a March-Madness-style bracket seeding real Austin shops, with public voting rounds each season. Shops campaign to their own followers to advance, importing their audiences. Publish matchup pages with honest tasting notes so every round adds evergreen content. (Fable enhanced action — also on the global worksheet.)

HOW TO APPROACH IT
1. Plan the cleanest way to accomplish this, then execute it well.
2. Use real, current information and match the existing quality of the site.

CONSTRAINTS
- Match the site's existing design, fonts, layout and voice — never a generic AI-template look.
- Keep the HTML valid; do not break nav, internal links, the promo bar, analytics, or ad units.
- Use only real, current, accurately-attributed information — no fabricated facts, numbers, or sources.
- Reuse the site's existing Google Analytics and AdSense.

WHAT "DONE" LOOKS LIKE
- The task is fully done, accurate, and live.
- Deployed to the droplet and verified on the live URL (screenshot from a Windows machine — never run a browser on the droplet), with a short summary of exactly what changed.

Think through the best approach first, then complete the whole task autonomously before stopping.

You are working on ONE specific website: austincoffeeshowdown.com (live at https://austincoffeeshowdown.com/), a Coffee site in the WholeTech network, served from /var/www/austincoffeeshowdown.com/ on the droplet (ssh root@143.198.182.180). Every instruction below is for austincoffeeshowdown.com and no other site — make all changes on austincoffeeshowdown.com. Act like a senior editor and web engineer who knows the Coffee niche cold.

GOAL (on austincoffeeshowdown.com)
Publish: A current, data-backed explainer in Coffee — on the site austincoffeeshowdown.com.

WHY IT MATTERS
Trending angle in Coffee. A fresh, data-backed page on this draws search and gives the site a current spine. Link it from the homepage and /sources/.

HOW TO APPROACH IT
1. Research the angle right now using at least 3 authoritative, current sources; capture the real, specific data points.
2. Write a focused page: a clear thesis, the evidence, what it concretely means for a Coffee reader, and a clear next step.
3. Cite and link each source inline and date the piece so its freshness is obvious.
4. Link it from the homepage and /sources/, and submit it via IndexNow.

CONSTRAINTS
- Match the site's existing design, fonts, layout and voice — never a generic AI-template look.
- Keep the HTML valid; do not break nav, internal links, the promo bar, analytics, or ad units.
- Use only real, current, accurately-attributed information — no fabricated facts, numbers, or sources.
- Reuse the site's existing Google Analytics and AdSense.

WHAT "DONE" LOOKS LIKE
- A reader learns something current and genuinely useful they could not get from a thin AI page.
- At least 3 real sources are cited and linked; no fabricated numbers.
- The page is internally linked and submitted for fast indexing.
- Deployed to the droplet and verified on the live URL (screenshot from a Windows machine — never run a browser on the droplet), with a short summary of exactly what changed.

Think through the best approach first, then complete the whole task autonomously before stopping.

You are working on ONE specific website: austincoffeeshowdown.com (live at https://austincoffeeshowdown.com/), a Coffee site in the WholeTech network, served from /var/www/austincoffeeshowdown.com/ on the droplet (ssh root@143.198.182.180). Every instruction below is for austincoffeeshowdown.com and no other site — make all changes on austincoffeeshowdown.com. Act like a senior editor and web engineer who knows the Coffee niche cold.

GOAL (on austincoffeeshowdown.com)
Publish the scoring rubric and invite rebuttals — on the site austincoffeeshowdown.com.

WHY IT MATTERS
Post the exact cupping rubric used in showdown judging, espresso, pour-over, space, service, and score shops against it transparently. Invite baristas to respond on the record. A visible methodology earns trust and gives coffee Twitter and Reddit something concrete to argue about and link. (Fable enhanced action — also on the global worksheet.)

HOW TO APPROACH IT
1. Plan the cleanest way to accomplish this, then execute it well.
2. Use real, current information and match the existing quality of the site.

CONSTRAINTS
- Match the site's existing design, fonts, layout and voice — never a generic AI-template look.
- Keep the HTML valid; do not break nav, internal links, the promo bar, analytics, or ad units.
- Use only real, current, accurately-attributed information — no fabricated facts, numbers, or sources.
- Reuse the site's existing Google Analytics and AdSense.

WHAT "DONE" LOOKS LIKE
- The task is fully done, accurate, and live.
- Deployed to the droplet and verified on the live URL (screenshot from a Windows machine — never run a browser on the droplet), with a short summary of exactly what changed.

Think through the best approach first, then complete the whole task autonomously before stopping.

You are working on ONE specific website: austincoffeeshowdown.com (live at https://austincoffeeshowdown.com/), a Coffee site in the WholeTech network, served from /var/www/austincoffeeshowdown.com/ on the droplet (ssh root@143.198.182.180). Every instruction below is for austincoffeeshowdown.com and no other site — make all changes on austincoffeeshowdown.com. Act like a senior editor and web engineer who knows the Coffee niche cold.

GOAL (on austincoffeeshowdown.com)
Publish: The question your Coffee audience is searching most right now — on the site austincoffeeshowdown.com.

WHY IT MATTERS
Trending angle in Coffee. A fresh, data-backed page on this draws search and gives the site a current spine. Link it from the homepage and /sources/.

HOW TO APPROACH IT
1. Research the angle right now using at least 3 authoritative, current sources; capture the real, specific data points.
2. Write a focused page: a clear thesis, the evidence, what it concretely means for a Coffee reader, and a clear next step.
3. Cite and link each source inline and date the piece so its freshness is obvious.
4. Link it from the homepage and /sources/, and submit it via IndexNow.

CONSTRAINTS
- Match the site's existing design, fonts, layout and voice — never a generic AI-template look.
- Keep the HTML valid; do not break nav, internal links, the promo bar, analytics, or ad units.
- Use only real, current, accurately-attributed information — no fabricated facts, numbers, or sources.
- Reuse the site's existing Google Analytics and AdSense.

WHAT "DONE" LOOKS LIKE
- A reader learns something current and genuinely useful they could not get from a thin AI page.
- At least 3 real sources are cited and linked; no fabricated numbers.
- The page is internally linked and submitted for fast indexing.
- Deployed to the droplet and verified on the live URL (screenshot from a Windows machine — never run a browser on the droplet), with a short summary of exactly what changed.

Think through the best approach first, then complete the whole task autonomously before stopping.

You are working on ONE specific website: austincoffeeshowdown.com (live at https://austincoffeeshowdown.com/), a Coffee site in the WholeTech network, served from /var/www/austincoffeeshowdown.com/ on the droplet (ssh root@143.198.182.180). Every instruction below is for austincoffeeshowdown.com and no other site — make all changes on austincoffeeshowdown.com. Act like a senior editor and web engineer who knows the Coffee niche cold.

GOAL (on austincoffeeshowdown.com)
Publish: A 'what changed in 2026' round-up for Coffee — on the site austincoffeeshowdown.com.

WHY IT MATTERS
Trending angle in Coffee. A fresh, data-backed page on this draws search and gives the site a current spine. Link it from the homepage and /sources/.

HOW TO APPROACH IT
1. Research the angle right now using at least 3 authoritative, current sources; capture the real, specific data points.
2. Write a focused page: a clear thesis, the evidence, what it concretely means for a Coffee reader, and a clear next step.
3. Cite and link each source inline and date the piece so its freshness is obvious.
4. Link it from the homepage and /sources/, and submit it via IndexNow.

CONSTRAINTS
- Match the site's existing design, fonts, layout and voice — never a generic AI-template look.
- Keep the HTML valid; do not break nav, internal links, the promo bar, analytics, or ad units.
- Use only real, current, accurately-attributed information — no fabricated facts, numbers, or sources.
- Reuse the site's existing Google Analytics and AdSense.

WHAT "DONE" LOOKS LIKE
- A reader learns something current and genuinely useful they could not get from a thin AI page.
- At least 3 real sources are cited and linked; no fabricated numbers.
- The page is internally linked and submitted for fast indexing.
- Deployed to the droplet and verified on the live URL (screenshot from a Windows machine — never run a browser on the droplet), with a short summary of exactly what changed.

Think through the best approach first, then complete the whole task autonomously before stopping.

You are working on ONE specific website: austincoffeeshowdown.com (live at https://austincoffeeshowdown.com/), a Coffee site in the WholeTech network, served from /var/www/austincoffeeshowdown.com/ on the droplet (ssh root@143.198.182.180). Every instruction below is for austincoffeeshowdown.com and no other site — make all changes on austincoffeeshowdown.com. Act like a senior editor and web engineer who knows the Coffee niche cold.

GOAL (on austincoffeeshowdown.com)
Home-brew gear guides with Amazon tag — on the site austincoffeeshowdown.com.

WHY IT MATTERS
Write buying guides for recreating shop drinks at home: gooseneck kettles, burr grinders, scales, V60 versus Aeropress, tagged colinfirthfan-20. Tie each guide to a bracket winner's method so recommendations stay concrete rather than generic gear-list filler. (Fable enhanced action — also on the global worksheet.)

HOW TO APPROACH IT
1. Plan the cleanest way to accomplish this, then execute it well.
2. Use real, current information and match the existing quality of the site.

CONSTRAINTS
- Match the site's existing design, fonts, layout and voice — never a generic AI-template look.
- Keep the HTML valid; do not break nav, internal links, the promo bar, analytics, or ad units.
- Use only real, current, accurately-attributed information — no fabricated facts, numbers, or sources.
- Reuse the site's existing Google Analytics and AdSense.

WHAT "DONE" LOOKS LIKE
- The task is fully done, accurate, and live.
- Deployed to the droplet and verified on the live URL (screenshot from a Windows machine — never run a browser on the droplet), with a short summary of exactly what changed.

Think through the best approach first, then complete the whole task autonomously before stopping.

You are working on ONE specific website: austincoffeeshowdown.com (live at https://austincoffeeshowdown.com/), a Coffee site in the WholeTech network, served from /var/www/austincoffeeshowdown.com/ on the droplet (ssh root@143.198.182.180). Every instruction below is for austincoffeeshowdown.com and no other site — make all changes on austincoffeeshowdown.com. Act like a senior editor and web engineer who knows the Coffee niche cold.

GOAL (on austincoffeeshowdown.com)
Labeled featured-roaster sponsorship slot — on the site austincoffeeshowdown.com.

WHY IT MATTERS
Offer one clearly marked Sponsored Roaster slot per season, a banner spot on bracket pages for a local roaster, flat rate, honest labeling, no influence on scores. The inquiry form drafts a Gmail notification only, never auto-sends. (Fable enhanced action — also on the global worksheet.)

HOW TO APPROACH IT
1. Plan the cleanest way to accomplish this, then execute it well.
2. Use real, current information and match the existing quality of the site.

CONSTRAINTS
- Match the site's existing design, fonts, layout and voice — never a generic AI-template look.
- Keep the HTML valid; do not break nav, internal links, the promo bar, analytics, or ad units.
- Use only real, current, accurately-attributed information — no fabricated facts, numbers, or sources.
- Reuse the site's existing Google Analytics and AdSense.

WHAT "DONE" LOOKS LIKE
- The task is fully done, accurate, and live.
- Deployed to the droplet and verified on the live URL (screenshot from a Windows machine — never run a browser on the droplet), with a short summary of exactly what changed.

Think through the best approach first, then complete the whole task autonomously before stopping.

You are working on ONE specific website: austincoffeeshowdown.com (live at https://austincoffeeshowdown.com/), a Coffee site in the WholeTech network, served from /var/www/austincoffeeshowdown.com/ on the droplet (ssh root@143.198.182.180). Every instruction below is for austincoffeeshowdown.com and no other site — make all changes on austincoffeeshowdown.com. Act like a senior editor and web engineer who knows the Coffee niche cold.

GOAL (on austincoffeeshowdown.com)
AdSense on guides, never on bracket votes — on the site austincoffeeshowdown.com.

WHY IT MATTERS
Run shared AdSense pub-7759195213529699 on crawl guides and gear pages where readers research, keeping the voting bracket ad-free so the competition feels editorial and shops keep promoting it. Protecting the flagship feature protects the whole traffic engine. (Fable enhanced action — also on the global worksheet.)

HOW TO APPROACH IT
1. Plan the cleanest way to accomplish this, then execute it well.
2. Use real, current information and match the existing quality of the site.

CONSTRAINTS
- Match the site's existing design, fonts, layout and voice — never a generic AI-template look.
- Keep the HTML valid; do not break nav, internal links, the promo bar, analytics, or ad units.
- Use only real, current, accurately-attributed information — no fabricated facts, numbers, or sources.
- Reuse the site's existing Google Analytics and AdSense.

WHAT "DONE" LOOKS LIKE
- The task is fully done, accurate, and live.
- Deployed to the droplet and verified on the live URL (screenshot from a Windows machine — never run a browser on the droplet), with a short summary of exactly what changed.

Think through the best approach first, then complete the whole task autonomously before stopping.

You are working on ONE specific website: austincoffeeshowdown.com (live at https://austincoffeeshowdown.com/), a Coffee site in the WholeTech network, served from /var/www/austincoffeeshowdown.com/ on the droplet (ssh root@143.198.182.180). Every instruction below is for austincoffeeshowdown.com and no other site — make all changes on austincoffeeshowdown.com. Act like a senior editor and web engineer who knows the Coffee niche cold.

GOAL (on austincoffeeshowdown.com)
Cupping-card and kraft-paper visual system — on the site austincoffeeshowdown.com.

WHY IT MATTERS
Art-direct from the cupping table: kraft-paper backgrounds, coffee-ring stain accents, a flavor-wheel radial graphic for score displays, and chalkboard-menu lettering on section heads. Every visual cue should smell like a roastery, not a generic review blog. (Fable enhanced action — also on the global worksheet.)

HOW TO APPROACH IT
1. Snapshot the current design first (save a dated copy under the site) so there is a rollback point.
2. Apply the change so the site looks distinctive and intentional — never a generic template — then run the Agents-First scorer.
3. Verify on the live render, not on a file:// preview.

CONSTRAINTS
- Match the site's existing design, fonts, layout and voice — never a generic AI-template look.
- Keep the HTML valid; do not break nav, internal links, the promo bar, analytics, or ad units.
- Use only real, current, accurately-attributed information — no fabricated facts, numbers, or sources.
- Reuse the site's existing Google Analytics and AdSense.

WHAT "DONE" LOOKS LIKE
- The Agents-First score is 90+ and the page looks designed rather than defaulted, with a rollback point saved.
- Deployed to the droplet and verified on the live URL (screenshot from a Windows machine — never run a browser on the droplet), with a short summary of exactly what changed.

Think through the best approach first, then complete the whole task autonomously before stopping.

You are working on ONE specific website: austincoffeeshowdown.com (live at https://austincoffeeshowdown.com/), a Coffee site in the WholeTech network, served from /var/www/austincoffeeshowdown.com/ on the droplet (ssh root@143.198.182.180). Every instruction below is for austincoffeeshowdown.com and no other site — make all changes on austincoffeeshowdown.com. Act like a senior editor and web engineer who knows the Coffee niche cold.

GOAL (on austincoffeeshowdown.com)
Deepen the main Coffee guide with current data and examples — on the site austincoffeeshowdown.com.

WHY IT MATTERS
Audit the flagship page; replace anything stale, add concrete numbers and a clear next step. Depth is what separates this from a thin template.

HOW TO APPROACH IT
1. Audit the flagship Coffee page and list exactly what is stale, thin, or vague.
2. Replace stale material with current data, add concrete real examples, tighten the structure, and add a clear next step for the reader.
3. Cross-link to /sources/ and the most related pages.

CONSTRAINTS
- Match the site's existing design, fonts, layout and voice — never a generic AI-template look.
- Keep the HTML valid; do not break nav, internal links, the promo bar, analytics, or ad units.
- Use only real, current, accurately-attributed information — no fabricated facts, numbers, or sources.
- Reuse the site's existing Google Analytics and AdSense.

WHAT "DONE" LOOKS LIKE
- Depth and accuracy match the best competitor page in Coffee.
- No filler and no fabricated specifics — every number is current and sourced.
- Deployed to the droplet and verified on the live URL (screenshot from a Windows machine — never run a browser on the droplet), with a short summary of exactly what changed.

Think through the best approach first, then complete the whole task autonomously before stopping.

You are working on ONE specific website: austincoffeeshowdown.com (live at https://austincoffeeshowdown.com/), a Coffee site in the WholeTech network, served from /var/www/austincoffeeshowdown.com/ on the droplet (ssh root@143.198.182.180). Every instruction below is for austincoffeeshowdown.com and no other site — make all changes on austincoffeeshowdown.com. Act like a senior editor and web engineer who knows the Coffee niche cold.

GOAL (on austincoffeeshowdown.com)
Bracket board styled as letterpress ticket stubs — on the site austincoffeeshowdown.com.

WHY IT MATTERS
Render the tournament as a physical bracket board: each shop seeded on a letterpress-style ticket stub with perforated edge, advancing stubs stamped WINNER in red ink. A distinctive bracket graphic becomes the shareable screenshot that markets each voting round. (Fable enhanced action — also on the global worksheet.)

HOW TO APPROACH IT
1. Snapshot the current design first (save a dated copy under the site) so there is a rollback point.
2. Apply the change so the site looks distinctive and intentional — never a generic template — then run the Agents-First scorer.
3. Verify on the live render, not on a file:// preview.

CONSTRAINTS
- Match the site's existing design, fonts, layout and voice — never a generic AI-template look.
- Keep the HTML valid; do not break nav, internal links, the promo bar, analytics, or ad units.
- Use only real, current, accurately-attributed information — no fabricated facts, numbers, or sources.
- Reuse the site's existing Google Analytics and AdSense.

WHAT "DONE" LOOKS LIKE
- The Agents-First score is 90+ and the page looks designed rather than defaulted, with a rollback point saved.
- Deployed to the droplet and verified on the live URL (screenshot from a Windows machine — never run a browser on the droplet), with a short summary of exactly what changed.

Think through the best approach first, then complete the whole task autonomously before stopping.

You are working on ONE specific website: austincoffeeshowdown.com (live at https://austincoffeeshowdown.com/), a Coffee site in the WholeTech network, served from /var/www/austincoffeeshowdown.com/ on the droplet (ssh root@143.198.182.180). Every instruction below is for austincoffeeshowdown.com and no other site — make all changes on austincoffeeshowdown.com. Act like a senior editor and web engineer who knows the Coffee niche cold.

GOAL (on austincoffeeshowdown.com)
Add an FAQ answering the top questions in Coffee — on the site austincoffeeshowdown.com.

WHY IT MATTERS
Pull the real questions people ask in Coffee and answer them plainly. Mark up with FAQ schema for rich results.

HOW TO APPROACH IT
1. Audit the flagship Coffee page and list exactly what is stale, thin, or vague.
2. Replace stale material with current data, add concrete real examples, tighten the structure, and add a clear next step for the reader.
3. Cross-link to /sources/ and the most related pages.

CONSTRAINTS
- Match the site's existing design, fonts, layout and voice — never a generic AI-template look.
- Keep the HTML valid; do not break nav, internal links, the promo bar, analytics, or ad units.
- Use only real, current, accurately-attributed information — no fabricated facts, numbers, or sources.
- Reuse the site's existing Google Analytics and AdSense.

WHAT "DONE" LOOKS LIKE
- Depth and accuracy match the best competitor page in Coffee.
- No filler and no fabricated specifics — every number is current and sourced.
- Deployed to the droplet and verified on the live URL (screenshot from a Windows machine — never run a browser on the droplet), with a short summary of exactly what changed.

Think through the best approach first, then complete the whole task autonomously before stopping.

You are working on ONE specific website: austincoffeeshowdown.com (live at https://austincoffeeshowdown.com/), a Coffee site in the WholeTech network, served from /var/www/austincoffeeshowdown.com/ on the droplet (ssh root@143.198.182.180). Every instruction below is for austincoffeeshowdown.com and no other site — make all changes on austincoffeeshowdown.com. Act like a senior editor and web engineer who knows the Coffee niche cold.

GOAL (on austincoffeeshowdown.com)
Best-coffee FAQ with dated, cited answers — on the site austincoffeeshowdown.com.

WHY IT MATTERS
Publish FAQ blocks answering best coffee in Austin, best espresso, best study cafe, each citing the current bracket results with a visible date. Add schema markup, run /root/af-deploy.py to 90+, and keep both sitemaps plus IndexNow pings current after each round. (Fable enhanced action — also on the global worksheet.)

HOW TO APPROACH IT
1. Plan the cleanest way to accomplish this, then execute it well.
2. Use real, current information and match the existing quality of the site.

CONSTRAINTS
- Match the site's existing design, fonts, layout and voice — never a generic AI-template look.
- Keep the HTML valid; do not break nav, internal links, the promo bar, analytics, or ad units.
- Use only real, current, accurately-attributed information — no fabricated facts, numbers, or sources.
- Reuse the site's existing Google Analytics and AdSense.

WHAT "DONE" LOOKS LIKE
- The task is fully done, accurate, and live.
- Deployed to the droplet and verified on the live URL (screenshot from a Windows machine — never run a browser on the droplet), with a short summary of exactly what changed.

Think through the best approach first, then complete the whole task autonomously before stopping.

You are working on ONE specific website: austincoffeeshowdown.com (live at https://austincoffeeshowdown.com/), a Coffee site in the WholeTech network, served from /var/www/austincoffeeshowdown.com/ on the droplet (ssh root@143.198.182.180). Every instruction below is for austincoffeeshowdown.com and no other site — make all changes on austincoffeeshowdown.com. Act like a senior editor and web engineer who knows the Coffee niche cold.

GOAL (on austincoffeeshowdown.com)
Sortable shop worksheet with practical columns — on the site austincoffeeshowdown.com.

WHY IT MATTERS
Build a master table of every featured shop, alldomains worksheet style: neighborhood, roaster, hours, wifi, outlets, parking, latest bracket finish, sortable with search chips. Practical columns make it the reference page students and remote workers bookmark. (Fable enhanced action — also on the global worksheet.)

HOW TO APPROACH IT
1. Plan the cleanest way to accomplish this, then execute it well.
2. Use real, current information and match the existing quality of the site.

CONSTRAINTS
- Match the site's existing design, fonts, layout and voice — never a generic AI-template look.
- Keep the HTML valid; do not break nav, internal links, the promo bar, analytics, or ad units.
- Use only real, current, accurately-attributed information — no fabricated facts, numbers, or sources.
- Reuse the site's existing Google Analytics and AdSense.

WHAT "DONE" LOOKS LIKE
- The task is fully done, accurate, and live.
- Deployed to the droplet and verified on the live URL (screenshot from a Windows machine — never run a browser on the droplet), with a short summary of exactly what changed.

Think through the best approach first, then complete the whole task autonomously before stopping.

You are working on ONE specific website: austincoffeeshowdown.com (live at https://austincoffeeshowdown.com/), a Coffee site in the WholeTech network, served from /var/www/austincoffeeshowdown.com/ on the droplet (ssh root@143.198.182.180). Every instruction below is for austincoffeeshowdown.com and no other site — make all changes on austincoffeeshowdown.com. Act like a senior editor and web engineer who knows the Coffee niche cold.

GOAL (on austincoffeeshowdown.com)
Apply a WT-Skins kit to lift the design and Agents-First score — on the site austincoffeeshowdown.com.

WHY IT MATTERS
The page looks generic without a kit. Roll the cluster's kit (snapshot the old design first) to get the AF score into the 90s.

HOW TO APPROACH IT
1. Snapshot the current design first (save a dated copy under the site) so there is a rollback point.
2. Apply the change so the site looks distinctive and intentional — never a generic template — then run the Agents-First scorer.
3. Verify on the live render, not on a file:// preview.

CONSTRAINTS
- Match the site's existing design, fonts, layout and voice — never a generic AI-template look.
- Keep the HTML valid; do not break nav, internal links, the promo bar, analytics, or ad units.
- Use only real, current, accurately-attributed information — no fabricated facts, numbers, or sources.
- Reuse the site's existing Google Analytics and AdSense.

WHAT "DONE" LOOKS LIKE
- The Agents-First score is 90+ and the page looks designed rather than defaulted, with a rollback point saved.
- Deployed to the droplet and verified on the live URL (screenshot from a Windows machine — never run a browser on the droplet), with a short summary of exactly what changed.

Think through the best approach first, then complete the whole task autonomously before stopping.

You are working on ONE specific website: austincoffeeshowdown.com (live at https://austincoffeeshowdown.com/), a Coffee site in the WholeTech network, served from /var/www/austincoffeeshowdown.com/ on the droplet (ssh root@143.198.182.180). Every instruction below is for austincoffeeshowdown.com and no other site — make all changes on austincoffeeshowdown.com. Act like a senior editor and web engineer who knows the Coffee niche cold.

GOAL (on austincoffeeshowdown.com)
Keep the live / news feed updated on cadence (2×/day) — on the site austincoffeeshowdown.com.

WHY IT MATTERS
Confirm the cron is publishing fresh, attributed Coffee items and pinging IndexNow each run.

HOW TO APPROACH IT
1. Confirm or set up the auto-publishing /live/ or /news/ feed for Coffee: attributed, links out, and running on a 2x/day cadence.
2. Make sure each run pings IndexNow and the items are genuinely fresh and relevant.

CONSTRAINTS
- Match the site's existing design, fonts, layout and voice — never a generic AI-template look.
- Keep the HTML valid; do not break nav, internal links, the promo bar, analytics, or ad units.
- Use only real, current, accurately-attributed information — no fabricated facts, numbers, or sources.
- Reuse the site's existing Google Analytics and AdSense.

WHAT "DONE" LOOKS LIKE
- The feed publishes fresh, correctly-attributed Coffee items on schedule and is discovered quickly.
- Deployed to the droplet and verified on the live URL (screenshot from a Windows machine — never run a browser on the droplet), with a short summary of exactly what changed.

Think through the best approach first, then complete the whole task autonomously before stopping.

You are working on ONE specific website: austincoffeeshowdown.com (live at https://austincoffeeshowdown.com/), a Coffee site in the WholeTech network, served from /var/www/austincoffeeshowdown.com/ on the droplet (ssh root@143.198.182.180). Every instruction below is for austincoffeeshowdown.com and no other site — make all changes on austincoffeeshowdown.com. Act like a senior editor and web engineer who knows the Coffee niche cold.

GOAL (on austincoffeeshowdown.com)
Submit updated URLs via IndexNow and refresh the sitemap — on the site austincoffeeshowdown.com.

WHY IT MATTERS
Ping IndexNow for new and changed pages so they're discovered fast; confirm the sitemap lists /act/, /sources/, /videos/.

HOW TO APPROACH IT
1. Implement the specific change (structured data / IndexNow / internal links) precisely and completely.
2. Validate it — rich-results test for schema, confirm the IndexNow submission, or check the internal-link graph.

CONSTRAINTS
- Match the site's existing design, fonts, layout and voice — never a generic AI-template look.
- Keep the HTML valid; do not break nav, internal links, the promo bar, analytics, or ad units.
- Use only real, current, accurately-attributed information — no fabricated facts, numbers, or sources.
- Reuse the site's existing Google Analytics and AdSense.

WHAT "DONE" LOOKS LIKE
- The change validates cleanly and the affected pages are eligible for richer results or faster discovery.
- Deployed to the droplet and verified on the live URL (screenshot from a Windows machine — never run a browser on the droplet), with a short summary of exactly what changed.

Think through the best approach first, then complete the whole task autonomously before stopping.

You are working on ONE specific website: austincoffeeshowdown.com (live at https://austincoffeeshowdown.com/), a Coffee site in the WholeTech network, served from /var/www/austincoffeeshowdown.com/ on the droplet (ssh root@143.198.182.180). Every instruction below is for austincoffeeshowdown.com and no other site — make all changes on austincoffeeshowdown.com. Act like a senior editor and web engineer who knows the Coffee niche cold.

GOAL (on austincoffeeshowdown.com)
Add Article + FAQ structured data — on the site austincoffeeshowdown.com.

WHY IT MATTERS
Schema.org markup earns richer results and helps answer-engines cite the page correctly.

HOW TO APPROACH IT
1. Implement the specific change (structured data / IndexNow / internal links) precisely and completely.
2. Validate it — rich-results test for schema, confirm the IndexNow submission, or check the internal-link graph.

CONSTRAINTS
- Match the site's existing design, fonts, layout and voice — never a generic AI-template look.
- Keep the HTML valid; do not break nav, internal links, the promo bar, analytics, or ad units.
- Use only real, current, accurately-attributed information — no fabricated facts, numbers, or sources.
- Reuse the site's existing Google Analytics and AdSense.

WHAT "DONE" LOOKS LIKE
- The change validates cleanly and the affected pages are eligible for richer results or faster discovery.
- Deployed to the droplet and verified on the live URL (screenshot from a Windows machine — never run a browser on the droplet), with a short summary of exactly what changed.

Think through the best approach first, then complete the whole task autonomously before stopping.

You are working on ONE specific website: austincoffeeshowdown.com (live at https://austincoffeeshowdown.com/), a Coffee site in the WholeTech network, served from /var/www/austincoffeeshowdown.com/ on the droplet (ssh root@143.198.182.180). Every instruction below is for austincoffeeshowdown.com and no other site — make all changes on austincoffeeshowdown.com. Act like a senior editor and web engineer who knows the Coffee niche cold.

GOAL (on austincoffeeshowdown.com)
Strengthen internal links between the core, /sources/ and this worksheet — on the site austincoffeeshowdown.com.

WHY IT MATTERS
A tight internal-link web spreads authority and keeps visitors moving through the site.

HOW TO APPROACH IT
1. Implement the specific change (structured data / IndexNow / internal links) precisely and completely.
2. Validate it — rich-results test for schema, confirm the IndexNow submission, or check the internal-link graph.

CONSTRAINTS
- Match the site's existing design, fonts, layout and voice — never a generic AI-template look.
- Keep the HTML valid; do not break nav, internal links, the promo bar, analytics, or ad units.
- Use only real, current, accurately-attributed information — no fabricated facts, numbers, or sources.
- Reuse the site's existing Google Analytics and AdSense.

WHAT "DONE" LOOKS LIKE
- The change validates cleanly and the affected pages are eligible for richer results or faster discovery.
- Deployed to the droplet and verified on the live URL (screenshot from a Windows machine — never run a browser on the droplet), with a short summary of exactly what changed.

Think through the best approach first, then complete the whole task autonomously before stopping.

You are working on ONE specific website: austincoffeeshowdown.com (live at https://austincoffeeshowdown.com/), a Coffee site in the WholeTech network, served from /var/www/austincoffeeshowdown.com/ on the droplet (ssh root@143.198.182.180). Every instruction below is for austincoffeeshowdown.com and no other site — make all changes on austincoffeeshowdown.com. Act like a senior editor and web engineer who knows the Coffee niche cold.

GOAL (on austincoffeeshowdown.com)
Verify affiliate links and ad placement resolve correctly — on the site austincoffeeshowdown.com.

WHY IT MATTERS
Spot-check Skimlinks / Amazon links and AdSense units; remove anything broken or low-value.

HOW TO APPROACH IT
1. Make the change — verify affiliate links resolve, or add the shared affiliate + AdSense setup.
2. Keep commerce low-key and well-integrated: subtle in-content links, not a bolted-on banner block.

CONSTRAINTS
- Match the site's existing design, fonts, layout and voice — never a generic AI-template look.
- Keep the HTML valid; do not break nav, internal links, the promo bar, analytics, or ad units.
- Use only real, current, accurately-attributed information — no fabricated facts, numbers, or sources.
- Reuse the site's existing Google Analytics and AdSense.

WHAT "DONE" LOOKS LIKE
- Links resolve and can earn, and placement is tasteful and does not hurt the reading experience.
- Deployed to the droplet and verified on the live URL (screenshot from a Windows machine — never run a browser on the droplet), with a short summary of exactly what changed.

Think through the best approach first, then complete the whole task autonomously before stopping.

You are working on ONE specific website: austincoffeeshowdown.com (live at https://austincoffeeshowdown.com/), a Coffee site in the WholeTech network, served from /var/www/austincoffeeshowdown.com/ on the droplet (ssh root@143.198.182.180). Every instruction below is for austincoffeeshowdown.com and no other site — make all changes on austincoffeeshowdown.com. Act like a senior editor and web engineer who knows the Coffee niche cold.

GOAL (on austincoffeeshowdown.com)
Refresh the /videos/ playlist with current, embeddable uploads — on the site austincoffeeshowdown.com.

WHY IT MATTERS
Swap any dead or stale videos; keep only authoritative, non-clickbait Coffee videos. Re-verify embeds.

HOW TO APPROACH IT
1. Find 6-10 authoritative, genuinely useful Coffee videos that ALLOW embedding — verify each embed actually plays.
2. Curate them on /videos/ with short context, and drop anything clickbait, get-rich-quick, or low-substance.

CONSTRAINTS
- Match the site's existing design, fonts, layout and voice — never a generic AI-template look.
- Keep the HTML valid; do not break nav, internal links, the promo bar, analytics, or ad units.
- Use only real, current, accurately-attributed information — no fabricated facts, numbers, or sources.
- Reuse the site's existing Google Analytics and AdSense.

WHAT "DONE" LOOKS LIKE
- Every embed plays, the set is authoritative and on-topic, and the page reuses the site's GA + AdSense.
- Deployed to the droplet and verified on the live URL (screenshot from a Windows machine — never run a browser on the droplet), with a short summary of exactly what changed.

Think through the best approach first, then complete the whole task autonomously before stopping.