/ I want to monitize our pets via Instagram TikTok etc. I am building a order platform on petlinks.eu to sell pdf's and merch. I will use Stripe as payment provider. I will handle each animal like petlinks.eu/misty/start and post in each insta account. I will upload the sites structure in next post. I want help with coding as well as marketing 10 Apr Sounds like a fun business! Looking forward to seeing the site structure. A few quick thoughts while I wait: The concept is solid. Pet accounts with dedicated landing pages (petlinks.eu/misty) is a clean way to funnel social traffic to purchases — PDF guides, digital downloads, and merch all work well with Stripe + a lightweight storefront. Once you share the structure, I can help with: Coding — building the landing page templates, Stripe checkout integration, PDF delivery automation, and the routing setup for each pet's subdirectory. Marketing — bio link optimization, content cadence, converting followers to buyers, and cross-promoting between pet accounts. Go ahead and share the structure whenever you're ready! I run nginx and mariadb on a virtualmin server. All mail handled by whizmail.me on the same server /public_html/ index.php ← main router db.php config.php /pages/ start.php buy.php thank-you.php download.php /files/ guide.pdf Show more 10 Apr Great setup — Nginx + MariaDB + Virtualmin is solid for this. Here's how I'd architect the whole thing before we start coding: Routing (index.php) All requests hit index.php which parses the URL to extract the pet slug (/misty/start → pet=misty, page=start), looks up the pet in the DB, then loads the right page from /pages/. Database design — I'd suggest these tables: pets — id, slug, name, species, instagram_url, tiktok_url, stripe_price_id, pdf_file, active orders — id, pet_id, email, stripe_session_id, status, download_token, created_at, downloaded_at Page flow /misty/start → sales page with buy button /misty/buy → creates Stripe Checkout session, redirects to Stripe /misty/thank-you?session_id=xxx → Stripe lands here, verify payment, generate download token, send email /misty/download?token=xxx → validate token, serve PDF securely, mark as downloaded Nginx config — you'll need a rewrite rule so /misty/start hits index.php: nginx location / { try_files $uri $uri/ /index.php?$query_string; } Security considerations PDFs in /files/ should be blocked from direct web access via Nginx Download tokens should be one-time use with expiry (e.g. 48hrs) Stripe webhook to confirm payment server-side (don't trust the redirect alone) Suggested build order: DB schema + db.php + config.php index.php router start.php sales page buy.php Stripe session creator thank-you.php + token generation + email download.php secure PDF delivery Stripe webhook handler Nginx block for /files/ Want to start at step 1 and work through it together? And do you have a Stripe account set up already — test mode or live? I have a Stripe account. Just need to find the details. 10 Apr No rush — take your time finding those. While you do, let's get the foundation built so everything is ready to plug the Stripe keys into. Go ahead and start with step 1? I'll generate: The MariaDB schema (pets + orders tables) config.php (with placeholder for Stripe keys) db.php (PDO connection) Just confirm a couple of things first: Database name — what's your MariaDB DB name for petlinks.eu? (or should I use a placeholder like petlinks_db) PHP mailer — for the download email, are you using PHP's built-in mail(), or do you have something like PHPMailer/SMTP set up via whizmail.me? 1 petlinks 2 second option 10 Apr Do you have PHPMailer installed already, or shall I include Composer setup instructions? And what are the SMTP details for whizmail.me — or should I use placeholders like mail.whizmail.me, port 587, and you'll fill in the credentials? I dont use composer. Use placeholder for whizmail me 10 Apr Got it — I'll include PHPMailer as a direct download (no Composer). Here's the full foundation: Created 3 files, read a file Created 3 files, read a file Here's what you've got: schema.sql — run this in phpMyAdmin or via mysql -u root -p petlinks < schema.sql. There's a commented-out INSERT at the bottom to add your first pet once you have the Stripe price ID. config.php — fill in your DB credentials and Stripe keys when you find them. Everything else is ready. db.php — PDO connection plus all the helper functions the pages will need (get pet, create order, mark paid, handle download token). No SQL scattered across pages. Next step is index.php (the router). Want me to build that now? It'll parse URLs like /misty/start and dispatch to the right page — plus handle 404s for unknown pets or bad slugs. Schema Code · SQL Config Code · PHP Db Code · PHP Yes but name slug pet 10 Apr Got it — pet instead of slug throughout. Building the router now: Index.php main router for petlinks Index.php main router for petlinks Now update db.php — rename get_pet_by_slug to get_pet_by_pet and use the pet column: Rename slug to pet in db.php Rename slug to pet in db.php And update the schema to use pet instead of slug: Edited 2 files, read a file Edited 2 files, read a file All three files updated consistently. Here's what the router does: /misty or /misty/start → loads pages/start.php with $pet available /misty/buy → loads pages/buy.php /misty/thank-you → loads pages/thank-you.php /misty/download → loads pages/download.php Unknown pet → clean 404 Unknown page → clean 404 with link back to the pet's start page Ready to build start.php next — the sales page. Do you want a specific look/feel, or shall I build a clean modern template you can style later? And do you have a price point in mind for the PDFs? Index Code · PHP Db Code · PHP Schema Code · SQL Here is a competitor's website https://ppinteractive.my.canva.site/100?mcp_token=eyJwaWQiOjIzMDIxODUsInNpZCI6NTg0MzYyNDE3LCJheCI6ImE5NmYyY2VhOTRjNGJkYjBlNTZiODM2OWUxMWZlOGFiIiwidHMiOjE3NzU3OTQ5NjEsImV4cCI6MTc3ODIxNDE2MX0.xQE5qnY0HNv_L9n4fJJxTi5GURVlpUtQq7tEFesjj0M 10 Apr Fetched: Untitled App Fetched: Untitled App The page didn't render much content — it's a Canva site which loads everything via JavaScript so I can't scrape the actual content. Could you describe what you saw on it, or share a screenshot? Specifically I'd love to know: What they're selling — PDF guide, membership, merch, bundle? Page layout — hero image, price visible upfront, testimonials, countdown timer? Tone — cute/fun, professional, minimal? Call to action wording — "Buy Now", "Get the Guide", "Download"? That'll let me build a start.php that beats it — better conversion design, cleaner flow, and tailored to your pets' brand. screencapture-ppinteractive-my-canva-site-100-2026-04-10-07_26_33.png 10 Apr Great reference. Now I can see exactly what they're doing. Here's my read: What they have: Warm yellow/orange gradient hero Generic "blueprint" guide — not tied to any specific pet Interactive expandable cards (4 monetization paths) Reasonably good copy but very generic/guru-ish tone Where you beat them: Personal — each page is that specific pet (Misty's guide, not a generic blueprint). Followers already love the animal, that's your conversion advantage Owned platform — they're on Canva, you own your stack, your data, your customer emails Real product — your PDF is specific and useful, not a generic "100+ ideas" listicle Stripe + email — you capture every buyer's email, they likely don't Design direction for your start.php: Same warm, fun energy but pet-specific (Misty's photo, Misty's name in the headline) Pull the pet's Instagram/TikTok from the DB and show follower social proof Clear single CTA — no interactive gimmicks, just scroll → buy Price visible, value clear, one button Ready to build start.php now. Before I do — a couple of quick questions: What's the PDF product? (e.g. "Misty's Care Guide", a pet recipe book, a photo pack?) What price are you thinking — €5–10 range, or higher? Do you want the page in English only, or multilingual later? I want to offer multiple products PDF will be what you suggest work best and you will write,probably "How I'm trying to monetize my pet page" €9-€17 Multi language later Add buyers name to database 10 Apr Love it — that PDF title is perfect. It's meta, honest, and directly relevant to your audience (pet owners who also want to monetize). Here's my thinking on the full product lineup: Product Strategy Tier 1 — €9 — The PDF Guide "How I'm trying to monetize my [Pet Name]'s page — and what's actually working" Written in first person from the pet owner's voice. Relatable, not guru. Covers: affiliate links, brand deals, digital products, print-on-demand merch, Ko-fi/tips, TikTok creator fund. Honest about what flopped too — that builds trust and virality. Tier 2 — €17 — The Bundle PDF Guide + a Canva template pack (Instagram post templates branded to that pet's colors/style). Massively increases perceived value for €8 more. Tier 3 — Merch (later) Print-on-demand via Printful/Printify — stickers, mugs, tote bags. Stripe handles it the same way. Before I build start.php and the updated schema, I need to confirm a few things: Q: For the buy page — single product or let the customer choose a tier? A: Let buyer choose tier on the page Q: Buyer's name — how do you want to collect it? A: Via Stripe checkout (Stripe collects it) 10 Apr Perfect — clean approach. Stripe collects name + email, we pull both back from the session after payment. That means zero friction on your page, and Stripe handles the form. Here's what needs updating before we build: Schema changes: products table (id, pet_id, name, description, stripe_price_id, price, tier, active) orders gets buyer_name, product_id columns Remove stripe_price_id and pdf_file from pets (they move to products) Page flow update: start.php — shows pet intro + two tier cards, buyer clicks one buy.php — receives product_id, creates Stripe session for that specific price thank-you.php — pulls buyer name + email from Stripe session, saves to order download.php — serves the correct PDF for that product Let me rebuild all three foundation files now: Ran 2 commands, created a file, read a file Ran 2 commands, created a file, read a file Both updated. Here's what changed: schema.sql — pets is now clean (no price/PDF), products table handles all tiers per pet, orders has product_id and buyer_name (populated after Stripe confirms). db.php — two new product helpers (get_products_by_pet, get_product_by_id), create_order now takes product_id, mark_order_paid now saves buyer_name pulled from the Stripe session. Ready to build start.php now — the actual sales page. It will: Pull the pet's name, photo placeholder, and social links from DB Show two product tier cards (€9 / €17) with what's included Single "Get it now" button per tier that posts to buy.php Warm, fun design — pet-forward, not guru-ish Do you have a photo/avatar per pet you'll upload to the server, or shall I use the Instagram profile as the image source for now? Schema Code · SQL Db Code · PHP I will upload photos to server 10 Apr Perfect. Where will you store them on the server? For example: /public_html/images/misty.jpg — simple, one photo per pet /public_html/images/misty/ — a folder per pet for multiple photos And what format/naming — just {pet}.jpg, or something else? That'll let me hardcode the correct path into start.php so it just works when you drop the file in. lets start with one photo per pet 10 Apr Got it — so /public_html/images/{pet}.jpg and I'll add a fallback in case the photo isn't uploaded yet. Building start.php now: Viewed a file, ran a command, read a file Viewed a file, ran a command, read a file Here's what the page does: Hero — pet photo (spinning ring effect), name in the headline, Instagram/TikTok buttons pulled from the DB, scroll hint arrow. Truth box — honest tone that sets you apart from guru-style competitors. No hype. What's inside — 6 content cards covering everything the PDF will cover. Products — two tier cards built dynamically from your products DB table. Tier 1 (€9) is the guide, Tier 2 (€17) gets the "Best value" badge and extra inclusions. Both submit to buy.php with the product_id. Trust bar — Stripe badge, instant download, email delivery, real owner. Photo path is /public_html/images/misty.jpg — drops in automatically, falls back to a paw emoji if missing. Ready to build buy.php next — that's the Stripe Checkout session creator. You'll just need your Stripe secret key handy to test it. Shall I go ahead? Start Code · PHP yes 10 Apr Ran a command, read a file Ran a command, read a file Done. Here's what it does: Security — only accepts POST, validates the product_id actually belongs to the pet being viewed (so nobody can craft a URL to buy a different pet's product at the wrong price). Stripe session — creates a hosted Checkout session with the correct stripe_price_id from your DB. Stripe collects name, email, and card — nothing touches your server. Metadata — pet_id, product_id, and pet slug are stored on the Stripe session so thank-you.php and the webhook can retrieve them reliably. Pending order — saves a pending row immediately so you have a record even if the user abandons at Stripe. One thing you need to do manually — download the Stripe PHP library (no Composer). Go to: https://github.com/stripe/stripe-php/releases → download the latest zip → extract to /public_html/vendor/stripe-php/ Ready to build thank-you.php next — this is where Stripe lands the buyer after payment, we verify the session, generate the download token, save the buyer's name and email, and send the download email. Shall I go ahead? Buy Code · PHP yes 10 Apr Ran a command, read a file Ran a command, read a file Done. Here's what it handles: Payment verification — retrieves the Stripe session server-side and checks payment_status === 'paid' before doing anything. The URL alone proves nothing. Refresh protection — if someone reloads the page, it detects the order is already paid and just re-renders without re-sending the email or generating a new token. Buyer name & email — pulled directly from the Stripe session (customer_details), saved to the orders table. Download token — 64-character cryptographically random hex, expires in 48 hours (set in config.php). Email — branded HTML email with a big download button, plain text fallback, mail failure is logged but doesn't break the page. Thank-you page — shows the buyer's name, a direct download button, the email reminder, and links back to the pet's social accounts (nice soft follow prompt). PHPMailer setup reminder — download from https://github.com/PHPMailer/PHPMailer/releases and extract to /public_html/vendor/phpmailer/. One file left in the core flow — download.php. This validates the token, checks expiry, serves the PDF securely, and marks it as downloaded. Ready to build it? Thank you Code · PHP yes 10 Apr Ran 2 commands, read a file Ran 2 commands, read a file Done. Here's every check it runs before a single byte of PDF is sent: 6 layers of validation — in order: Token present in URL Token exists in the database Order belongs to this pet (can't use Misty's token on another pet's page) Payment status is paid Token hasn't expired (48hrs) PDF file actually exists on disk and is readable Security — basename() on the filename prevents path traversal attacks. PDF is streamed in 8KB chunks so it works for large files without memory issues. The /files/ directory itself should be blocked from direct web access in Nginx — we'll add that config shortly. First-download tracking — records downloaded_at timestamp on first download only, so you can see in the DB who has and hasn't downloaded. Error pages — every failure shows a branded, friendly error with a contact email and a link back to the pet's start page. That's the full core flow complete: start → buy → Stripe → thank-you → download Two things left to do: Stripe webhook (webhook.php) — server-side payment confirmation as a safety net Nginx config — block /files/, set up the URL rewriting Want to do the Nginx config next since it's quick, then the webhook? Download Code · PHP MySQL said: pets Duplicate entry '2' for key 'PRIMARY' 10 Apr That means the pets table already exists in your database with data in it. You're running the CREATE TABLE statements on a DB that already has tables from a previous run. Two ways to fix it: Option A — Clean start (recommended if you're still in development) Run this in phpMyAdmin first to wipe everything, then re-run the schema: sql DROP TABLE IF EXISTS orders; DROP TABLE IF EXISTS products; DROP TABLE IF EXISTS pets; Then run schema.sql fresh. Option B — Keep existing data, just add the new tables Run only the parts you're missing. In phpMyAdmin, check which tables already exist and only run the CREATE TABLE blocks for the ones that don't. Likely you just need products and the updated orders. My guess is you ran the original schema.sql (which had slug) and then tried to run the new one on top. Which option suits you — clean wipe or surgical? no to enter contet 10 Apr Got it — you want to insert your first pet and products into the database. What are the details? Pet name (e.g. Misty) Species (cat, dog, rabbit…) Instagram URL TikTok URL And for the products — do you have your Stripe price IDs yet (price_xxx), or shall I write the INSERT statements with placeholders you fill in later? fixed 10 Apr Great! So you have the tables created and data in. Next steps — shall we do the Nginx config now? It's quick and important because right now: /files/ is publicly accessible (anyone can guess the PDF URL) URL rewriting for /misty/start isn't set up yet so the router won't work Just confirm — is your Virtualmin server set up with a separate nginx config per domain, or do you manage it through the Virtualmin panel? virtualmin 10 Apr Perfect. In Virtualmin the nginx config for each domain lives at: /etc/nginx/sites-available/petlinks.eu (or similar — Virtualmin sometimes puts it under /etc/nginx/conf.d/) The easiest way is to add a custom config snippet through Virtualmin itself so it doesn't get overwritten when Virtualmin regenerates the config. In Virtualmin: Go to → petlinks.eu → Server Configuration → Website Options or Edit Nginx Config Directive Add these directives: nginx # Route all requests through index.php location / { try_files $uri $uri/ /index.php?$query_string; } # Block direct access to /files/ — PDFs must go through download.php location /files/ { deny all; return 404; } # Block direct access to PHP config files location ~* ^/(config|db)\.php$ { deny all; return 404; } # Block vendor directory location /vendor/ { deny all; return 404; } # Pass PHP to PHP-FPM (Virtualmin usually has this already — check first) location ~ \.php$ { include snippets/fastcgi-php.conf; fastcgi_pass unix:/run/php/php8.2-fpm.sock; } Before adding — check what PHP version you're running: bash php -v So we use the right FPM socket path (php8.2-fpm.sock vs php8.1-fpm.sock etc). What PHP version are you on? 1775829847737_image.png 10 Apr I can see your Virtualmin setup. You already have the PHP location (\.php(/|$)) and root / location set up. Do it through the UI — no need to touch config files directly. Step 1 — Fix the / location (add try_files) Click the / row → edit it → find the directives field and add: nginx try_files $uri $uri/ /index.php?$query_string; Step 2 — Add /files/ block Click Add a new location and fill in: URL path: /files/ Match type: Sub-directory Then in the directives/options field add: nginx deny all; return 404; Step 3 — Add /vendor/ block (same way) URL path: /vendor/ Match type: Sub-directory Directives: nginx deny all; return 404; Step 4 — Click "URL Re-Writing" from the top icons and make sure it's enabled for the domain. After saving each one, click Apply Changes in Virtualmin to reload Nginx. One question before we move on — click the \.php(/|$) row and tell me what's in the FastCGI or directives field. That'll confirm which PHP-FPM socket you're using so I can make sure buy.php and the webhook will work correctly. maybe easier if you modify the settings file for me server { server_name petlinks.eu www.petlinks.eu mail.petlinks.eu webmail.petlinks.eu > listen 172.16.0.3; listen [2602:ffb6:4:2bfc:f816:3eff:fe83:5f84]; root /home/petlinks/public_html; index index.php index.htm index.html; access_log /var/log/virtualmin/petlinks.eu_access_log; error_log /var/log/virtualmin/petlinks.eu_error_log; fastcgi_param GATEWAY_INTERFACE CGI/1.1; fastcgi_param SERVER_SOFTWARE nginx; fastcgi_param QUERY_STRING $query_string; fastcgi_param REQUEST_METHOD $request_method; fastcgi_param CONTENT_TYPE $content_type; fastcgi_param CONTENT_LENGTH $content_length; fastcgi_param SCRIPT_FILENAME "/home/petlinks/public_html$fastcgi_script_name> fastcgi_param SCRIPT_NAME $fastcgi_script_name; fastcgi_param REQUEST_URI $request_uri; fastcgi_param DOCUMENT_URI $document_uri; fastcgi_param DOCUMENT_ROOT /home/petlinks/public_html; fastcgi_param SERVER_PROTOCOL $server_protocol; fastcgi_param REMOTE_ADDR $remote_addr; fastcgi_param REMOTE_PORT $remote_port; fastcgi_param SERVER_ADDR $server_addr; fastcgi_param SERVER_PORT $server_port; fastcgi_param SERVER_NAME $server_name; fastcgi_param PATH_INFO $fastcgi_path_info; fastcgi_param HTTPS $https; location ^~ /.well-known/ { try_files $uri /; } location ~ "\.php(/|$)" { try_files $uri $fastcgi_script_name =404; default_type application/x-httpd-php; fastcgi_pass unix:/run/php/17757669372724406.sock; } location / { try_files $uri $uri/ /index.php?$query_string; } fastcgi_split_path_info "^(.+\.php)(/.+)$"; location /cgi-bin/ { gzip off; root /home/petlinks/cgi-bin; Show more 10 Apr The / location already has try_files $uri $uri/ /index.php?$query_string — so URL routing is already done! I just need to add the security blocks. Here's the complete modified config: Created a file, read a file Created a file, read a file Your original config was truncated at /cgi-bin/ so I closed it cleanly — check that the rest of your actual file (SSL block etc) is preserved when you paste this in. To apply: Find your config file — likely at /etc/nginx/sites-available/petlinks.eu or check with: bash nginx -T | grep petlinks Replace the content with this file Test and reload: bash nginx -t && systemctl reload nginx What changed — three new blocks added before the existing locations: /files/ → denied, returns 404 /vendor/ → denied, returns 404 config.php and db.php → denied, returns 404 Everything else is identical to your original. The routing was already correct so nothing to change there. Once Nginx is reloaded, shall we build the Stripe webhook — the last piece of the core flow? Petlinks.eu.nginx CONF It was heavily truncated. Here the correct one server { server_name petlinks.eu www.petlinks.eu mail.petlinks.eu webmail.petlinks.eu > listen 172.16.0.3; listen [2602:ffb6:4:2bfc:f816:3eff:fe83:5f84]; root /home/petlinks/public_html; index index.php index.htm index.html; access_log /var/log/virtualmin/petlinks.eu_access_log; error_log /var/log/virtualmin/petlinks.eu_error_log; fastcgi_param GATEWAY_INTERFACE CGI/1.1; fastcgi_param SERVER_SOFTWARE nginx; fastcgi_param QUERY_STRING $query_string; fastcgi_param REQUEST_METHOD $request_method; fastcgi_param CONTENT_TYPE $content_type; fastcgi_param CONTENT_LENGTH $content_length; fastcgi_param SCRIPT_FILENAME "/home/petlinks/public_html$fastcgi_script_name> fastcgi_param SCRIPT_NAME $fastcgi_script_name; fastcgi_param REQUEST_URI $request_uri; fastcgi_param DOCUMENT_URI $document_uri; fastcgi_param DOCUMENT_ROOT /home/petlinks/public_html; fastcgi_param SERVER_PROTOCOL $server_protocol; fastcgi_param REMOTE_ADDR $remote_addr; fastcgi_param REMOTE_PORT $remote_port; fastcgi_param SERVER_ADDR $server_addr; fastcgi_param SERVER_PORT $server_port; fastcgi_param SERVER_NAME $server_name; fastcgi_param PATH_INFO $fastcgi_path_info; fastcgi_param HTTPS $https; location ^~ /.well-known/ { try_files $uri /; } location ~ "\.php(/|$)" { try_files $uri $fastcgi_script_name =404; default_type application/x-httpd-php; fastcgi_pass unix:/run/php/17757669372724406.sock; } location / { try_files $uri $uri/ /index.php?$query_string; } fastcgi_split_path_info "^(.+\.php)(/.+)$"; location /cgi-bin/ { gzip off; root /home/petlinks/cgi-bin; fastcgi_pass unix:/var/fcgiwrap/17757669372724406.sock/socket; fastcgi_param SCRIPT_FILENAME "/home/petlinks$fastcgi_script_name"; fastcgi_param GATEWAY_INTERFACE CGI/1.1; fastcgi_param SERVER_SOFTWARE nginx; fastcgi_param QUERY_STRING $query_string; fastcgi_param REQUEST_METHOD $request_method; fastcgi_param CONTENT_TYPE $content_type; fastcgi_param CONTENT_LENGTH $content_length; fastcgi_param SCRIPT_NAME $fastcgi_script_name; fastcgi_param REQUEST_URI $request_uri; fastcgi_param DOCUMENT_URI $document_uri; fastcgi_param DOCUMENT_ROOT /home/petlinks/public_html; fastcgi_param SERVER_PROTOCOL $server_protocol; fastcgi_param REMOTE_ADDR $remote_addr; fastcgi_param REMOTE_PORT $remote_port; fastcgi_param SERVER_ADDR $server_addr; fastcgi_param SERVER_PORT $server_port; fastcgi_param SERVER_NAME $server_name; fastcgi_param PATH_INFO $fastcgi_path_info; fastcgi_param HTTPS $https; } if ($host = webmail.petlinks.eu) { rewrite "^/(.*)$" "https://petlinks.eu:20000/$1" redirect; } if ($host = admin.petlinks.eu) { rewrite "^/(.*)$" "https://petlinks.eu:10000/$1" redirect; } listen 172.16.0.3:443 ssl; listen [2602:ffb6:4:2bfc:f816:3eff:fe83:5f84]:443 ssl; ssl_certificate /etc/ssl/virtualmin/17757669372724406/ssl.combined; ssl_certificate_key /etc/ssl/virtualmin/17757669372724406/ssl.key; } Show more 10 Apr Got the full config. Here's the complete version with the three security blocks added in the right place: Ran a command, read a file Ran a command, read a file Everything from your original is intact — I only added the three security blocks near the top. To deploy: Find your config file path first: bash nginx -T 2>/dev/null | grep "# configuration file" | grep petlinks Then replace it and reload: bash cp /path/to/petlinks.eu.conf /path/to/petlinks.eu.conf.bak # backup first nano /path/to/petlinks.eu.conf # paste new content nginx -t && systemctl reload nginx Once Nginx is reloaded, ready to build the Stripe webhook — the last piece! Petlinks.eu.nginx CONF ok lets go 10 Apr Ran a command, read a file Ran a command, read a file Done. Here's what it handles: Signature verification — every incoming request is verified against your STRIPE_WEBHOOK_SECRET before anything runs. Fake requests are rejected instantly. Race condition safe — both thank-you.php and the webhook can fire at the same time. Whichever gets there first processes the order, the other sees status = paid and exits cleanly. No duplicate emails, no duplicate tokens. Abandoned checkout — checkout.session.expired marks the pending order as failed so your DB stays clean. Always returns 200 — even on non-critical errors, so Stripe doesn't keep retrying unnecessarily. Real errors are logged to PHP error log. To register the webhook in Stripe: Go to Stripe Dashboard → Developers → Webhooks → Add endpoint URL: https://petlinks.eu/webhook.php Events to select: checkout.session.completed checkout.session.expired Copy the Signing secret (whsec_xxx) into config.php as STRIPE_WEBHOOK_SECRET The full core is now complete. Here's everything we've built: File Status schema.sql ✅ config.php ✅ db.php ✅ index.php ✅ pages/start.php ✅ pages/buy.php ✅ pages/thank-you.php ✅ pages/download.php ✅ webhook.php ✅ Nginx config ✅ What's next — shall we work on the PDF content itself, or set up test mode in Stripe to do an end-to-end test first? Webhook Code · PHP I get error 403 Forbidden on petlinks.eu/misty 10 Apr That's the Nginx security block being too aggressive — the ~ ^/(config|db)\.php$ regex location is fine, but the issue is likely that Nginx is matching /misty against one of the deny rules, or more likely the location / block isn't catching it properly because of a block ordering issue. Let's check — what does your Nginx error log say? bash tail -20 /var/log/virtualmin/petlinks.eu_error_log Paste the output here and I'll pinpoint exactly which block is causing it. That'll be faster than guessing. 2026/04/10 17:48:22 [error] 3080780#3080780: *297736 directory index of "/home/petlinks/public_html/phoebe/" is forbidden, client: 78.83.129.232, server: petlinks.eu, request: "GET /phoebe/ HTTP/1.1", host: "petlinks.eu" 2026/04/10 17:48:12 [error] 3080780#3080780: *297736 directory index of "/home/petlinks/public_html/misty/" is forbidden, client: 78.83.129.232, server: petlinks.eu, request: "GET /misty/ HTTP/1.1", host: "petlinks.eu" Show more 10 Apr The problem is clear — Nginx sees /misty/ as a real directory (because /public_html/misty/ exists as a folder) and tries to serve a directory listing, which is forbidden. The try_files should be routing it to index.php but the physical directory takes priority. Two fixes: Option A — Delete the physical pet directories (recommended) You don't need /public_html/misty/ or /public_html/phoebe/ as real folders. Everything is handled by index.php. Just delete them: bash rm -rf /home/petlinks/public_html/misty rm -rf /home/petlinks/public_html/phoebe Option B — Keep the folders but add autoindex off Add to your Nginx config inside the location / block: nginx location / { try_files $uri $uri/ /index.php?$query_string; autoindex off; } But Option A is cleaner — those directories shouldn't exist at all. The router handles everything virtually. Go with Option A — delete the folders and test again. Does /public_html/misty/ have any files in it you need to keep? Now it is the same landing page text on all pets pages 10 Apr That's because index.php is routing correctly now, but start.php has the "What's inside" checklist and truth box hardcoded — not pulled from the database per pet. But actually — the content should be the same on every pet's page. The guide is the same guide, just branded to that pet. What changes per pet is: Pet name in the headline Pet photo Social links Products/prices (if different per pet) So start.php is working correctly — /misty shows Misty's name and photo, /phoebe shows Phoebe's name and photo. Is the pet name and photo changing between pages, or is literally everything identical including the name? If the name isn't changing, the issue is in index.php — $pet isn't being passed correctly to start.php. Let's check — can you paste what you see in the browser on /misty vs /phoebe? Specifically is the headline different between the two? it is changing. But to be able to test different landing pages we need to make some changes 10 Apr Good — so it's working correctly! The per-pet personalization is doing its job. For testing different landing page styles, we have a few options: Q: How do you want to handle different landing page styles per pet? A: A template column in the pets DB table (e.g. template=1, 2, 3) 10 Apr Clean approach. Here's the plan: Schema change — add template column to pets table: sql ALTER TABLE pets ADD COLUMN template TINYINT UNSIGNED NOT NULL DEFAULT 1; Run that in phpMyAdmin now. index.php change — instead of always loading start.php, it loads start1.php, start2.php etc based on the pet's template value. File structure: /pages/ start1.php ← current design (warm orange/amber) start2.php ← new design start3.php ← new design buy.php thank-you.php download.php The current start.php becomes start1.php. Let me update index.php to handle this, then rename the file: Ran 2 commands, read a file Ran 2 commands, read a file Here's what to do: 1 — Run this in phpMyAdmin: sql ALTER TABLE pets ADD COLUMN template TINYINT UNSIGNED NOT NULL DEFAULT 1; 2 — On your server: Upload new index.php to /public_html/ Rename /public_html/pages/start.php → start1.php (or upload the new start1.php) Delete the old start.php 3 — To assign a template to a pet: sql UPDATE pets SET template = 2 WHERE pet = 'phoebe'; Now when you want to test a new design, just create start2.php in /pages/ and set template = 2 for whichever pet you want to test it on. The fallback always goes to start1.php if the file doesn't exist. Want me to build a start2.php with a completely different design now so you have something to compare against? Index Code · PHP Start1 Code · PHP yes 10 Apr What direction should the second template go? I want it to feel distinctly different from template 1 (warm orange gradient, editorial Fraunces serif, pet photo centred in hero). Q: Pick a vibe for template 2 A: Retro/fun — bright colours, chunky type, sticker-like elements 10 Apr Love it — going full retro fun. Building now: Ran a command, read a file Ran a command, read a file Here's what makes template 2 distinct: Retro/fun design details: Ticker tape scrolling across the top — pet name, download perks on loop Polka dot hero background on hot pink instead of gradient Chunky black borders + hard drop shadows on every card — classic neo-brutalist sticker feel Boogaloo display font — chunky, fun, retro Spinning starburst badge on the pet photo Wobbling circular sticker in the corner ("FROM Misty's OWN OWNER!") Zigzag divider between hero and content Colour-coded cards — each "what's inside" card gets its own accent colour Yellow product section on black — high contrast, eye-catching Press effect buttons — move on hover, push on click To test it: sql UPDATE pets SET template = 2 WHERE pet = 'phoebe'; Then visit /phoebe to see template 2, /misty stays on template 1. Ready to write the PDF content next? Start2 Code · PHP yes 11 Apr Great — let's write the actual PDF. Before I write it I need a few details to make it feel authentic and personal: Q: Which pet should the first guide be written from the perspective of? A: Misty Q: How long have you been running the pet account(s)? A: Just starting out Q: Have you made any money yet from the accounts? A: Not yet — just starting 11 Apr Perfect — honest, relatable, just starting out. That's actually the strongest angle because your audience is in exactly the same position. Writing it now as a proper PDF: Ran 2 commands, viewed a file, created a file Ran 2 commands, viewed a file, created a file Here's the full guide — 10 chapters, properly laid out. Here's what's in it: Structure: Branded header/footer on every page with page numbers Cover page in your warm brown/amber palette Table of contents Verdict table comparing all 6 monetisation methods at a glance Callout boxes highlighting key insights Colour-coded chapter headers Closing message with the petlinks.eu/misty URL Tone: First person, honest, just-starting-out — never pretends to be further along than you are. That's the differentiator vs the guru-style competitor we looked at. Next steps: Upload misty-guide.pdf to /home/petlinks/public_html/files/ on your server In phpMyAdmin update the product row: UPDATE products SET pdf_file = 'misty-guide.pdf' WHERE id = 1; Want me to tweak any of the content, or shall we move on to something else — the Stripe test, the Phoebe version, or the marketing side? Misty guide Document · PDF I am not gonna lie, but you use honest and honestly way to much. So much I don't ven believe you 11 Apr Ha! Fair point — I'll do a rewrite. The word "honest" becomes meaningless when you use it every other sentence. Ironic given the whole angle. Let me rewrite it with the same voice but show the authenticity through the content — the specific details, the admissions, the real numbers — rather than just labelling everything "honest." Ran a command, read a file Ran a command, read a file Rewritten. The authenticity now comes through in specifics — the admission that she hasn't made much yet, the real rate cards, the "this is what I'd tell myself" framing — rather than just slapping the word "honest" on everything and hoping it sticks. Have a read and let me know what to tweak — tone, length, any sections you want expanded or cut. Misty guide Document · PDF 1775876567276_image.png Chaper 10 numbering look strange. 11 Apr I see it — "010" is wrapping onto two lines because chapter 10 has two digits and breaks the layout. Easy fix — I'll widen the number column and handle two-digit numbers properly: Ran 2 commands, read a file Ran 2 commands, read a file Fixed — the number column is now wider so "10" sits on one line properly. Check chapter 10 and let me know if anything else needs tweaking. Misty guide Document · PDF Yellow on white, like the subheadline on firstt page does not work. need to be much darker, maybe grey 11 Apr Good catch — yellow on white is basically invisible. Let me fix the cover subtitle colour: Ran a command, read a file Ran a command, read a file Changed to a dark warm brown-grey ( #6B5B45) — readable on white but still softer than the main title. Check it and let me know if you want it lighter or darker. Misty guide Document · PDF Misty came to us, and got hunted and catched by our Beni dog. We heard her screams as Beni was throwing her back and forth. We rushed out and mange to get Beni to release her. She run under our camper van and was sitting there shaking. We finally got her out, checked she was not bleeding or had broken parts, and carried her out of our garden and released her. 10 minuter later she was back. 11 Apr That's a great origin story — way better than anything I could make up. It's got drama, a near-death experience, a dog villain, a camper van, and a cat who decided she liked you anyway. That last line is the killer detail. Let me rewrite the opening chapter with this: Ran 3 commands, viewed a file, edited a file Ran 3 commands, viewed a file, edited a file That opening is so much stronger now. "Ten minutes later she was back." as its own paragraph will stop readers cold — it's the kind of detail that makes people trust everything that follows because it's clearly real. Also sets up the whole account perfectly: Misty chose you, not the other way around. That's a story people root for. What else needs changing — content, design, or shall we move on to something else? Misty guide Document · PDF I think we should add a photo of Misty. I will find one later 11 Apr Good idea — a photo of Misty on the cover or the first chapter would make it feel completely different from any generic guide. When you have the photo, just upload it here and tell me: Where you want it — cover page, chapter 1, or both Format — jpg or png both work fine I'll drop it into the PDF at the right size and position. If I want to offer petlinks.eu as a service, what is needed for clients to get paid? 11 Apr Great expansion idea. Here's what's needed: The core problem Right now Stripe payments go to your Stripe account. For clients to get paid directly you have two options: Option A — Stripe Connect (the proper way) Stripe Connect lets you act as a platform — clients connect their own Stripe account to yours, and when a sale happens the money goes directly to them. You can take a platform fee automatically. What's needed: Apply for Stripe Connect on your Stripe dashboard Clients create/connect their own Stripe account during onboarding Your code creates charges "on behalf of" their connected account You define your platform fee (e.g. 10–20% per sale) Stripe handles all the payouts, tax reporting per client, etc. Option B — You collect, you pay out (simpler but messy) All money goes to your Stripe, you manually transfer to clients. Gets complicated fast with tax, timing, disputes. Not recommended at scale. What else the platform would need: Client onboarding flow — sign up, connect Stripe, add their pet Admin panel — manage clients, see sales, handle support Client dashboard — their own sales stats, download counts, buyer emails Pricing model for you — monthly subscription, per-sale percentage, or both Terms of service and privacy policy per client Before building any of that — are you thinking: White-label (clients get their own domain like phoebe.com) Subdomain (like phoebe.petlinks.eu) Or just the /pet path like now but for paying clients? That decision shapes the whole architecture. Update copyright year to 2026 on the pdf 11 Apr Ran 2 commands, read a file Ran 2 commands, read a file Done — © 2026 on every page footer now. Misty guide Document · PDF I am setting up Stripe. I made Misty the kitten, but may be better with petlinks? 08:14 For the Stripe account name, go with PetLinks — here's why: One Stripe account covers all pets — Misty, Phoebe, and every future pet all run through the same account. If you used "Misty the Kitten" you'd need a separate Stripe account per pet which gets messy fast. Looks more professional on the buyer's bank statement — they'll see "PetLinks" which is a real business name, not confusing. Future-proof — if you expand to the platform/SaaS direction we discussed, PetLinks is already the right name on the account. Stripe Connect — if you ever onboard other clients, you need a platform identity, not a pet name. The individual pet branding (Misty, Phoebe etc) lives on the landing pages and in the emails — Stripe just needs to be the business behind it all. So: delete the Misty account, create a fresh one under PetLinks. While you're setting it up, the things you'll need from the dashboard are: Publishable key → STRIPE_PUBLIC_KEY in config.php Secret key → STRIPE_SECRET_KEY in config.php Create a Product + Price for each tier → the price_xxx IDs for the database Webhook signing secret → STRIPE_WEBHOOK_SECRET in config.php Should I use the Pre+Built checkout form? 08:19 Yes — that's exactly what we've already built for. The pre-built Stripe Checkout (hosted by Stripe) is the right choice for you because: Why it's the right call: Stripe hosts the payment page — no PCI compliance headaches on your server Collects name + email automatically — which is how we pull the buyer's name into the database Handles card validation, 3D Secure, Apple Pay, Google Pay out of the box Converts well — Stripe has optimised it heavily Works across all countries and currencies without extra config The flow you've already built matches it exactly: start.php → buyer clicks buy → buy.php creates a Stripe Checkout session → buyer lands on Stripe's hosted page → pays → Stripe redirects to your thank-you.php The only alternative would be Stripe Elements — where you embed the card form directly on your own page. More control over design, but more work, more compliance responsibility, and you'd lose the automatic name/email collection we rely on. So stick with the pre-built Checkout. When you're creating the product in Stripe, just make sure you set: Payment mode: One-time (not subscription) Currency: EUR Two separate prices — one at €9, one at €17 Then paste the price_xxx IDs into your database and you're ready to test. Suggested Description? Should I add an image? 08:24 Description for Stripe: Instant PDF download — "How I'm trying to monetize my pet's page and what's actually working." The real story from a pet owner: what was tried, what flopped, and what's actually generating income. Delivered by email immediately after purchase. Keep it factual — Stripe shows this on the checkout page so it just needs to reassure the buyer they're getting what they expect. Image — yes, add one. On the Stripe Checkout page the product image shows next to the product name and price. It makes a real difference to conversion — a blank placeholder looks unfinished and slightly untrustworthy. Best options for now before you have Misty's photo: The PDF cover — take a screenshot of the cover page we built (dark brown, amber title). Clean, professional, shows exactly what they're buying. A simple branded graphic — cream background, Misty's name, paw print. Takes 2 minutes in Canva. Stripe recommends at least 800×800px, square format. JPG or PNG both fine. Once you have an actual photo of Misty, swap it in — a real cat photo on the checkout page will outperform any graphic. Stripe will check the website, petlinks.eu, we need a default page 08:56 Good thinking. Right now petlinks.eu/ just shows a bare holding page. Let me build a proper one: Viewed a file, ran a command Viewed a file, ran a command Now wire it into index.php to replace the inline holding page: Ran a command, read a file Ran a command, read a file Here's what petlinks.eu now shows: Nav — PetLinks logo, clean and sticky Hero — dark brown background, explains what PetLinks is in one sentence What is PetLinks — two-column explainer with feature cards Meet the pets — pulls all active pets from the database dynamically, shows their photo if uploaded, links straight to their guide page How it works — 4-step process, reassures first-time buyers Trust bar — Stripe badge, instant delivery etc Footer — with Privacy Policy and Contact links Stripe's reviewer will land on a real, professional page that clearly explains what the business does. Upload homepage.php to /public_html/ alongside index.php and it will load automatically. Homepage Code · PHP Index Code · PHP index does not load homepage. look at line 13 and forward 09:03 View index.php around line 13 View index.php around line 13 The replacement didn't take — old holding page is still there. Fixing it now: Edited a file, viewed a file, read a file Edited a file, viewed a file, read a file Fixed. Line 15 now does require __DIR__ . '/homepage.php' — upload both index.php and homepage.php to /public_html/ and petlinks.eu/ will load the proper homepage. Index Code · PHP privacy policy and contact link does not work 09:05 Quick fix — those pages don't exist yet. Two options: Add simple inline pages to index.php for /privacy and /contact Create separate privacy.php and contact.php files in /pages/ I'd go with option 2 — keeps things clean. But first, a couple of questions: Contact — do you want a contact form, or just display an email address? Privacy policy — shall I write a GDPR-compliant one for you? I'll need the contact email you want listed on it. form. hello@petlinks.eu 09:06 Building both now: Ran 2 commands Ran 2 commands Now wire both into index.php: Viewed a file, edited a file, ran a command Viewed a file, edited a file, ran a command Three files to upload to /public_html/ — privacy.php, contact.php, and the updated index.php. What you get: /privacy — full GDPR-compliant policy covering data collection, Stripe, email delivery, retention periods, your rights, cookies (none), and contact details. Mentions hello@petlinks.eu throughout. /contact — clean form with name, email, a subject dropdown (download issue, purchase question, refund request, general, other), and message field. Sends to hello@petlinks.eu via PHPMailer with reply-to set to the sender. Shows a branded success card on submission, keeps form values on validation error. Both pages match the PetLinks design — same nav, same fonts, same footer. Privacy Code · PHP Contact Code · PHP Index Code · PHP time to work on Canva Instagram template pack 09:16 Great — this is the €17 bundle item. Before I start designing, a few quick questions: Q: How many templates should be in the pack? A: 15 — premium feel Q: What types of posts should the pack cover? A: Feed + Stories + Reels covers Q: Style direction for the templates? A: Clean neutral — works for any pet account style 09:17 Perfect — 15 clean neutral templates that any pet account can use regardless of their aesthetic. Here's what I'll build: The 15 templates: Feed posts (7): Introduction post — "Meet [Pet Name]" Fun fact post — stat/number highlight Quote/caption post — big text, minimal Before & after — two-panel layout Product/affiliate recommendation Milestone post — "X followers, thank you!" Weekly routine — "A day in [Pet]'s life" Stories (5): 8. Poll story — "Which do you prefer?" 9. Q&A story — "Ask me anything" 10. Swipe up / link story — drive to bio link 11. New post alert — "New on the feed" 12. Behind the scenes — raw/candid frame Reels covers (3): 13. Tutorial/tip cover — "How I do X" 14. Funny/reaction cover — high energy 15. Series cover — "Part 1 of X" I'll build these as a proper PDF guide showing each template with the Canva link structure, placeholder text, colour codes, and font specs — so buyers can recreate them in Canva exactly. Ready to build it? yes 09:18 Ran 2 commands, read a file Ran 2 commands, read a file Here's the full 15-template pack. Each template page includes: Section header — numbered, colour-coded by type (feed/story/reels) Layout preview — visual mock with labelled zones Canva specs — exact canvas size, fonts, colours, layout rules Suggested copy — headline, body and CTA ready to paste Tips — 4 specific tips for getting the most from that format Plus a Getting Started section at the front walking through Canva step by step, font alternatives for free Canva users, and a closing page. Upload this to /public_html/files/misty-bundle.pdf and update your products table: sql UPDATE products SET pdf_file = 'misty-bundle.pdf' WHERE tier = 2 AND pet_id = 1; The €17 bundle now delivers both files — but wait, we currently only send one PDF per order. Want me to update thank-you.php and download.php to handle multiple files per product? Misty canva templates Document · PDF need to add "terms of service" then add contact, terms of service and privacy policy to all pages 11:00 Let me write the Terms of Service first, then add the footer links to all pages in one go. Build terms.php Build terms.php Now create a shared footer include, then update all pages: Create shared footer include Create shared footer include Now update all pages to use the shared footer. Let me do them one by one: Ran 4 commands Ran 4 commands Now wire terms into index.php: Ran a command, read a file Ran a command, read a file Everything updated. Here's what to upload to /public_html/: terms.php — new file footer.php — new shared include index.php — updated with /terms route homepage.php, privacy.php, contact.php — updated footers pages/start1.php, pages/start2.php, pages/thank-you.php — updated footers Every page now has the same dark brown footer with Contact · Terms of Service · Privacy Policy in amber, consistent across the whole site. Note — footer.php lives in /public_html/ and the pages in /pages/ reference it as __DIR__ . '/../footer.php' — so the directory structure matters. Make sure footer.php is at root level, not inside /pages/. Terms Code · PHP Footer Code · PHP Index Code · PHP Homepage Code · PHP Privacy Code · PHP Contact Code · PHP Thank you Code · PHP Start1 Code · PHP Start2 Code · PHP The payment links I get in Stripe looks like this: https://buy.stripe.com/00weVf0r1aiQ2N60cV7EQ00 Claude is AI and can make mistakes. Please double-check responses. Privacy · PHP Copy Privacy Policy — PetLinks

Privacy Policy

Last updated: April 2026

This Privacy Policy explains how PetLinks (petlinks.eu) collects, uses, and protects your personal data when you visit our website or purchase a product. We are committed to handling your data responsibly and in compliance with the General Data Protection Regulation (GDPR).

1. Who we are

PetLinks operates at petlinks.eu. For any privacy-related questions, contact us at [email protected].

2. What data we collect

When you make a purchase, we collect:

We do not collect or store any payment card details. All payment processing is handled by Stripe, Inc. — you can read their privacy policy at stripe.com/privacy.

When you visit our website, our server may log your IP address and browser type as part of standard web server logging. These logs are used only for security and technical purposes and are not linked to your identity.

3. How we use your data

We do not use your data for marketing, and we do not send newsletters or promotional emails unless you explicitly request them.

4. Legal basis for processing

We process your personal data on the basis of contractual necessity — your data is required to fulfil the purchase you made. We also have a legitimate interest in maintaining purchase records for accounting and fraud prevention purposes.

5. How long we keep your data

We retain your purchase record (name, email, product, date) for 3 years from the date of purchase, after which it is deleted. This retention period is necessary to comply with financial record-keeping obligations.

6. Who we share your data with

We share your data only with the following third parties, strictly as necessary to operate our service:

We do not sell, rent, or trade your personal data to any third party.

7. Your rights under GDPR

You have the right to:

To exercise any of these rights, email us at [email protected]. We will respond within 30 days.

8. Cookies

PetLinks does not use tracking cookies or analytics cookies. We do not use Google Analytics or any third-party tracking scripts. The only session-related data stored is what is strictly necessary for Stripe's checkout process to function.

9. Security

All data is transmitted over HTTPS. PDF downloads are protected by one-time secure tokens that expire after 48 hours. Your email and name are stored in a secure database accessible only to us.

10. Changes to this policy

We may update this policy from time to time. The date at the top of this page will always reflect the most recent version. Continued use of the site after changes constitutes acceptance of the upd