Your page is invisible to the machines that decide what gets recommended.
Schema Forge turns any page into copy-paste-ready JSON-LD, llms.txt, and FAQ schema — then validates that your markup matches the page, so Google shows rich results and AI assistants quote you accurately.
Get first access at launch
Waitlist perks: first access in signup order · founding price locked for your first year.
No spam. One email at launch, one if we kill it. That's it.
At a glance
| Problem | Search engines and AI assistants read structured data, not your design. Hand-written JSON-LD breaks silently; most generators stop at one schema type and never check it again. |
|---|---|
| Input | A URL (we read the page) — or a 2-minute form if the page doesn't exist yet |
| Output | Copy-paste-ready JSON-LD + llms.txt + FAQ schema · validation report with line-level errors · schema-vs-page mismatch warnings |
| Time to value | Under 5 minutes |
| Price | Free: 1 site, 5 generations / month · Pro: $9/mo |
| Status | Pre-launch — waitlist open |
What it generates — and what it catches
Generation gets you started. Validation is what keeps you indexed: schema that drifts from your page is treated as spam.
| GENERATES | |
| Page schema | SoftwareApplication · Product + Offer · FAQPage · HowTo · Article · Organization · LocalBusiness · BreadcrumbList — type auto-detected from your content, confirmed by you |
|---|---|
| llms.txt | The plain-text summary AI crawlers read first, generated from the same facts as your JSON-LD so the two never disagree |
| FAQ schema | Builds FAQPage markup from your existing Q&A copy — or shows you which buyer questions your page never answers |
| CATCHES | |
| Syntax errors | Invalid JSON, missing required properties, wrong types — with the exact line and the fix |
| Eligibility gaps | Markup that parses but won't qualify for Google rich results (missing Offer, image, dates) |
| Mismatches | Claims in your schema that aren't on the visible page: stale prices, invented ratings, FAQs that were edited in HTML but not in markup (Pro) |
| Silent breakage | Weekly re-validation (Pro) — catches the redesign or CMS update that wiped your markup three weeks ago |
What the output looks like
Sample output — illustrative, not a real customer's data.
URL example-shop.com/products/desk-mat DETECTED Product → generated Product + Offer + BreadcrumbList ✓ GENERATED <script type="application/ld+json"> …1.2 KB, copy below ✓ GENERATED llms.txt …14 lines, copy below VALIDATION 2 errors · 1 warning E1 offers.price is "29.99" but page shows $24.99 (sale price) → use priceSpecification with both, or update schema on sale E2 aggregateRating present but no reviews visible on page → remove it: invisible ratings risk a Google spam action W1 FAQPage has 2 questions; your page answers 5 → 3 unmarked Q&As found, one click to add E1–E2 are mismatch checks (Pro). Syntax + eligibility: free.
How it works
- Paste a URL. Schema Forge reads the visible page — headings, prices, Q&As, org details — and drafts every schema type the page qualifies for, plus llms.txt.
- Review the facts. Every generated value is shown next to where it appears on your page. Nothing invented: if a fact isn't on the page, it isn't in your markup.
- Paste one script tag. Copy into your page head or CMS custom-code field. Pro re-validates weekly and emails you when markup breaks or drifts from the page.
Pricing
| Free — $0 | Pro — $9/mo | |
| Sites | 1 | Unlimited |
| Generations | 5 / month | Unlimited |
| Schema types | Product · Article · FAQPage | All 8 types |
| llms.txt | ✓ | ✓ |
| Validation | Syntax + rich-result eligibility | + schema-vs-page mismatch diff |
| Monitoring | — | Weekly re-validation + email alerts |
Questions
What is JSON-LD and why should I care?
JSON-LD is the structured-data format search engines and AI crawlers read to understand what your page actually is — a product, an article, a FAQ — and its hard facts: price, ratings, availability. Pages with valid structured data are eligible for rich results on Google and get represented more accurately inside AI answers. Pages without it force machines to guess.
How is this different from Google's Rich Results Test?
Google's tool only validates what you already wrote, only for Google's rich-result types. Schema Forge generates the markup from your page in the first place, covers AI-side files like llms.txt, checks that your schema matches your visible copy, and re-validates weekly so silent breakage gets caught.
Why not just use a free schema generator?
Free generators are fine for day one — they output a static snippet and stop there. The failure mode comes later: prices change, FAQs get edited, a redesign wipes the script tag, and nothing tells you. Schema Forge's value is the validation loop: schema-vs-page mismatch detection and weekly re-checks that catch drift before Google or an AI crawler does.
Which schema types does it generate?
SoftwareApplication, Product with Offer, FAQPage, HowTo, Article, Organization, LocalBusiness, and BreadcrumbList at launch. The generator picks the right type from your page content and asks before assuming.
What is llms.txt?
llms.txt is a plain-text file at your site root that gives AI crawlers a clean, factual summary of what you offer — name, problem, inputs, outputs, pricing, contact. It's the first thing many AI tools read. Schema Forge generates it from the same facts as your JSON-LD so the two never disagree.
Will it exaggerate my page to score better?
No — the opposite. Schema Forge flags claims in your schema that don't appear on the visible page (fake ratings, wrong prices, invented FAQs). Mismatched schema is treated as spam by Google and erodes trust with AI crawlers, so the validator treats honesty as a hard rule.
Do you store my pages?
Pages are fetched on demand, parsed, and discarded — we keep the generated markup and validation results, not your page content. Nothing is used for training. You can delete your account and its data with one email.
Do I need to know how to code?
No. You paste a URL, review the generated facts, and copy one script tag into your page head (or your CMS's custom-code field). The validator tells you in plain language what to change if something breaks later.
When does it launch?
Early access ships within a few weeks of this waitlist validating demand. Signups are invited in order, and founding pricing is locked for your first year. If we decide not to build it, you get one honest email saying so.
Still here? That's usually a yes.
First access in signup order · founding price locked for your first year.