@ontosdk/next 1.3.9 → 1.4.0

package/dist/index.d.mts

@@ -1,5 +1,11 @@
 export { OntoConfig, OntoConfig as OntoConfigType, OntoRoute, OntoRoute as OntoRouteType, PageType, generateLlmsTxt } from './config.mjs';
+export { ontoMiddleware } from './middleware.mjs';
+export { OntoHead } from './OntoHead.mjs';
+export { OntoProvider } from './OntoProvider.mjs';
 export { AIOMethodologySchema, AboutPageSchema, OrganizationSchema, generateAIOMethodologySchema, generateAboutPageSchema, generateOrganizationSchema, generateSchemaForPageType, serializeSchema } from './schemas.mjs';
+import 'next/server';
+import 'react/jsx-runtime';
+import 'react';
 
 interface ExtractionResult {
     markdown: string;

package/dist/index.d.ts

@@ -1,5 +1,11 @@
 export { OntoConfig, OntoConfig as OntoConfigType, OntoRoute, OntoRoute as OntoRouteType, PageType, generateLlmsTxt } from './config.js';
+export { ontoMiddleware } from './middleware.js';
+export { OntoHead } from './OntoHead.js';
+export { OntoProvider } from './OntoProvider.js';
 export { AIOMethodologySchema, AboutPageSchema, OrganizationSchema, generateAIOMethodologySchema, generateAboutPageSchema, generateOrganizationSchema, generateSchemaForPageType, serializeSchema } from './schemas.js';
+import 'next/server';
+import 'react/jsx-runtime';
+import 'react';
 
 interface ExtractionResult {
     markdown: string;

package/dist/index.js

@@ -1,11 +1,15 @@
-"use strict";var P=Object.create;var c=Object.defineProperty;var L=Object.getOwnPropertyDescriptor;var C=Object.getOwnPropertyNames;var H=Object.getPrototypeOf,R=Object.prototype.hasOwnProperty;var v=(t,e)=>{for(var n in e)c(t,n,{get:e[n],enumerable:!0})},d=(t,e,n,o)=>{if(e&&typeof e=="object"||typeof e=="function")for(let i of C(e))!R.call(t,i)&&i!==n&&c(t,i,{get:()=>e[i],enumerable:!(o=L(e,i))||o.enumerable});return t};var f=(t,e,n)=>(n=t!=null?P(H(t)):{},d(e||!t||!t.__esModule?c(n,"default",{value:t,enumerable:!0}):n,t)),I=t=>d(c({},"__esModule",{value:!0}),t);var j={};v(j,{extractContent:()=>x,generateAIOMethodologySchema:()=>u,generateAboutPageSchema:()=>p,generateLlmsTxt:()=>O,generateOrganizationSchema:()=>l,generateSchemaForPageType:()=>w,serializeSchema:()=>z});module.exports=I(j);var y=f(require("cheerio")),S=f(require("turndown")),M=new S.default({headingStyle:"atx",codeBlockStyle:"fenced"});function x(t,e="Generated Output"){let n=t.length,o=y.load(t),i=o("title").text()||o("h1").first().text()||"Untitled Page",g=o('meta[name="description"]').attr("content")||"No description found.",a=[];o('script[type="application/ld+json"]').each((h,$)=>{try{let b=o($).html()||"",A=JSON.parse(b);a.push(A)}catch{}}),o("script, style, noscript, iframe, svg, nav, footer, meta, link, header").remove();let s="";o("main").length>0?s=o("main").html()||"":o("article").length>0?s=o("article").html()||"":s=o("body").html()||"";let k=M.turndown(s),r=[`# ${i}`,`> ${g}`,"",`**Source:** ${e}`,`**Extracted:** ${new Date().toISOString()}`,"","---",""].join(`
-`)+k;a.length>0&&(r+=`
+"use strict";var U=Object.create;var S=Object.defineProperty;var G=Object.getOwnPropertyDescriptor;var D=Object.getOwnPropertyNames;var F=Object.getPrototypeOf,J=Object.prototype.hasOwnProperty;var W=(t,e)=>{for(var n in e)S(t,n,{get:e[n],enumerable:!0})},z=(t,e,n,o)=>{if(e&&typeof e=="object"||typeof e=="function")for(let a of D(e))!J.call(t,a)&&a!==n&&S(t,a,{get:()=>e[a],enumerable:!(o=G(e,a))||o.enumerable});return t};var B=(t,e,n)=>(n=t!=null?U(F(t)):{},z(e||!t||!t.__esModule?S(n,"default",{value:t,enumerable:!0}):n,t)),X=t=>z(S({},"__esModule",{value:!0}),t);var V={};W(V,{OntoHead:()=>E,OntoProvider:()=>_,extractContent:()=>I,generateAIOMethodologySchema:()=>k,generateAboutPageSchema:()=>P,generateLlmsTxt:()=>O,generateOrganizationSchema:()=>C,generateSchemaForPageType:()=>w,ontoMiddleware:()=>v,serializeSchema:()=>A});module.exports=X(V);var R=B(require("cheerio")),M=B(require("turndown")),K=new M.default({headingStyle:"atx",codeBlockStyle:"fenced"});function I(t,e="Generated Output"){let n=t.length,o=R.load(t),a=o("title").text()||o("h1").first().text()||"Untitled Page",h=o('meta[name="description"]').attr("content")||"No description found.",m=[];o('script[type="application/ld+json"]').each((r,c)=>{try{let T=o(c).html()||"",x=JSON.parse(T);m.push(x)}catch{}}),o("script, style, noscript, iframe, svg, nav, footer, meta, link, header").remove();let p="";o("main").length>0?p=o("main").html()||"":o("article").length>0?p=o("article").html()||"":p=o("body").html()||"";let i=K.turndown(p),s=[`# ${a}`,`> ${h}`,"",`**Source:** ${e}`,`**Extracted:** ${new Date().toISOString()}`,"","---",""].join(`
+`)+i;m.length>0&&(s+=`
 
 ---
 ## Structured Data (JSON-LD)
 \`\`\`json
-`,a.forEach(h=>{r+=JSON.stringify(h,null,2)+`
-`}),r+="```\n");let m=r.length,T=n>0?(n-m)/n*100:0;return{markdown:r,metadata:{title:i,description:g,jsonLd:a},stats:{originalHtmlSize:n,markdownSize:m,tokenReductionRatio:T}}}function O(t){let e=[];if(e.push(`# ${t.name}`),e.push(""),e.push(`> ${t.summary}`),e.push(""),t.routes&&t.routes.length>0){e.push("## Key Routes"),e.push("");for(let n of t.routes){let o=`${t.baseUrl}${n.path}`;e.push(`- [${n.path}](${o}): ${n.description}`)}e.push("")}if(t.externalLinks&&t.externalLinks.length>0){e.push("## Resources"),e.push("");for(let n of t.externalLinks)n.description?e.push(`- [${n.title}](${n.url}): ${n.description}`):e.push(`- [${n.title}](${n.url})`);e.push("")}if(t.sections&&t.sections.length>0)for(let n of t.sections)e.push(`## ${n.heading}`),e.push(""),e.push(n.content),e.push("");return e.join(`
+`,m.forEach(r=>{s+=JSON.stringify(r,null,2)+`
+`}),s+="```\n");let f=s.length,y=n>0?(n-f)/n*100:0;return{markdown:s,metadata:{title:a,description:h,jsonLd:m},stats:{originalHtmlSize:n,markdownSize:f,tokenReductionRatio:y}}}function O(t){let e=[];if(e.push(`# ${t.name}`),e.push(""),e.push(`> ${t.summary}`),e.push(""),t.routes&&t.routes.length>0){e.push("## Key Routes"),e.push("");for(let n of t.routes){let o=`${t.baseUrl}${n.path}`;e.push(`- [${n.path}](${o}): ${n.description}`)}e.push("")}if(t.externalLinks&&t.externalLinks.length>0){e.push("## Resources"),e.push("");for(let n of t.externalLinks)n.description?e.push(`- [${n.title}](${n.url}): ${n.description}`):e.push(`- [${n.title}](${n.url})`);e.push("")}if(t.sections&&t.sections.length>0)for(let n of t.sections)e.push(`## ${n.heading}`),e.push(""),e.push(n.content),e.push("");return e.join(`
 `).trim()+`
-`}function u(t,e){return{"@context":"https://schema.org","@type":"HowTo",name:"AIO Score Calculation Methodology",description:"AI Optimization (AIO) Score measures how well a website is optimized for AI agents and LLM crawlers. Scored out of 100 points based on four key metrics.",step:[{"@type":"HowToStep",name:"Content Negotiation",text:"Check if the site responds to Accept: text/markdown header. Weight: 40%. Penalty: -30 points if missing. This ensures AI bots receive optimized content instead of heavy HTML.",position:1},{"@type":"HowToStep",name:"Token Efficiency (React Tax)",text:"Measure the ratio of visible text to total HTML size. Weight: 35%. Penalty: -30 points if HTML > 100KB but text < 1KB. Detects JavaScript-heavy sites that are difficult for AI to parse.",position:2},{"@type":"HowToStep",name:"Structured Data",text:"Verify presence of JSON-LD structured data (Schema.org). Weight: 25%. Penalty: -25 points if missing. Enables AI to confidently extract pricing, products, and entities.",position:3},{"@type":"HowToStep",name:"Semantic HTML",text:"Check for semantic tags like <main> and <article>. Bonus: +15 points if present. Helps AI agents separate navigation from core content.",position:4}]}}function l(t,e){if(!t.organization)return null;let n={"@context":"https://schema.org","@type":"Organization",name:t.organization.name};return t.organization.url&&(n.url=t.organization.url),t.organization.description&&(n.description=t.organization.description),t.organization.logo&&(n.logo=t.organization.logo),t.organization.foundingDate&&(n.foundingDate=t.organization.foundingDate),n}function p(t,e){let n=l(t,e),o={"@context":"https://schema.org","@type":"AboutPage",name:`About ${t.name}`,url:e};return t.summary&&(o.description=t.summary),n&&(o.mainEntity=n),o}function w(t,e,n){switch(t){case"scoring":return u(e,n);case"about":return p(e,n);default:return null}}function z(t){return t?JSON.stringify(t,null,2):null}0&&(module.exports={extractContent,generateAIOMethodologySchema,generateAboutPageSchema,generateLlmsTxt,generateOrganizationSchema,generateSchemaForPageType,serializeSchema});
+`}var l=require("next/server");var L=[{name:"GPTBot",company:"OpenAI"},{name:"ChatGPT-User",company:"OpenAI"},{name:"OAI-SearchBot",company:"OpenAI"},{name:"Googlebot",company:"Google"},{name:"Google-CloudVertexBot",company:"Google"},{name:"Google-Extended",company:"Google"},{name:"GoogleOther",company:"Google"},{name:"ClaudeBot",company:"Anthropic"},{name:"Claude-User",company:"Anthropic"},{name:"anthropic-ai",company:"Anthropic"},{name:"PerplexityBot",company:"Perplexity"},{name:"Perplexity-User",company:"Perplexity"},{name:"Meta-ExternalAgent",company:"Meta"},{name:"Meta-ExternalFetcher",company:"Meta"},{name:"FacebookBot",company:"Meta"},{name:"CCBot",company:"Common Crawl"},{name:"Bytespider",company:"ByteDance"},{name:"Applebot-Extended",company:"Apple"},{name:"cohere-ai",company:"Cohere"},{name:"YouBot",company:"You.com"}],Y=L.map(t=>t.name);function $(t){if(!t)return;let e=t.toLowerCase();return L.find(n=>e.includes(n.name.toLowerCase()))}async function v(t,e){let n=t.headers.get("user-agent"),o=t.nextUrl.clone(),a=$(n),h=t.headers.get("accept")||"",m=!!a,p=h.includes("text/markdown");if(m||p){if(o.pathname.startsWith("/_next"))return l.NextResponse.next();if(o.pathname==="/llms.txt")try{if(e){let r=O(e),c=new l.NextResponse(r,{headers:{"Content-Type":"text/plain; charset=utf-8","Cache-Control":"public, max-age=3600, s-maxage=3600, stale-while-revalidate=86400"}});return a&&c.headers.set("X-Onto-Bot",`${a.name} (${a.company})`),c}else{o.pathname="/llms.txt";let r=l.NextResponse.rewrite(o);return r.headers.set("Content-Type","text/plain; charset=utf-8"),r.headers.set("Cache-Control","public, max-age=3600, s-maxage=3600, stale-while-revalidate=86400"),a&&r.headers.set("X-Onto-Bot",`${a.name} (${a.company})`),r}}catch(r){console.error("[Onto] Failed to generate llms.txt:",r),o.pathname="/llms.txt";let c=l.NextResponse.rewrite(o);return c.headers.set("Content-Type","text/plain; charset=utf-8"),a&&c.headers.set("X-Onto-Bot",`${a.name} (${a.company})`),c}if(o.pathname.includes("."))return l.NextResponse.next();let i=o.pathname;(i==="/"||i==="")&&(i="/index"),i.endsWith("/")&&i!=="/"&&(i=i.slice(0,-1));let g={"Content-Type":"text/markdown; charset=utf-8","Cache-Control":"public, max-age=3600, s-maxage=3600, stale-while-revalidate=86400"};a&&(g["X-Onto-Bot"]=`${a.name} (${a.company})`);let s=process.env.ONTO_API_KEY,f=process.env.ONTO_DASHBOARD_URL||"https://app.buildonto.dev";if(s){fetch(`${f}/api/track`,{method:"POST",headers:{"x-onto-key":s,"Content-Type":"application/json"},body:JSON.stringify({route:o.pathname,userAgent:n,bot:a?a.name:null,company:a?a.company:null})}).catch(()=>{});try{let r=await fetch(`${f}/api/sdk/inject?route=${o.pathname}`,{headers:{"x-onto-key":s},signal:AbortSignal.timeout(1500)});if(r.ok){let{injection:c}=await r.json();if(c){let T=`${o.origin}/.onto${i}.md`,x=await fetch(T);if(x.ok){let j=`${await x.text()}
+
+---
+
+${c}`;return new l.NextResponse(j,{headers:{...g,"X-Onto-Injected":"true"}})}}}}catch(r){console.error("[Onto] Injection failed",r)}}o.pathname=`/.onto${i}.md`;let y=l.NextResponse.rewrite(o);return a&&y.headers.set("X-Onto-Bot",`${a.name} (${a.company})`),y}return l.NextResponse.next()}var H=require("next/navigation"),u=require("react/jsx-runtime");function E(){let e=(0,H.usePathname)();(e==="/"||e==="")&&(e="/index"),e.endsWith("/")&&e!=="/"&&(e=e.slice(0,-1));let n=`/.onto${e}.md`;return(0,u.jsxs)(u.Fragment,{children:[(0,u.jsx)("link",{rel:"alternate",type:"text/markdown",href:n,title:"AI-optimized Markdown version"}),(0,u.jsx)("link",{rel:"alternate",type:"text/plain",href:"/llms.txt",title:"LLM-readable site manifest"})]})}var N=require("next/navigation"),b=require("react");function k(t,e){return{"@context":"https://schema.org","@type":"HowTo",name:"AIO Score Calculation Methodology",description:"AI Optimization (AIO) Score measures how well a website is optimized for AI agents and LLM crawlers. Scored out of 100 points based on four key metrics.",step:[{"@type":"HowToStep",name:"Content Negotiation",text:"Check if the site responds to Accept: text/markdown header. Weight: 40%. Penalty: -30 points if missing. This ensures AI bots receive optimized content instead of heavy HTML.",position:1},{"@type":"HowToStep",name:"Token Efficiency (React Tax)",text:"Measure the ratio of visible text to total HTML size. Weight: 35%. Penalty: -30 points if HTML > 100KB but text < 1KB. Detects JavaScript-heavy sites that are difficult for AI to parse.",position:2},{"@type":"HowToStep",name:"Structured Data",text:"Verify presence of JSON-LD structured data (Schema.org). Weight: 25%. Penalty: -25 points if missing. Enables AI to confidently extract pricing, products, and entities.",position:3},{"@type":"HowToStep",name:"Semantic HTML",text:"Check for semantic tags like <main> and <article>. Bonus: +15 points if present. Helps AI agents separate navigation from core content.",position:4}]}}function C(t,e){if(!t.organization)return null;let n={"@context":"https://schema.org","@type":"Organization",name:t.organization.name};return t.organization.url&&(n.url=t.organization.url),t.organization.description&&(n.description=t.organization.description),t.organization.logo&&(n.logo=t.organization.logo),t.organization.foundingDate&&(n.foundingDate=t.organization.foundingDate),n}function P(t,e){let n=C(t,e),o={"@context":"https://schema.org","@type":"AboutPage",name:`About ${t.name}`,url:e};return t.summary&&(o.description=t.summary),n&&(o.mainEntity=n),o}function w(t,e,n){switch(t){case"scoring":return k(e,n);case"about":return P(e,n);default:return null}}function A(t){return t?JSON.stringify(t,null,2):null}var d=require("react/jsx-runtime");function _({baseUrl:t,children:e,config:n}){let o=(0,N.usePathname)(),a=t.endsWith("/")?t.slice(0,-1):t,h=`${a}${o}?format=md`,m=`${a}${o}`,p=(0,b.useMemo)(()=>n?.routes&&n.routes.find(s=>s.path===o)?.pageType||"default",[n,o]),i=(0,b.useMemo)(()=>{if(!n||p==="default")return null;let g=w(p,n,m);return A(g)},[n,p,m]);return(0,d.jsxs)(d.Fragment,{children:[(0,d.jsx)("link",{rel:"alternate",type:"text/markdown",href:h,title:"AI-optimized Markdown version"}),i&&(0,d.jsx)("script",{type:"application/ld+json",dangerouslySetInnerHTML:{__html:i}}),e]})}0&&(module.exports={OntoHead,OntoProvider,extractContent,generateAIOMethodologySchema,generateAboutPageSchema,generateLlmsTxt,generateOrganizationSchema,generateSchemaForPageType,ontoMiddleware,serializeSchema});
 //# sourceMappingURL=index.js.map

package/dist/index.js.map

@@ -1 +1 @@
-{"version":3,"sources":["../src/index.ts","../src/extractor.ts","../src/config.ts","../src/schemas.ts"],"sourcesContent":["// We cannot use Webpack plugins reliably in Next.js Turbopack due to WorkerError restrictions.\r\n// Users must instead run `npx onto-next` as a postbuild script.\r\nexport { extractContent } from './extractor';\r\nexport { OntoConfig, OntoRoute, generateLlmsTxt } from './config';\r\nexport type { OntoConfig as OntoConfigType, OntoRoute as OntoRouteType, PageType } from './config';\r\nexport {\r\n  generateAIOMethodologySchema,\r\n  generateOrganizationSchema,\r\n  generateAboutPageSchema,\r\n  generateSchemaForPageType,\r\n  serializeSchema\r\n} from './schemas';\r\nexport type {\r\n  AIOMethodologySchema,\r\n  OrganizationSchema,\r\n  AboutPageSchema\r\n} from './schemas';\r\n","import * as cheerio from 'cheerio';\r\nimport TurndownService from 'turndown';\r\n\r\nconst turndownService = new TurndownService({\r\n    headingStyle: 'atx',\r\n    codeBlockStyle: 'fenced',\r\n});\r\n\r\n// Configure turndown to keep some layout or handle semantic tags differently if needed\r\n\r\nexport interface ExtractionResult {\r\n    markdown: string;\r\n    metadata: {\r\n        title: string;\r\n        description: string;\r\n        jsonLd: any[];\r\n    };\r\n    stats: {\r\n        originalHtmlSize: number;\r\n        markdownSize: number;\r\n        tokenReductionRatio: number;\r\n    };\r\n}\r\n\r\n/**\r\n * Extracts pure semantic markdown and metadata from rendered Next.js HTML strings.\r\n * @param html The raw HTML string.\r\n * @param sourceUrl (Optional) the URL this was generated from, to attach as metadata.\r\n * @returns {ExtractionResult} The extracted payload.\r\n */\r\nexport function extractContent(html: string, sourceUrl: string = 'Generated Output'): ExtractionResult {\r\n    const originalSize = html.length;\r\n\r\n    const $ = cheerio.load(html);\r\n\r\n    // 1. Extract Metadata BEFORE removing structure\r\n    const title = $('title').text() || $('h1').first().text() || 'Untitled Page';\r\n    const description = $('meta[name=\"description\"]').attr('content') || 'No description found.';\r\n\r\n    const jsonLdScripts: any[] = [];\r\n    $('script[type=\"application/ld+json\"]').each((_, el) => {\r\n        try {\r\n            const raw = $(el).html() || '';\r\n            const parsed = JSON.parse(raw);\r\n            jsonLdScripts.push(parsed);\r\n        } catch {\r\n            // ignore bad json\r\n        }\r\n    });\r\n\r\n    // 2. Strip noise (React boilerplate, styles, unnecessary tags)\r\n    $('script, style, noscript, iframe, svg, nav, footer, meta, link, header').remove();\r\n\r\n    // Optionally remove typical Next.js hidden wrappers if they don't contain real content.\r\n    // Next.js uses <div id=\"__next\"> but we mostly just want semantic content.\r\n\r\n    // 3. Find the entry point for content\r\n    // Prefer <main> or <article> over <body>\r\n    let contentHtml = '';\r\n    if ($('main').length > 0) {\r\n        contentHtml = $('main').html() || '';\r\n    } else if ($('article').length > 0) {\r\n        contentHtml = $('article').html() || '';\r\n    } else {\r\n        contentHtml = $('body').html() || '';\r\n    }\r\n\r\n    // 4. Convert to Markdown\r\n    let markdown = turndownService.turndown(contentHtml);\r\n\r\n    // 5. Optionally inject Metadata header\r\n    const headerLines = [\r\n        `# ${title}`,\r\n        `> ${description}`,\r\n        ``,\r\n        `**Source:** ${sourceUrl}`,\r\n        `**Extracted:** ${new Date().toISOString()}`,\r\n        ``,\r\n        `---`,\r\n        ``\r\n    ];\r\n\r\n    let finalMarkdown = headerLines.join('\\n') + markdown;\r\n\r\n    // Add JSON-LD section if exists\r\n    if (jsonLdScripts.length > 0) {\r\n        finalMarkdown += '\\n\\n---\\n## Structured Data (JSON-LD)\\n```json\\n';\r\n        jsonLdScripts.forEach(j => {\r\n            finalMarkdown += JSON.stringify(j, null, 2) + '\\n';\r\n        });\r\n        finalMarkdown += '```\\n';\r\n    }\r\n\r\n    const markdownSize = finalMarkdown.length;\r\n    const tokenReductionRatio = originalSize > 0 ? ((originalSize - markdownSize) / originalSize) * 100 : 0;\r\n\r\n    return {\r\n        markdown: finalMarkdown,\r\n        metadata: {\r\n            title,\r\n            description,\r\n            jsonLd: jsonLdScripts\r\n        },\r\n        stats: {\r\n            originalHtmlSize: originalSize,\r\n            markdownSize,\r\n            tokenReductionRatio\r\n        }\r\n    };\r\n}\r\n\r\nexport async function generateStaticPayloads(nextAppDirDir: string, ontoPublicDir: string) {\r\n    const fs = await import('fs');\r\n    const path = await import('path');\r\n    const { glob } = await import('glob');\r\n\r\n    if (!fs.existsSync(nextAppDirDir)) {\r\n        return;\r\n    }\r\n\r\n    const files = await glob('**/*.html', { cwd: nextAppDirDir });\r\n    if (files.length === 0) return;\r\n\r\n    if (!fs.existsSync(ontoPublicDir)) {\r\n        fs.mkdirSync(ontoPublicDir, { recursive: true });\r\n    }\r\n\r\n    let totalFilesProcessed = 0;\r\n\r\n    for (const file of files) {\r\n        const inputPath = path.join(nextAppDirDir, file);\r\n        const outputPathRelative = file.replace(/\\.html$/, '.md');\r\n        const outputPath = path.join(ontoPublicDir, outputPathRelative);\r\n\r\n        try {\r\n            const htmlContent = fs.readFileSync(inputPath, 'utf8');\r\n\r\n            let routeName = file.replace(/\\.html$/, '');\r\n            if (routeName === 'index') routeName = '/';\r\n            else routeName = `/${routeName}`;\r\n\r\n            const result = extractContent(htmlContent, routeName);\r\n\r\n            const outputDir = path.dirname(outputPath);\r\n            if (!fs.existsSync(outputDir)) {\r\n                fs.mkdirSync(outputDir, { recursive: true });\r\n            }\r\n\r\n            fs.writeFileSync(outputPath, result.markdown, 'utf8');\r\n            totalFilesProcessed++;\r\n        } catch (e: any) {\r\n            console.error(`[Onto] Failed to process ${file}: ${e.message}`);\r\n        }\r\n    }\r\n    console.log(`[Onto] Successfully generated ${totalFilesProcessed} semantic markdown endpoints.`);\r\n}\r\n","/**\r\n * Configuration schema for onto.config.ts\r\n * Used to dynamically generate llms.txt and other AI discovery files\r\n */\r\n\r\nexport type PageType = 'scoring' | 'about' | 'default';\r\n\r\nexport interface OntoRoute {\r\n  /**\r\n   * The URL path (e.g., '/docs', '/api/reference')\r\n   */\r\n  path: string;\r\n  /**\r\n   * Description of what this route contains\r\n   */\r\n  description: string;\r\n  /**\r\n   * Optional: Page type for automatic JSON-LD schema injection\r\n   * - 'scoring': Injects Methodology schema with AIO scoring weights (40/35/25)\r\n   * - 'about': Injects Organization/AboutPage schema\r\n   * - 'default': No automatic schema injection\r\n   */\r\n  pageType?: PageType;\r\n}\r\n\r\nexport interface OntoConfig {\r\n  /**\r\n   * The name of your project or site (required)\r\n   * Used as the H1 heading in llms.txt\r\n   */\r\n  name: string;\r\n\r\n  /**\r\n   * A short summary of your project (required)\r\n   * Displayed as a blockquote in llms.txt\r\n   * Should contain key information necessary for understanding the rest of the file\r\n   */\r\n  summary: string;\r\n\r\n  /**\r\n   * The base URL of your site (e.g., 'https://example.com')\r\n   */\r\n  baseUrl: string;\r\n\r\n  /**\r\n   * Optional: Additional sections to include in llms.txt\r\n   * Each section can contain any markdown content\r\n   */\r\n  sections?: {\r\n    heading: string;\r\n    content: string;\r\n  }[];\r\n\r\n  /**\r\n   * Key routes that AI agents should know about\r\n   * These will be formatted as a markdown list in llms.txt\r\n   */\r\n  routes?: OntoRoute[];\r\n\r\n  /**\r\n   * Optional: Links to external resources (documentation, API references, etc.)\r\n   */\r\n  externalLinks?: {\r\n    title: string;\r\n    url: string;\r\n    description?: string;\r\n  }[];\r\n\r\n  /**\r\n   * Optional: Organization information for JSON-LD schemas\r\n   */\r\n  organization?: {\r\n    name: string;\r\n    description?: string;\r\n    url?: string;\r\n    logo?: string;\r\n    foundingDate?: string;\r\n  };\r\n}\r\n\r\n/**\r\n * Generate llms.txt content from OntoConfig\r\n * Follows the llms.txt specification:\r\n * - H1 with project name\r\n * - Blockquote with summary\r\n * - Additional markdown sections\r\n */\r\nexport function generateLlmsTxt(config: OntoConfig): string {\r\n  const lines: string[] = [];\r\n\r\n  // H1: Project name (required)\r\n  lines.push(`# ${config.name}`);\r\n  lines.push('');\r\n\r\n  // Blockquote: Summary (required)\r\n  lines.push(`> ${config.summary}`);\r\n  lines.push('');\r\n\r\n  // Key Routes section (if provided)\r\n  if (config.routes && config.routes.length > 0) {\r\n    lines.push('## Key Routes');\r\n    lines.push('');\r\n    for (const route of config.routes) {\r\n      const fullUrl = `${config.baseUrl}${route.path}`;\r\n      lines.push(`- [${route.path}](${fullUrl}): ${route.description}`);\r\n    }\r\n    lines.push('');\r\n  }\r\n\r\n  // External Links section (if provided)\r\n  if (config.externalLinks && config.externalLinks.length > 0) {\r\n    lines.push('## Resources');\r\n    lines.push('');\r\n    for (const link of config.externalLinks) {\r\n      if (link.description) {\r\n        lines.push(`- [${link.title}](${link.url}): ${link.description}`);\r\n      } else {\r\n        lines.push(`- [${link.title}](${link.url})`);\r\n      }\r\n    }\r\n    lines.push('');\r\n  }\r\n\r\n  // Custom sections (if provided)\r\n  if (config.sections && config.sections.length > 0) {\r\n    for (const section of config.sections) {\r\n      lines.push(`## ${section.heading}`);\r\n      lines.push('');\r\n      lines.push(section.content);\r\n      lines.push('');\r\n    }\r\n  }\r\n\r\n  return lines.join('\\n').trim() + '\\n';\r\n}\r\n","/**\r\n * JSON-LD Schema generators for automatic structured data injection\r\n * Follows Schema.org standards for AI-friendly metadata\r\n */\r\n\r\nimport { OntoConfig } from './config';\r\n\r\n/**\r\n * Standard AIO (AI Optimization) scoring methodology\r\n * Based on the Onto scoring algorithm:\r\n * - Content Negotiation: 40% (30 points)\r\n * - React Tax / Token Efficiency: 35% (30 points)\r\n * - Structured Data: 25% (25 points)\r\n * - Semantic HTML: Bonus (15 points)\r\n */\r\nexport interface AIOMethodologySchema {\r\n  '@context': 'https://schema.org';\r\n  '@type': 'HowTo';\r\n  name: string;\r\n  description: string;\r\n  step: Array<{\r\n    '@type': 'HowToStep';\r\n    name: string;\r\n    text: string;\r\n    position: number;\r\n  }>;\r\n}\r\n\r\n/**\r\n * Generate AIO Scoring Methodology JSON-LD schema\r\n * This explains to AI agents how the scoring system works\r\n */\r\nexport function generateAIOMethodologySchema(\r\n  config: OntoConfig,\r\n  pageUrl: string\r\n): AIOMethodologySchema {\r\n  return {\r\n    '@context': 'https://schema.org',\r\n    '@type': 'HowTo',\r\n    name: 'AIO Score Calculation Methodology',\r\n    description: 'AI Optimization (AIO) Score measures how well a website is optimized for AI agents and LLM crawlers. Scored out of 100 points based on four key metrics.',\r\n    step: [\r\n      {\r\n        '@type': 'HowToStep',\r\n        name: 'Content Negotiation',\r\n        text: 'Check if the site responds to Accept: text/markdown header. Weight: 40%. Penalty: -30 points if missing. This ensures AI bots receive optimized content instead of heavy HTML.',\r\n        position: 1\r\n      },\r\n      {\r\n        '@type': 'HowToStep',\r\n        name: 'Token Efficiency (React Tax)',\r\n        text: 'Measure the ratio of visible text to total HTML size. Weight: 35%. Penalty: -30 points if HTML > 100KB but text < 1KB. Detects JavaScript-heavy sites that are difficult for AI to parse.',\r\n        position: 2\r\n      },\r\n      {\r\n        '@type': 'HowToStep',\r\n        name: 'Structured Data',\r\n        text: 'Verify presence of JSON-LD structured data (Schema.org). Weight: 25%. Penalty: -25 points if missing. Enables AI to confidently extract pricing, products, and entities.',\r\n        position: 3\r\n      },\r\n      {\r\n        '@type': 'HowToStep',\r\n        name: 'Semantic HTML',\r\n        text: 'Check for semantic tags like <main> and <article>. Bonus: +15 points if present. Helps AI agents separate navigation from core content.',\r\n        position: 4\r\n      }\r\n    ]\r\n  };\r\n}\r\n\r\n/**\r\n * Organization schema for About pages\r\n */\r\nexport interface OrganizationSchema {\r\n  '@context': 'https://schema.org';\r\n  '@type': 'Organization';\r\n  name: string;\r\n  url?: string;\r\n  description?: string;\r\n  logo?: string;\r\n  foundingDate?: string;\r\n}\r\n\r\n/**\r\n * Generate Organization JSON-LD schema for About pages\r\n */\r\nexport function generateOrganizationSchema(\r\n  config: OntoConfig,\r\n  pageUrl: string\r\n): OrganizationSchema | null {\r\n  if (!config.organization) {\r\n    return null;\r\n  }\r\n\r\n  const schema: OrganizationSchema = {\r\n    '@context': 'https://schema.org',\r\n    '@type': 'Organization',\r\n    name: config.organization.name\r\n  };\r\n\r\n  if (config.organization.url) {\r\n    schema.url = config.organization.url;\r\n  }\r\n\r\n  if (config.organization.description) {\r\n    schema.description = config.organization.description;\r\n  }\r\n\r\n  if (config.organization.logo) {\r\n    schema.logo = config.organization.logo;\r\n  }\r\n\r\n  if (config.organization.foundingDate) {\r\n    schema.foundingDate = config.organization.foundingDate;\r\n  }\r\n\r\n  return schema;\r\n}\r\n\r\n/**\r\n * AboutPage schema combining Organization and WebPage\r\n */\r\nexport interface AboutPageSchema {\r\n  '@context': 'https://schema.org';\r\n  '@type': 'AboutPage';\r\n  name: string;\r\n  url: string;\r\n  description?: string;\r\n  mainEntity?: OrganizationSchema;\r\n}\r\n\r\n/**\r\n * Generate AboutPage JSON-LD schema\r\n */\r\nexport function generateAboutPageSchema(\r\n  config: OntoConfig,\r\n  pageUrl: string\r\n): AboutPageSchema {\r\n  const orgSchema = generateOrganizationSchema(config, pageUrl);\r\n\r\n  const schema: AboutPageSchema = {\r\n    '@context': 'https://schema.org',\r\n    '@type': 'AboutPage',\r\n    name: `About ${config.name}`,\r\n    url: pageUrl\r\n  };\r\n\r\n  if (config.summary) {\r\n    schema.description = config.summary;\r\n  }\r\n\r\n  if (orgSchema) {\r\n    schema.mainEntity = orgSchema;\r\n  }\r\n\r\n  return schema;\r\n}\r\n\r\n/**\r\n * Determine which schema to generate based on page type\r\n */\r\nexport function generateSchemaForPageType(\r\n  pageType: 'scoring' | 'about' | 'default',\r\n  config: OntoConfig,\r\n  pageUrl: string\r\n): any | null {\r\n  switch (pageType) {\r\n    case 'scoring':\r\n      return generateAIOMethodologySchema(config, pageUrl);\r\n    case 'about':\r\n      return generateAboutPageSchema(config, pageUrl);\r\n    case 'default':\r\n    default:\r\n      return null;\r\n  }\r\n}\r\n\r\n/**\r\n * Serialize schema to JSON-LD script tag content\r\n */\r\nexport function serializeSchema(schema: any | null): string | null {\r\n  if (!schema) {\r\n    return null;\r\n  }\r\n  return JSON.stringify(schema, null, 2);\r\n}\r\n"],"mappings":"0jBAAA,IAAAA,EAAA,GAAAC,EAAAD,EAAA,oBAAAE,EAAA,iCAAAC,EAAA,4BAAAC,EAAA,oBAAAC,EAAA,+BAAAC,EAAA,8BAAAC,EAAA,oBAAAC,IAAA,eAAAC,EAAAT,GCAA,IAAAU,EAAyB,sBACzBC,EAA4B,uBAEtBC,EAAkB,IAAI,EAAAC,QAAgB,CACxC,aAAc,MACd,eAAgB,QACpB,CAAC,EAwBM,SAASC,EAAeC,EAAcC,EAAoB,mBAAsC,CACnG,IAAMC,EAAeF,EAAK,OAEpBG,EAAY,OAAKH,CAAI,EAGrBI,EAAQD,EAAE,OAAO,EAAE,KAAK,GAAKA,EAAE,IAAI,EAAE,MAAM,EAAE,KAAK,GAAK,gBACvDE,EAAcF,EAAE,0BAA0B,EAAE,KAAK,SAAS,GAAK,wBAE/DG,EAAuB,CAAC,EAC9BH,EAAE,oCAAoC,EAAE,KAAK,CAACI,EAAGC,IAAO,CACpD,GAAI,CACA,IAAMC,EAAMN,EAAEK,CAAE,EAAE,KAAK,GAAK,GACtBE,EAAS,KAAK,MAAMD,CAAG,EAC7BH,EAAc,KAAKI,CAAM,CAC7B,MAAQ,CAER,CACJ,CAAC,EAGDP,EAAE,uEAAuE,EAAE,OAAO,EAOlF,IAAIQ,EAAc,GACdR,EAAE,MAAM,EAAE,OAAS,EACnBQ,EAAcR,EAAE,MAAM,EAAE,KAAK,GAAK,GAC3BA,EAAE,SAAS,EAAE,OAAS,EAC7BQ,EAAcR,EAAE,SAAS,EAAE,KAAK,GAAK,GAErCQ,EAAcR,EAAE,MAAM,EAAE,KAAK,GAAK,GAItC,IAAIS,EAAWf,EAAgB,SAASc,CAAW,EAc/CE,EAXgB,CAChB,KAAKT,CAAK,GACV,KAAKC,CAAW,GAChB,GACA,eAAeJ,CAAS,GACxB,kBAAkB,IAAI,KAAK,EAAE,YAAY,CAAC,GAC1C,GACA,MACA,EACJ,EAEgC,KAAK;AAAA,CAAI,EAAIW,EAGzCN,EAAc,OAAS,IACvBO,GAAiB;AAAA;AAAA;AAAA;AAAA;AAAA,EACjBP,EAAc,QAAQQ,GAAK,CACvBD,GAAiB,KAAK,UAAUC,EAAG,KAAM,CAAC,EAAI;AAAA,CAClD,CAAC,EACDD,GAAiB,SAGrB,IAAME,EAAeF,EAAc,OAC7BG,EAAsBd,EAAe,GAAMA,EAAea,GAAgBb,EAAgB,IAAM,EAEtG,MAAO,CACH,SAAUW,EACV,SAAU,CACN,MAAAT,EACA,YAAAC,EACA,OAAQC,CACZ,EACA,MAAO,CACH,iBAAkBJ,EAClB,aAAAa,EACA,oBAAAC,CACJ,CACJ,CACJ,CCtBO,SAASC,EAAgBC,EAA4B,CAC1D,IAAMC,EAAkB,CAAC,EAWzB,GARAA,EAAM,KAAK,KAAKD,EAAO,IAAI,EAAE,EAC7BC,EAAM,KAAK,EAAE,EAGbA,EAAM,KAAK,KAAKD,EAAO,OAAO,EAAE,EAChCC,EAAM,KAAK,EAAE,EAGTD,EAAO,QAAUA,EAAO,OAAO,OAAS,EAAG,CAC7CC,EAAM,KAAK,eAAe,EAC1BA,EAAM,KAAK,EAAE,EACb,QAAWC,KAASF,EAAO,OAAQ,CACjC,IAAMG,EAAU,GAAGH,EAAO,OAAO,GAAGE,EAAM,IAAI,GAC9CD,EAAM,KAAK,MAAMC,EAAM,IAAI,KAAKC,CAAO,MAAMD,EAAM,WAAW,EAAE,CAClE,CACAD,EAAM,KAAK,EAAE,CACf,CAGA,GAAID,EAAO,eAAiBA,EAAO,cAAc,OAAS,EAAG,CAC3DC,EAAM,KAAK,cAAc,EACzBA,EAAM,KAAK,EAAE,EACb,QAAWG,KAAQJ,EAAO,cACpBI,EAAK,YACPH,EAAM,KAAK,MAAMG,EAAK,KAAK,KAAKA,EAAK,GAAG,MAAMA,EAAK,WAAW,EAAE,EAEhEH,EAAM,KAAK,MAAMG,EAAK,KAAK,KAAKA,EAAK,GAAG,GAAG,EAG/CH,EAAM,KAAK,EAAE,CACf,CAGA,GAAID,EAAO,UAAYA,EAAO,SAAS,OAAS,EAC9C,QAAWK,KAAWL,EAAO,SAC3BC,EAAM,KAAK,MAAMI,EAAQ,OAAO,EAAE,EAClCJ,EAAM,KAAK,EAAE,EACbA,EAAM,KAAKI,EAAQ,OAAO,EAC1BJ,EAAM,KAAK,EAAE,EAIjB,OAAOA,EAAM,KAAK;AAAA,CAAI,EAAE,KAAK,EAAI;AAAA,CACnC,CCtGO,SAASK,EACdC,EACAC,EACsB,CACtB,MAAO,CACL,WAAY,qBACZ,QAAS,QACT,KAAM,oCACN,YAAa,2JACb,KAAM,CACJ,CACE,QAAS,YACT,KAAM,sBACN,KAAM,iLACN,SAAU,CACZ,EACA,CACE,QAAS,YACT,KAAM,+BACN,KAAM,4LACN,SAAU,CACZ,EACA,CACE,QAAS,YACT,KAAM,kBACN,KAAM,2KACN,SAAU,CACZ,EACA,CACE,QAAS,YACT,KAAM,gBACN,KAAM,0IACN,SAAU,CACZ,CACF,CACF,CACF,CAkBO,SAASC,EACdF,EACAC,EAC2B,CAC3B,GAAI,CAACD,EAAO,aACV,OAAO,KAGT,IAAMG,EAA6B,CACjC,WAAY,qBACZ,QAAS,eACT,KAAMH,EAAO,aAAa,IAC5B,EAEA,OAAIA,EAAO,aAAa,MACtBG,EAAO,IAAMH,EAAO,aAAa,KAG/BA,EAAO,aAAa,cACtBG,EAAO,YAAcH,EAAO,aAAa,aAGvCA,EAAO,aAAa,OACtBG,EAAO,KAAOH,EAAO,aAAa,MAGhCA,EAAO,aAAa,eACtBG,EAAO,aAAeH,EAAO,aAAa,cAGrCG,CACT,CAiBO,SAASC,EACdJ,EACAC,EACiB,CACjB,IAAMI,EAAYH,EAA2BF,EAAQC,CAAO,EAEtDE,EAA0B,CAC9B,WAAY,qBACZ,QAAS,YACT,KAAM,SAASH,EAAO,IAAI,GAC1B,IAAKC,CACP,EAEA,OAAID,EAAO,UACTG,EAAO,YAAcH,EAAO,SAG1BK,IACFF,EAAO,WAAaE,GAGfF,CACT,CAKO,SAASG,EACdC,EACAP,EACAC,EACY,CACZ,OAAQM,EAAU,CAChB,IAAK,UACH,OAAOR,EAA6BC,EAAQC,CAAO,EACrD,IAAK,QACH,OAAOG,EAAwBJ,EAAQC,CAAO,EAEhD,QACE,OAAO,IACX,CACF,CAKO,SAASO,EAAgBL,EAAmC,CACjE,OAAKA,EAGE,KAAK,UAAUA,EAAQ,KAAM,CAAC,EAF5B,IAGX","names":["index_exports","__export","extractContent","generateAIOMethodologySchema","generateAboutPageSchema","generateLlmsTxt","generateOrganizationSchema","generateSchemaForPageType","serializeSchema","__toCommonJS","cheerio","import_turndown","turndownService","TurndownService","extractContent","html","sourceUrl","originalSize","$","title","description","jsonLdScripts","_","el","raw","parsed","contentHtml","markdown","finalMarkdown","j","markdownSize","tokenReductionRatio","generateLlmsTxt","config","lines","route","fullUrl","link","section","generateAIOMethodologySchema","config","pageUrl","generateOrganizationSchema","schema","generateAboutPageSchema","orgSchema","generateSchemaForPageType","pageType","serializeSchema"]}
+{"version":3,"sources":["../src/index.ts","../src/extractor.ts","../src/config.ts","../src/middleware.ts","../src/bots.ts","../src/OntoHead.tsx","../src/OntoProvider.tsx","../src/schemas.ts"],"sourcesContent":["// We cannot use Webpack plugins reliably in Next.js Turbopack due to WorkerError restrictions.\r\n// Users must instead run `npx onto-next` as a postbuild script.\r\nexport { extractContent } from './extractor';\r\nexport { OntoConfig, OntoRoute, generateLlmsTxt } from './config';\r\nexport type { OntoConfig as OntoConfigType, OntoRoute as OntoRouteType, PageType } from './config';\r\nexport { ontoMiddleware } from './middleware';\r\nexport { OntoHead } from './OntoHead';\r\nexport { OntoProvider } from './OntoProvider';\r\nexport {\r\n  generateAIOMethodologySchema,\r\n  generateOrganizationSchema,\r\n  generateAboutPageSchema,\r\n  generateSchemaForPageType,\r\n  serializeSchema\r\n} from './schemas';\r\nexport type {\r\n  AIOMethodologySchema,\r\n  OrganizationSchema,\r\n  AboutPageSchema\r\n} from './schemas';\r\n","import * as cheerio from 'cheerio';\r\nimport TurndownService from 'turndown';\r\n\r\nconst turndownService = new TurndownService({\r\n    headingStyle: 'atx',\r\n    codeBlockStyle: 'fenced',\r\n});\r\n\r\n// Configure turndown to keep some layout or handle semantic tags differently if needed\r\n\r\nexport interface ExtractionResult {\r\n    markdown: string;\r\n    metadata: {\r\n        title: string;\r\n        description: string;\r\n        jsonLd: any[];\r\n    };\r\n    stats: {\r\n        originalHtmlSize: number;\r\n        markdownSize: number;\r\n        tokenReductionRatio: number;\r\n    };\r\n}\r\n\r\n/**\r\n * Extracts pure semantic markdown and metadata from rendered Next.js HTML strings.\r\n * @param html The raw HTML string.\r\n * @param sourceUrl (Optional) the URL this was generated from, to attach as metadata.\r\n * @returns {ExtractionResult} The extracted payload.\r\n */\r\nexport function extractContent(html: string, sourceUrl: string = 'Generated Output'): ExtractionResult {\r\n    const originalSize = html.length;\r\n\r\n    const $ = cheerio.load(html);\r\n\r\n    // 1. Extract Metadata BEFORE removing structure\r\n    const title = $('title').text() || $('h1').first().text() || 'Untitled Page';\r\n    const description = $('meta[name=\"description\"]').attr('content') || 'No description found.';\r\n\r\n    const jsonLdScripts: any[] = [];\r\n    $('script[type=\"application/ld+json\"]').each((_, el) => {\r\n        try {\r\n            const raw = $(el).html() || '';\r\n            const parsed = JSON.parse(raw);\r\n            jsonLdScripts.push(parsed);\r\n        } catch {\r\n            // ignore bad json\r\n        }\r\n    });\r\n\r\n    // 2. Strip noise (React boilerplate, styles, unnecessary tags)\r\n    $('script, style, noscript, iframe, svg, nav, footer, meta, link, header').remove();\r\n\r\n    // Optionally remove typical Next.js hidden wrappers if they don't contain real content.\r\n    // Next.js uses <div id=\"__next\"> but we mostly just want semantic content.\r\n\r\n    // 3. Find the entry point for content\r\n    // Prefer <main> or <article> over <body>\r\n    let contentHtml = '';\r\n    if ($('main').length > 0) {\r\n        contentHtml = $('main').html() || '';\r\n    } else if ($('article').length > 0) {\r\n        contentHtml = $('article').html() || '';\r\n    } else {\r\n        contentHtml = $('body').html() || '';\r\n    }\r\n\r\n    // 4. Convert to Markdown\r\n    let markdown = turndownService.turndown(contentHtml);\r\n\r\n    // 5. Optionally inject Metadata header\r\n    const headerLines = [\r\n        `# ${title}`,\r\n        `> ${description}`,\r\n        ``,\r\n        `**Source:** ${sourceUrl}`,\r\n        `**Extracted:** ${new Date().toISOString()}`,\r\n        ``,\r\n        `---`,\r\n        ``\r\n    ];\r\n\r\n    let finalMarkdown = headerLines.join('\\n') + markdown;\r\n\r\n    // Add JSON-LD section if exists\r\n    if (jsonLdScripts.length > 0) {\r\n        finalMarkdown += '\\n\\n---\\n## Structured Data (JSON-LD)\\n```json\\n';\r\n        jsonLdScripts.forEach(j => {\r\n            finalMarkdown += JSON.stringify(j, null, 2) + '\\n';\r\n        });\r\n        finalMarkdown += '```\\n';\r\n    }\r\n\r\n    const markdownSize = finalMarkdown.length;\r\n    const tokenReductionRatio = originalSize > 0 ? ((originalSize - markdownSize) / originalSize) * 100 : 0;\r\n\r\n    return {\r\n        markdown: finalMarkdown,\r\n        metadata: {\r\n            title,\r\n            description,\r\n            jsonLd: jsonLdScripts\r\n        },\r\n        stats: {\r\n            originalHtmlSize: originalSize,\r\n            markdownSize,\r\n            tokenReductionRatio\r\n        }\r\n    };\r\n}\r\n\r\nexport async function generateStaticPayloads(nextAppDirDir: string, ontoPublicDir: string) {\r\n    const fs = await import('fs');\r\n    const path = await import('path');\r\n    const { glob } = await import('glob');\r\n\r\n    if (!fs.existsSync(nextAppDirDir)) {\r\n        return;\r\n    }\r\n\r\n    const files = await glob('**/*.html', { cwd: nextAppDirDir });\r\n    if (files.length === 0) return;\r\n\r\n    if (!fs.existsSync(ontoPublicDir)) {\r\n        fs.mkdirSync(ontoPublicDir, { recursive: true });\r\n    }\r\n\r\n    let totalFilesProcessed = 0;\r\n\r\n    for (const file of files) {\r\n        const inputPath = path.join(nextAppDirDir, file);\r\n        const outputPathRelative = file.replace(/\\.html$/, '.md');\r\n        const outputPath = path.join(ontoPublicDir, outputPathRelative);\r\n\r\n        try {\r\n            const htmlContent = fs.readFileSync(inputPath, 'utf8');\r\n\r\n            let routeName = file.replace(/\\.html$/, '');\r\n            if (routeName === 'index') routeName = '/';\r\n            else routeName = `/${routeName}`;\r\n\r\n            const result = extractContent(htmlContent, routeName);\r\n\r\n            const outputDir = path.dirname(outputPath);\r\n            if (!fs.existsSync(outputDir)) {\r\n                fs.mkdirSync(outputDir, { recursive: true });\r\n            }\r\n\r\n            fs.writeFileSync(outputPath, result.markdown, 'utf8');\r\n            totalFilesProcessed++;\r\n        } catch (e: any) {\r\n            console.error(`[Onto] Failed to process ${file}: ${e.message}`);\r\n        }\r\n    }\r\n    console.log(`[Onto] Successfully generated ${totalFilesProcessed} semantic markdown endpoints.`);\r\n}\r\n","/**\r\n * Configuration schema for onto.config.ts\r\n * Used to dynamically generate llms.txt and other AI discovery files\r\n */\r\n\r\nexport type PageType = 'scoring' | 'about' | 'default';\r\n\r\nexport interface OntoRoute {\r\n  /**\r\n   * The URL path (e.g., '/docs', '/api/reference')\r\n   */\r\n  path: string;\r\n  /**\r\n   * Description of what this route contains\r\n   */\r\n  description: string;\r\n  /**\r\n   * Optional: Page type for automatic JSON-LD schema injection\r\n   * - 'scoring': Injects Methodology schema with AIO scoring weights (40/35/25)\r\n   * - 'about': Injects Organization/AboutPage schema\r\n   * - 'default': No automatic schema injection\r\n   */\r\n  pageType?: PageType;\r\n}\r\n\r\nexport interface OntoConfig {\r\n  /**\r\n   * The name of your project or site (required)\r\n   * Used as the H1 heading in llms.txt\r\n   */\r\n  name: string;\r\n\r\n  /**\r\n   * A short summary of your project (required)\r\n   * Displayed as a blockquote in llms.txt\r\n   * Should contain key information necessary for understanding the rest of the file\r\n   */\r\n  summary: string;\r\n\r\n  /**\r\n   * The base URL of your site (e.g., 'https://example.com')\r\n   */\r\n  baseUrl: string;\r\n\r\n  /**\r\n   * Optional: Additional sections to include in llms.txt\r\n   * Each section can contain any markdown content\r\n   */\r\n  sections?: {\r\n    heading: string;\r\n    content: string;\r\n  }[];\r\n\r\n  /**\r\n   * Key routes that AI agents should know about\r\n   * These will be formatted as a markdown list in llms.txt\r\n   */\r\n  routes?: OntoRoute[];\r\n\r\n  /**\r\n   * Optional: Links to external resources (documentation, API references, etc.)\r\n   */\r\n  externalLinks?: {\r\n    title: string;\r\n    url: string;\r\n    description?: string;\r\n  }[];\r\n\r\n  /**\r\n   * Optional: Organization information for JSON-LD schemas\r\n   */\r\n  organization?: {\r\n    name: string;\r\n    description?: string;\r\n    url?: string;\r\n    logo?: string;\r\n    foundingDate?: string;\r\n  };\r\n}\r\n\r\n/**\r\n * Generate llms.txt content from OntoConfig\r\n * Follows the llms.txt specification:\r\n * - H1 with project name\r\n * - Blockquote with summary\r\n * - Additional markdown sections\r\n */\r\nexport function generateLlmsTxt(config: OntoConfig): string {\r\n  const lines: string[] = [];\r\n\r\n  // H1: Project name (required)\r\n  lines.push(`# ${config.name}`);\r\n  lines.push('');\r\n\r\n  // Blockquote: Summary (required)\r\n  lines.push(`> ${config.summary}`);\r\n  lines.push('');\r\n\r\n  // Key Routes section (if provided)\r\n  if (config.routes && config.routes.length > 0) {\r\n    lines.push('## Key Routes');\r\n    lines.push('');\r\n    for (const route of config.routes) {\r\n      const fullUrl = `${config.baseUrl}${route.path}`;\r\n      lines.push(`- [${route.path}](${fullUrl}): ${route.description}`);\r\n    }\r\n    lines.push('');\r\n  }\r\n\r\n  // External Links section (if provided)\r\n  if (config.externalLinks && config.externalLinks.length > 0) {\r\n    lines.push('## Resources');\r\n    lines.push('');\r\n    for (const link of config.externalLinks) {\r\n      if (link.description) {\r\n        lines.push(`- [${link.title}](${link.url}): ${link.description}`);\r\n      } else {\r\n        lines.push(`- [${link.title}](${link.url})`);\r\n      }\r\n    }\r\n    lines.push('');\r\n  }\r\n\r\n  // Custom sections (if provided)\r\n  if (config.sections && config.sections.length > 0) {\r\n    for (const section of config.sections) {\r\n      lines.push(`## ${section.heading}`);\r\n      lines.push('');\r\n      lines.push(section.content);\r\n      lines.push('');\r\n    }\r\n  }\r\n\r\n  return lines.join('\\n').trim() + '\\n';\r\n}\r\n","import { NextRequest, NextResponse } from 'next/server';\r\nimport { matchBot } from './bots';\r\nimport { generateLlmsTxt, OntoConfig } from './config';\r\n\r\nexport async function ontoMiddleware(request: any, config?: OntoConfig) {\r\n    const userAgent = request.headers.get('user-agent');\r\n    const url = request.nextUrl.clone();\r\n    const matched = matchBot(userAgent);\r\n\r\n    const accept = request.headers.get('accept') || '';\r\n\r\n    const isAiBot = !!matched;\r\n    const isMarkdownRequested = accept.includes('text/markdown');\r\n\r\n    // If traffic is identified as an AI Bot or markdown is requested\r\n    if (isAiBot || isMarkdownRequested) {\r\n\r\n        // Ignore internal next.js requests & static assets (but not llms.txt)\r\n        if (url.pathname.startsWith('/_next')) {\r\n            return NextResponse.next();\r\n        }\r\n\r\n        // --- llms.txt Auto-Discovery ---\r\n        // Serve the llms.txt manifest to AI agents\r\n        if (url.pathname === '/llms.txt') {\r\n            try {\r\n                if (config) {\r\n                    // Generate llms.txt dynamically from config\r\n                    const llmsTxtContent = generateLlmsTxt(config);\r\n                    const response = new NextResponse(llmsTxtContent, {\r\n                        headers: {\r\n                            'Content-Type': 'text/plain; charset=utf-8',\r\n                            'Cache-Control': 'public, max-age=3600, s-maxage=3600, stale-while-revalidate=86400',\r\n                        }\r\n                    });\r\n\r\n                    if (matched) {\r\n                        response.headers.set('X-Onto-Bot', `${matched.name} (${matched.company})`);\r\n                    }\r\n\r\n                    return response;\r\n                } else {\r\n                    // Fallback: try to serve static llms.txt from public folder\r\n                    url.pathname = '/llms.txt';\r\n                    const response = NextResponse.rewrite(url);\r\n                    response.headers.set('Content-Type', 'text/plain; charset=utf-8');\r\n                    response.headers.set('Cache-Control', 'public, max-age=3600, s-maxage=3600, stale-while-revalidate=86400');\r\n                    if (matched) {\r\n                        response.headers.set('X-Onto-Bot', `${matched.name} (${matched.company})`);\r\n                    }\r\n                    return response;\r\n                }\r\n            } catch (error) {\r\n                console.error('[Onto] Failed to generate llms.txt:', error);\r\n                // Fallback to static file on error\r\n                url.pathname = '/llms.txt';\r\n                const response = NextResponse.rewrite(url);\r\n                response.headers.set('Content-Type', 'text/plain; charset=utf-8');\r\n                if (matched) {\r\n                    response.headers.set('X-Onto-Bot', `${matched.name} (${matched.company})`);\r\n                }\r\n                return response;\r\n            }\r\n        }\r\n\r\n        // Skip other static assets\r\n        if (url.pathname.includes('.')) {\r\n            return NextResponse.next();\r\n        }\r\n\r\n        // Determine the corresponding payload path\r\n        let payloadPath = url.pathname;\r\n        if (payloadPath === '/' || payloadPath === '') {\r\n            payloadPath = '/index';\r\n        }\r\n\r\n        // Strip trailing slash if present\r\n        if (payloadPath.endsWith('/') && payloadPath !== '/') {\r\n            payloadPath = payloadPath.slice(0, -1);\r\n        }\r\n\r\n        // Common response headers for all bot responses\r\n        const botHeaders: Record<string, string> = {\r\n            'Content-Type': 'text/markdown; charset=utf-8',\r\n            'Cache-Control': 'public, max-age=3600, s-maxage=3600, stale-while-revalidate=86400',\r\n        };\r\n        if (matched) {\r\n            botHeaders['X-Onto-Bot'] = `${matched.name} (${matched.company})`;\r\n        }\r\n\r\n        // --- Onto Control Plane Integration (Premium) ---\r\n        const ONTO_API_KEY = process.env.ONTO_API_KEY;\r\n        const DASHBOARD_URL = process.env.ONTO_DASHBOARD_URL || 'https://app.buildonto.dev';\r\n\r\n        if (ONTO_API_KEY) {\r\n            // 1. Fire-and-forget tracking — includes structured bot info\r\n            fetch(`${DASHBOARD_URL}/api/track`, {\r\n                method: 'POST',\r\n                headers: {\r\n                    'x-onto-key': ONTO_API_KEY,\r\n                    'Content-Type': 'application/json'\r\n                },\r\n                body: JSON.stringify({\r\n                    route: url.pathname,\r\n                    userAgent: userAgent,\r\n                    bot: matched ? matched.name : null,\r\n                    company: matched ? matched.company : null,\r\n                })\r\n            }).catch(() => {});\r\n\r\n            // 2. Dynamic Context Injection\r\n            try {\r\n                const injectRes = await fetch(`${DASHBOARD_URL}/api/sdk/inject?route=${url.pathname}`, {\r\n                    headers: { 'x-onto-key': ONTO_API_KEY },\r\n                    signal: AbortSignal.timeout(1500)\r\n                });\r\n\r\n                if (injectRes.ok) {\r\n                    const { injection } = await injectRes.json();\r\n                    \r\n                    if (injection) {\r\n                        const localMdUrl = `${url.origin}/.onto${payloadPath}.md`;\r\n                        const mdRes = await fetch(localMdUrl);\r\n                        \r\n                        if (mdRes.ok) {\r\n                            const baseMarkdown = await mdRes.text();\r\n                            const finalMarkdown = `${baseMarkdown}\\n\\n---\\n\\n${injection}`;\r\n                            \r\n                            return new NextResponse(finalMarkdown, {\r\n                                headers: {\r\n                                    ...botHeaders,\r\n                                    'X-Onto-Injected': 'true'\r\n                                }\r\n                            });\r\n                        }\r\n                    }\r\n                }\r\n            } catch (err) {\r\n                console.error('[Onto] Injection failed', err);\r\n            }\r\n        }\r\n        // ------------------------------------------------\r\n\r\n        url.pathname = `/.onto${payloadPath}.md`;\r\n\r\n        // Rewrite implicitly serves the target URL transparently to the client.\r\n        const response = NextResponse.rewrite(url);\r\n        if (matched) {\r\n            response.headers.set('X-Onto-Bot', `${matched.name} (${matched.company})`);\r\n        }\r\n        return response;\r\n    }\r\n\r\n    return NextResponse.next();\r\n}\r\n\r\n// Re-export the bot registry for consumers who want to extend or inspect it\r\nexport { AI_BOT_USER_AGENTS, matchBot } from './bots';\r\nexport type { AiBot } from './bots';\r\n\r\n\r\n","/**\n * Comprehensive registry of AI bot user-agent strings.\n * The middleware uses this list to detect AI crawlers and serve optimized markdown.\n */\n\nexport interface AiBot {\n    /** The user-agent substring to match against */\n    name: string;\n    /** The company operating this bot */\n    company: string;\n}\n\n/**\n * Structured registry of all known AI bots, grouped by company.\n * Useful for analytics and the Control Plane dashboard.\n */\nexport const AI_BOTS: AiBot[] = [\n    // OpenAI\n    { name: 'GPTBot',            company: 'OpenAI' },\n    { name: 'ChatGPT-User',     company: 'OpenAI' },\n    { name: 'OAI-SearchBot',    company: 'OpenAI' },\n\n    // Google\n    { name: 'Googlebot',              company: 'Google' },\n    { name: 'Google-CloudVertexBot',   company: 'Google' },\n    { name: 'Google-Extended',         company: 'Google' },\n    { name: 'GoogleOther',             company: 'Google' },\n\n    // Anthropic\n    { name: 'ClaudeBot',       company: 'Anthropic' },\n    { name: 'Claude-User',    company: 'Anthropic' },\n    { name: 'anthropic-ai',   company: 'Anthropic' },\n\n    // Perplexity\n    { name: 'PerplexityBot',   company: 'Perplexity' },\n    { name: 'Perplexity-User', company: 'Perplexity' },\n\n    // Meta\n    { name: 'Meta-ExternalAgent',   company: 'Meta' },\n    { name: 'Meta-ExternalFetcher', company: 'Meta' },\n    { name: 'FacebookBot',         company: 'Meta' },\n\n    // Common Crawl (used by most smaller AI companies)\n    { name: 'CCBot', company: 'Common Crawl' },\n\n    // Other notable AI crawlers\n    { name: 'Bytespider',        company: 'ByteDance' },\n    { name: 'Applebot-Extended', company: 'Apple' },\n    { name: 'cohere-ai',        company: 'Cohere' },\n    { name: 'YouBot',           company: 'You.com' },\n];\n\n/**\n * Flat list of user-agent substrings for fast matching in the middleware.\n */\nexport const AI_BOT_USER_AGENTS: string[] = AI_BOTS.map(bot => bot.name);\n\n/**\n * Given a raw user-agent string, returns the matched AiBot entry or undefined.\n * Comparison is case-insensitive to handle inconsistent agent casing.\n */\nexport function matchBot(userAgent: string | null): AiBot | undefined {\n    if (!userAgent) return undefined;\n    const lowerUA = userAgent.toLowerCase();\n    return AI_BOTS.find(bot => \n        lowerUA.includes(bot.name.toLowerCase())\n    );\n}\n","'use client';\n\nimport { usePathname } from 'next/navigation';\n\n/**\n * OntoHead — Auto-Discovery component for AI agents.\n * \n * Injects `<link rel=\"alternate\">` tags into the page `<head>` so AI crawlers\n * can discover the optimized markdown endpoint for the current route.\n * \n * Usage in a Next.js App Router layout:\n * ```tsx\n * import { OntoHead } from '@ontosdk/next/components';\n * \n * export default function RootLayout({ children }) {\n *   return (\n *     <html>\n *       <head>\n *         <OntoHead />\n *       </head>\n *       <body>{children}</body>\n *     </html>\n *   );\n * }\n * ```\n */\nexport function OntoHead() {\n    const pathname = usePathname();\n\n    // Map current route to its .onto markdown payload path\n    let payloadPath = pathname;\n    if (payloadPath === '/' || payloadPath === '') {\n        payloadPath = '/index';\n    }\n    if (payloadPath.endsWith('/') && payloadPath !== '/') {\n        payloadPath = payloadPath.slice(0, -1);\n    }\n\n    const markdownHref = `/.onto${payloadPath}.md`;\n\n    return (\n        <>\n            {/* Per-page markdown alternate for AI agents */}\n            <link\n                rel=\"alternate\"\n                type=\"text/markdown\"\n                href={markdownHref}\n                title=\"AI-optimized Markdown version\"\n            />\n            {/* Site-wide llms.txt manifest */}\n            <link\n                rel=\"alternate\"\n                type=\"text/plain\"\n                href=\"/llms.txt\"\n                title=\"LLM-readable site manifest\"\n            />\n        </>\n    );\n}\n","'use client';\r\n\r\nimport { usePathname } from 'next/navigation';\r\nimport { ReactNode, useMemo } from 'react';\r\nimport type { OntoConfig, PageType } from './config';\r\nimport { generateSchemaForPageType, serializeSchema } from './schemas';\r\n\r\nexport interface OntoProviderProps {\r\n  /**\r\n   * The base URL of your site (e.g., 'https://example.com')\r\n   * Used to construct the full href for the AI discovery link tag.\r\n   */\r\n  baseUrl: string;\r\n  /**\r\n   * Child components to render\r\n   */\r\n  children: ReactNode;\r\n  /**\r\n   * Optional: Onto configuration for automatic JSON-LD schema injection\r\n   * If provided, the provider will automatically inject JSON-LD schemas\r\n   * based on the page type configuration\r\n   */\r\n  config?: OntoConfig;\r\n}\r\n\r\n/**\r\n * OntoProvider — Automatic AI Discovery Provider\r\n *\r\n * Wraps your application and automatically injects:\r\n * 1. `<link rel=\"alternate\">` tags for AI discovery\r\n * 2. JSON-LD structured data schemas based on page type\r\n *\r\n * With config, automatically generates JSON-LD schemas:\r\n * - 'scoring' pages get Methodology schema with AIO weights (40/35/25)\r\n * - 'about' pages get Organization/AboutPage schema\r\n *\r\n * Usage in a Next.js App Router layout:\r\n * ```tsx\r\n * import { OntoProvider } from '@ontosdk/next/provider';\r\n * import config from '../onto.config';\r\n *\r\n * export default function RootLayout({ children }) {\r\n *   return (\r\n *     <OntoProvider baseUrl=\"https://example.com\" config={config}>\r\n *       <html>\r\n *         <head />\r\n *         <body>{children}</body>\r\n *       </html>\r\n *     </OntoProvider>\r\n *   );\r\n * }\r\n * ```\r\n */\r\nexport function OntoProvider({ baseUrl, children, config }: OntoProviderProps) {\r\n  const pathname = usePathname();\r\n\r\n  // Construct the full URL with the current path and ?format=md query string\r\n  const cleanBaseUrl = baseUrl.endsWith('/') ? baseUrl.slice(0, -1) : baseUrl;\r\n  const markdownHref = `${cleanBaseUrl}${pathname}?format=md`;\r\n  const fullPageUrl = `${cleanBaseUrl}${pathname}`;\r\n\r\n  // Determine page type from config routes\r\n  const pageType: PageType = useMemo(() => {\r\n    if (!config?.routes) return 'default';\r\n\r\n    const matchingRoute = config.routes.find(route => route.path === pathname);\r\n    return matchingRoute?.pageType || 'default';\r\n  }, [config, pathname]);\r\n\r\n  // Generate JSON-LD schema based on page type\r\n  const jsonLdSchema = useMemo(() => {\r\n    if (!config || pageType === 'default') return null;\r\n\r\n    const schema = generateSchemaForPageType(pageType, config, fullPageUrl);\r\n    return serializeSchema(schema);\r\n  }, [config, pageType, fullPageUrl]);\r\n\r\n  return (\r\n    <>\r\n      <link\r\n        rel=\"alternate\"\r\n        type=\"text/markdown\"\r\n        href={markdownHref}\r\n        title=\"AI-optimized Markdown version\"\r\n      />\r\n      {jsonLdSchema && (\r\n        <script\r\n          type=\"application/ld+json\"\r\n          dangerouslySetInnerHTML={{ __html: jsonLdSchema }}\r\n        />\r\n      )}\r\n      {children}\r\n    </>\r\n  );\r\n}\r\n","/**\r\n * JSON-LD Schema generators for automatic structured data injection\r\n * Follows Schema.org standards for AI-friendly metadata\r\n */\r\n\r\nimport { OntoConfig } from './config';\r\n\r\n/**\r\n * Standard AIO (AI Optimization) scoring methodology\r\n * Based on the Onto scoring algorithm:\r\n * - Content Negotiation: 40% (30 points)\r\n * - React Tax / Token Efficiency: 35% (30 points)\r\n * - Structured Data: 25% (25 points)\r\n * - Semantic HTML: Bonus (15 points)\r\n */\r\nexport interface AIOMethodologySchema {\r\n  '@context': 'https://schema.org';\r\n  '@type': 'HowTo';\r\n  name: string;\r\n  description: string;\r\n  step: Array<{\r\n    '@type': 'HowToStep';\r\n    name: string;\r\n    text: string;\r\n    position: number;\r\n  }>;\r\n}\r\n\r\n/**\r\n * Generate AIO Scoring Methodology JSON-LD schema\r\n * This explains to AI agents how the scoring system works\r\n */\r\nexport function generateAIOMethodologySchema(\r\n  config: OntoConfig,\r\n  pageUrl: string\r\n): AIOMethodologySchema {\r\n  return {\r\n    '@context': 'https://schema.org',\r\n    '@type': 'HowTo',\r\n    name: 'AIO Score Calculation Methodology',\r\n    description: 'AI Optimization (AIO) Score measures how well a website is optimized for AI agents and LLM crawlers. Scored out of 100 points based on four key metrics.',\r\n    step: [\r\n      {\r\n        '@type': 'HowToStep',\r\n        name: 'Content Negotiation',\r\n        text: 'Check if the site responds to Accept: text/markdown header. Weight: 40%. Penalty: -30 points if missing. This ensures AI bots receive optimized content instead of heavy HTML.',\r\n        position: 1\r\n      },\r\n      {\r\n        '@type': 'HowToStep',\r\n        name: 'Token Efficiency (React Tax)',\r\n        text: 'Measure the ratio of visible text to total HTML size. Weight: 35%. Penalty: -30 points if HTML > 100KB but text < 1KB. Detects JavaScript-heavy sites that are difficult for AI to parse.',\r\n        position: 2\r\n      },\r\n      {\r\n        '@type': 'HowToStep',\r\n        name: 'Structured Data',\r\n        text: 'Verify presence of JSON-LD structured data (Schema.org). Weight: 25%. Penalty: -25 points if missing. Enables AI to confidently extract pricing, products, and entities.',\r\n        position: 3\r\n      },\r\n      {\r\n        '@type': 'HowToStep',\r\n        name: 'Semantic HTML',\r\n        text: 'Check for semantic tags like <main> and <article>. Bonus: +15 points if present. Helps AI agents separate navigation from core content.',\r\n        position: 4\r\n      }\r\n    ]\r\n  };\r\n}\r\n\r\n/**\r\n * Organization schema for About pages\r\n */\r\nexport interface OrganizationSchema {\r\n  '@context': 'https://schema.org';\r\n  '@type': 'Organization';\r\n  name: string;\r\n  url?: string;\r\n  description?: string;\r\n  logo?: string;\r\n  foundingDate?: string;\r\n}\r\n\r\n/**\r\n * Generate Organization JSON-LD schema for About pages\r\n */\r\nexport function generateOrganizationSchema(\r\n  config: OntoConfig,\r\n  pageUrl: string\r\n): OrganizationSchema | null {\r\n  if (!config.organization) {\r\n    return null;\r\n  }\r\n\r\n  const schema: OrganizationSchema = {\r\n    '@context': 'https://schema.org',\r\n    '@type': 'Organization',\r\n    name: config.organization.name\r\n  };\r\n\r\n  if (config.organization.url) {\r\n    schema.url = config.organization.url;\r\n  }\r\n\r\n  if (config.organization.description) {\r\n    schema.description = config.organization.description;\r\n  }\r\n\r\n  if (config.organization.logo) {\r\n    schema.logo = config.organization.logo;\r\n  }\r\n\r\n  if (config.organization.foundingDate) {\r\n    schema.foundingDate = config.organization.foundingDate;\r\n  }\r\n\r\n  return schema;\r\n}\r\n\r\n/**\r\n * AboutPage schema combining Organization and WebPage\r\n */\r\nexport interface AboutPageSchema {\r\n  '@context': 'https://schema.org';\r\n  '@type': 'AboutPage';\r\n  name: string;\r\n  url: string;\r\n  description?: string;\r\n  mainEntity?: OrganizationSchema;\r\n}\r\n\r\n/**\r\n * Generate AboutPage JSON-LD schema\r\n */\r\nexport function generateAboutPageSchema(\r\n  config: OntoConfig,\r\n  pageUrl: string\r\n): AboutPageSchema {\r\n  const orgSchema = generateOrganizationSchema(config, pageUrl);\r\n\r\n  const schema: AboutPageSchema = {\r\n    '@context': 'https://schema.org',\r\n    '@type': 'AboutPage',\r\n    name: `About ${config.name}`,\r\n    url: pageUrl\r\n  };\r\n\r\n  if (config.summary) {\r\n    schema.description = config.summary;\r\n  }\r\n\r\n  if (orgSchema) {\r\n    schema.mainEntity = orgSchema;\r\n  }\r\n\r\n  return schema;\r\n}\r\n\r\n/**\r\n * Determine which schema to generate based on page type\r\n */\r\nexport function generateSchemaForPageType(\r\n  pageType: 'scoring' | 'about' | 'default',\r\n  config: OntoConfig,\r\n  pageUrl: string\r\n): any | null {\r\n  switch (pageType) {\r\n    case 'scoring':\r\n      return generateAIOMethodologySchema(config, pageUrl);\r\n    case 'about':\r\n      return generateAboutPageSchema(config, pageUrl);\r\n    case 'default':\r\n    default:\r\n      return null;\r\n  }\r\n}\r\n\r\n/**\r\n * Serialize schema to JSON-LD script tag content\r\n */\r\nexport function serializeSchema(schema: any | null): string | null {\r\n  if (!schema) {\r\n    return null;\r\n  }\r\n  return JSON.stringify(schema, null, 2);\r\n}\r\n"],"mappings":"0jBAAA,IAAAA,EAAA,GAAAC,EAAAD,EAAA,cAAAE,EAAA,iBAAAC,EAAA,mBAAAC,EAAA,iCAAAC,EAAA,4BAAAC,EAAA,oBAAAC,EAAA,+BAAAC,EAAA,8BAAAC,EAAA,mBAAAC,EAAA,oBAAAC,IAAA,eAAAC,EAAAZ,GCAA,IAAAa,EAAyB,sBACzBC,EAA4B,uBAEtBC,EAAkB,IAAI,EAAAC,QAAgB,CACxC,aAAc,MACd,eAAgB,QACpB,CAAC,EAwBM,SAASC,EAAeC,EAAcC,EAAoB,mBAAsC,CACnG,IAAMC,EAAeF,EAAK,OAEpBG,EAAY,OAAKH,CAAI,EAGrBI,EAAQD,EAAE,OAAO,EAAE,KAAK,GAAKA,EAAE,IAAI,EAAE,MAAM,EAAE,KAAK,GAAK,gBACvDE,EAAcF,EAAE,0BAA0B,EAAE,KAAK,SAAS,GAAK,wBAE/DG,EAAuB,CAAC,EAC9BH,EAAE,oCAAoC,EAAE,KAAK,CAACI,EAAGC,IAAO,CACpD,GAAI,CACA,IAAMC,EAAMN,EAAEK,CAAE,EAAE,KAAK,GAAK,GACtBE,EAAS,KAAK,MAAMD,CAAG,EAC7BH,EAAc,KAAKI,CAAM,CAC7B,MAAQ,CAER,CACJ,CAAC,EAGDP,EAAE,uEAAuE,EAAE,OAAO,EAOlF,IAAIQ,EAAc,GACdR,EAAE,MAAM,EAAE,OAAS,EACnBQ,EAAcR,EAAE,MAAM,EAAE,KAAK,GAAK,GAC3BA,EAAE,SAAS,EAAE,OAAS,EAC7BQ,EAAcR,EAAE,SAAS,EAAE,KAAK,GAAK,GAErCQ,EAAcR,EAAE,MAAM,EAAE,KAAK,GAAK,GAItC,IAAIS,EAAWf,EAAgB,SAASc,CAAW,EAc/CE,EAXgB,CAChB,KAAKT,CAAK,GACV,KAAKC,CAAW,GAChB,GACA,eAAeJ,CAAS,GACxB,kBAAkB,IAAI,KAAK,EAAE,YAAY,CAAC,GAC1C,GACA,MACA,EACJ,EAEgC,KAAK;AAAA,CAAI,EAAIW,EAGzCN,EAAc,OAAS,IACvBO,GAAiB;AAAA;AAAA;AAAA;AAAA;AAAA,EACjBP,EAAc,QAAQQ,GAAK,CACvBD,GAAiB,KAAK,UAAUC,EAAG,KAAM,CAAC,EAAI;AAAA,CAClD,CAAC,EACDD,GAAiB,SAGrB,IAAME,EAAeF,EAAc,OAC7BG,EAAsBd,EAAe,GAAMA,EAAea,GAAgBb,EAAgB,IAAM,EAEtG,MAAO,CACH,SAAUW,EACV,SAAU,CACN,MAAAT,EACA,YAAAC,EACA,OAAQC,CACZ,EACA,MAAO,CACH,iBAAkBJ,EAClB,aAAAa,EACA,oBAAAC,CACJ,CACJ,CACJ,CCtBO,SAASC,EAAgBC,EAA4B,CAC1D,IAAMC,EAAkB,CAAC,EAWzB,GARAA,EAAM,KAAK,KAAKD,EAAO,IAAI,EAAE,EAC7BC,EAAM,KAAK,EAAE,EAGbA,EAAM,KAAK,KAAKD,EAAO,OAAO,EAAE,EAChCC,EAAM,KAAK,EAAE,EAGTD,EAAO,QAAUA,EAAO,OAAO,OAAS,EAAG,CAC7CC,EAAM,KAAK,eAAe,EAC1BA,EAAM,KAAK,EAAE,EACb,QAAWC,KAASF,EAAO,OAAQ,CACjC,IAAMG,EAAU,GAAGH,EAAO,OAAO,GAAGE,EAAM,IAAI,GAC9CD,EAAM,KAAK,MAAMC,EAAM,IAAI,KAAKC,CAAO,MAAMD,EAAM,WAAW,EAAE,CAClE,CACAD,EAAM,KAAK,EAAE,CACf,CAGA,GAAID,EAAO,eAAiBA,EAAO,cAAc,OAAS,EAAG,CAC3DC,EAAM,KAAK,cAAc,EACzBA,EAAM,KAAK,EAAE,EACb,QAAWG,KAAQJ,EAAO,cACpBI,EAAK,YACPH,EAAM,KAAK,MAAMG,EAAK,KAAK,KAAKA,EAAK,GAAG,MAAMA,EAAK,WAAW,EAAE,EAEhEH,EAAM,KAAK,MAAMG,EAAK,KAAK,KAAKA,EAAK,GAAG,GAAG,EAG/CH,EAAM,KAAK,EAAE,CACf,CAGA,GAAID,EAAO,UAAYA,EAAO,SAAS,OAAS,EAC9C,QAAWK,KAAWL,EAAO,SAC3BC,EAAM,KAAK,MAAMI,EAAQ,OAAO,EAAE,EAClCJ,EAAM,KAAK,EAAE,EACbA,EAAM,KAAKI,EAAQ,OAAO,EAC1BJ,EAAM,KAAK,EAAE,EAIjB,OAAOA,EAAM,KAAK;AAAA,CAAI,EAAE,KAAK,EAAI;AAAA,CACnC,CCtIA,IAAAK,EAA0C,uBCgBnC,IAAMC,EAAmB,CAE5B,CAAE,KAAM,SAAqB,QAAS,QAAS,EAC/C,CAAE,KAAM,eAAoB,QAAS,QAAS,EAC9C,CAAE,KAAM,gBAAoB,QAAS,QAAS,EAG9C,CAAE,KAAM,YAA0B,QAAS,QAAS,EACpD,CAAE,KAAM,wBAA2B,QAAS,QAAS,EACrD,CAAE,KAAM,kBAA2B,QAAS,QAAS,EACrD,CAAE,KAAM,cAA2B,QAAS,QAAS,EAGrD,CAAE,KAAM,YAAmB,QAAS,WAAY,EAChD,CAAE,KAAM,cAAkB,QAAS,WAAY,EAC/C,CAAE,KAAM,eAAkB,QAAS,WAAY,EAG/C,CAAE,KAAM,gBAAmB,QAAS,YAAa,EACjD,CAAE,KAAM,kBAAmB,QAAS,YAAa,EAGjD,CAAE,KAAM,qBAAwB,QAAS,MAAO,EAChD,CAAE,KAAM,uBAAwB,QAAS,MAAO,EAChD,CAAE,KAAM,cAAuB,QAAS,MAAO,EAG/C,CAAE,KAAM,QAAS,QAAS,cAAe,EAGzC,CAAE,KAAM,aAAqB,QAAS,WAAY,EAClD,CAAE,KAAM,oBAAqB,QAAS,OAAQ,EAC9C,CAAE,KAAM,YAAoB,QAAS,QAAS,EAC9C,CAAE,KAAM,SAAoB,QAAS,SAAU,CACnD,EAKaC,EAA+BD,EAAQ,IAAIE,GAAOA,EAAI,IAAI,EAMhE,SAASC,EAASC,EAA6C,CAClE,GAAI,CAACA,EAAW,OAChB,IAAMC,EAAUD,EAAU,YAAY,EACtC,OAAOJ,EAAQ,KAAKE,GAChBG,EAAQ,SAASH,EAAI,KAAK,YAAY,CAAC,CAC3C,CACJ,CD/DA,eAAsBI,EAAeC,EAAcC,EAAqB,CACpE,IAAMC,EAAYF,EAAQ,QAAQ,IAAI,YAAY,EAC5CG,EAAMH,EAAQ,QAAQ,MAAM,EAC5BI,EAAUC,EAASH,CAAS,EAE5BI,EAASN,EAAQ,QAAQ,IAAI,QAAQ,GAAK,GAE1CO,EAAU,CAAC,CAACH,EACZI,EAAsBF,EAAO,SAAS,eAAe,EAG3D,GAAIC,GAAWC,EAAqB,CAGhC,GAAIL,EAAI,SAAS,WAAW,QAAQ,EAChC,OAAO,eAAa,KAAK,EAK7B,GAAIA,EAAI,WAAa,YACjB,GAAI,CACA,GAAIF,EAAQ,CAER,IAAMQ,EAAiBC,EAAgBT,CAAM,EACvCU,EAAW,IAAI,eAAaF,EAAgB,CAC9C,QAAS,CACL,eAAgB,4BAChB,gBAAiB,mEACrB,CACJ,CAAC,EAED,OAAIL,GACAO,EAAS,QAAQ,IAAI,aAAc,GAAGP,EAAQ,IAAI,KAAKA,EAAQ,OAAO,GAAG,EAGtEO,CACX,KAAO,CAEHR,EAAI,SAAW,YACf,IAAMQ,EAAW,eAAa,QAAQR,CAAG,EACzC,OAAAQ,EAAS,QAAQ,IAAI,eAAgB,2BAA2B,EAChEA,EAAS,QAAQ,IAAI,gBAAiB,mEAAmE,EACrGP,GACAO,EAAS,QAAQ,IAAI,aAAc,GAAGP,EAAQ,IAAI,KAAKA,EAAQ,OAAO,GAAG,EAEtEO,CACX,CACJ,OAASC,EAAO,CACZ,QAAQ,MAAM,sCAAuCA,CAAK,EAE1DT,EAAI,SAAW,YACf,IAAMQ,EAAW,eAAa,QAAQR,CAAG,EACzC,OAAAQ,EAAS,QAAQ,IAAI,eAAgB,2BAA2B,EAC5DP,GACAO,EAAS,QAAQ,IAAI,aAAc,GAAGP,EAAQ,IAAI,KAAKA,EAAQ,OAAO,GAAG,EAEtEO,CACX,CAIJ,GAAIR,EAAI,SAAS,SAAS,GAAG,EACzB,OAAO,eAAa,KAAK,EAI7B,IAAIU,EAAcV,EAAI,UAClBU,IAAgB,KAAOA,IAAgB,MACvCA,EAAc,UAIdA,EAAY,SAAS,GAAG,GAAKA,IAAgB,MAC7CA,EAAcA,EAAY,MAAM,EAAG,EAAE,GAIzC,IAAMC,EAAqC,CACvC,eAAgB,+BAChB,gBAAiB,mEACrB,EACIV,IACAU,EAAW,YAAY,EAAI,GAAGV,EAAQ,IAAI,KAAKA,EAAQ,OAAO,KAIlE,IAAMW,EAAe,QAAQ,IAAI,aAC3BC,EAAgB,QAAQ,IAAI,oBAAsB,4BAExD,GAAID,EAAc,CAEd,MAAM,GAAGC,CAAa,aAAc,CAChC,OAAQ,OACR,QAAS,CACL,aAAcD,EACd,eAAgB,kBACpB,EACA,KAAM,KAAK,UAAU,CACjB,MAAOZ,EAAI,SACX,UAAWD,EACX,IAAKE,EAAUA,EAAQ,KAAO,KAC9B,QAASA,EAAUA,EAAQ,QAAU,IACzC,CAAC,CACL,CAAC,EAAE,MAAM,IAAM,CAAC,CAAC,EAGjB,GAAI,CACA,IAAMa,EAAY,MAAM,MAAM,GAAGD,CAAa,yBAAyBb,EAAI,QAAQ,GAAI,CACnF,QAAS,CAAE,aAAcY,CAAa,EACtC,OAAQ,YAAY,QAAQ,IAAI,CACpC,CAAC,EAED,GAAIE,EAAU,GAAI,CACd,GAAM,CAAE,UAAAC,CAAU,EAAI,MAAMD,EAAU,KAAK,EAE3C,GAAIC,EAAW,CACX,IAAMC,EAAa,GAAGhB,EAAI,MAAM,SAASU,CAAW,MAC9CO,EAAQ,MAAM,MAAMD,CAAU,EAEpC,GAAIC,EAAM,GAAI,CAEV,IAAMC,EAAgB,GADD,MAAMD,EAAM,KAAK,CACD;AAAA;AAAA;AAAA;AAAA,EAAcF,CAAS,GAE5D,OAAO,IAAI,eAAaG,EAAe,CACnC,QAAS,CACL,GAAGP,EACH,kBAAmB,MACvB,CACJ,CAAC,CACL,CACJ,CACJ,CACJ,OAASQ,EAAK,CACV,QAAQ,MAAM,0BAA2BA,CAAG,CAChD,CACJ,CAGAnB,EAAI,SAAW,SAASU,CAAW,MAGnC,IAAMF,EAAW,eAAa,QAAQR,CAAG,EACzC,OAAIC,GACAO,EAAS,QAAQ,IAAI,aAAc,GAAGP,EAAQ,IAAI,KAAKA,EAAQ,OAAO,GAAG,EAEtEO,CACX,CAEA,OAAO,eAAa,KAAK,CAC7B,CExJA,IAAAY,EAA4B,2BAuCpBC,EAAA,6BAfD,SAASC,GAAW,CAIvB,IAAIC,KAHa,eAAY,GAIzBA,IAAgB,KAAOA,IAAgB,MACvCA,EAAc,UAEdA,EAAY,SAAS,GAAG,GAAKA,IAAgB,MAC7CA,EAAcA,EAAY,MAAM,EAAG,EAAE,GAGzC,IAAMC,EAAe,SAASD,CAAW,MAEzC,SACI,oBAEI,oBAAC,QACG,IAAI,YACJ,KAAK,gBACL,KAAMC,EACN,MAAM,gCACV,KAEA,OAAC,QACG,IAAI,YACJ,KAAK,aACL,KAAK,YACL,MAAM,6BACV,GACJ,CAER,CCxDA,IAAAC,EAA4B,2BAC5BC,EAAmC,iBC6B5B,SAASC,EACdC,EACAC,EACsB,CACtB,MAAO,CACL,WAAY,qBACZ,QAAS,QACT,KAAM,oCACN,YAAa,2JACb,KAAM,CACJ,CACE,QAAS,YACT,KAAM,sBACN,KAAM,iLACN,SAAU,CACZ,EACA,CACE,QAAS,YACT,KAAM,+BACN,KAAM,4LACN,SAAU,CACZ,EACA,CACE,QAAS,YACT,KAAM,kBACN,KAAM,2KACN,SAAU,CACZ,EACA,CACE,QAAS,YACT,KAAM,gBACN,KAAM,0IACN,SAAU,CACZ,CACF,CACF,CACF,CAkBO,SAASC,EACdF,EACAC,EAC2B,CAC3B,GAAI,CAACD,EAAO,aACV,OAAO,KAGT,IAAMG,EAA6B,CACjC,WAAY,qBACZ,QAAS,eACT,KAAMH,EAAO,aAAa,IAC5B,EAEA,OAAIA,EAAO,aAAa,MACtBG,EAAO,IAAMH,EAAO,aAAa,KAG/BA,EAAO,aAAa,cACtBG,EAAO,YAAcH,EAAO,aAAa,aAGvCA,EAAO,aAAa,OACtBG,EAAO,KAAOH,EAAO,aAAa,MAGhCA,EAAO,aAAa,eACtBG,EAAO,aAAeH,EAAO,aAAa,cAGrCG,CACT,CAiBO,SAASC,EACdJ,EACAC,EACiB,CACjB,IAAMI,EAAYH,EAA2BF,EAAQC,CAAO,EAEtDE,EAA0B,CAC9B,WAAY,qBACZ,QAAS,YACT,KAAM,SAASH,EAAO,IAAI,GAC1B,IAAKC,CACP,EAEA,OAAID,EAAO,UACTG,EAAO,YAAcH,EAAO,SAG1BK,IACFF,EAAO,WAAaE,GAGfF,CACT,CAKO,SAASG,EACdC,EACAP,EACAC,EACY,CACZ,OAAQM,EAAU,CAChB,IAAK,UACH,OAAOR,EAA6BC,EAAQC,CAAO,EACrD,IAAK,QACH,OAAOG,EAAwBJ,EAAQC,CAAO,EAEhD,QACE,OAAO,IACX,CACF,CAKO,SAASO,EAAgBL,EAAmC,CACjE,OAAKA,EAGE,KAAK,UAAUA,EAAQ,KAAM,CAAC,EAF5B,IAGX,CD3GI,IAAAM,EAAA,6BAzBG,SAASC,EAAa,CAAE,QAAAC,EAAS,SAAAC,EAAU,OAAAC,CAAO,EAAsB,CAC7E,IAAMC,KAAW,eAAY,EAGvBC,EAAeJ,EAAQ,SAAS,GAAG,EAAIA,EAAQ,MAAM,EAAG,EAAE,EAAIA,EAC9DK,EAAe,GAAGD,CAAY,GAAGD,CAAQ,aACzCG,EAAc,GAAGF,CAAY,GAAGD,CAAQ,GAGxCI,KAAqB,WAAQ,IAC5BL,GAAQ,QAESA,EAAO,OAAO,KAAKM,GAASA,EAAM,OAASL,CAAQ,GACnD,UAAY,UACjC,CAACD,EAAQC,CAAQ,CAAC,EAGfM,KAAe,WAAQ,IAAM,CACjC,GAAI,CAACP,GAAUK,IAAa,UAAW,OAAO,KAE9C,IAAMG,EAASC,EAA0BJ,EAAUL,EAAQI,CAAW,EACtE,OAAOM,EAAgBF,CAAM,CAC/B,EAAG,CAACR,EAAQK,EAAUD,CAAW,CAAC,EAElC,SACE,oBACE,oBAAC,QACC,IAAI,YACJ,KAAK,gBACL,KAAMD,EACN,MAAM,gCACR,EACCI,MACC,OAAC,UACC,KAAK,sBACL,wBAAyB,CAAE,OAAQA,CAAa,EAClD,EAEDR,GACH,CAEJ","names":["index_exports","__export","OntoHead","OntoProvider","extractContent","generateAIOMethodologySchema","generateAboutPageSchema","generateLlmsTxt","generateOrganizationSchema","generateSchemaForPageType","ontoMiddleware","serializeSchema","__toCommonJS","cheerio","import_turndown","turndownService","TurndownService","extractContent","html","sourceUrl","originalSize","$","title","description","jsonLdScripts","_","el","raw","parsed","contentHtml","markdown","finalMarkdown","j","markdownSize","tokenReductionRatio","generateLlmsTxt","config","lines","route","fullUrl","link","section","import_server","AI_BOTS","AI_BOT_USER_AGENTS","bot","matchBot","userAgent","lowerUA","ontoMiddleware","request","config","userAgent","url","matched","matchBot","accept","isAiBot","isMarkdownRequested","llmsTxtContent","generateLlmsTxt","response","error","payloadPath","botHeaders","ONTO_API_KEY","DASHBOARD_URL","injectRes","injection","localMdUrl","mdRes","finalMarkdown","err","import_navigation","import_jsx_runtime","OntoHead","payloadPath","markdownHref","import_navigation","import_react","generateAIOMethodologySchema","config","pageUrl","generateOrganizationSchema","schema","generateAboutPageSchema","orgSchema","generateSchemaForPageType","pageType","serializeSchema","import_jsx_runtime","OntoProvider","baseUrl","children","config","pathname","cleanBaseUrl","markdownHref","fullPageUrl","pageType","route","jsonLdSchema","schema","generateSchemaForPageType","serializeSchema"]}

package/dist/index.mjs

@@ -1,11 +1,15 @@
-import*as p from"cheerio";import O from"turndown";var w=new O({headingStyle:"atx",codeBlockStyle:"fenced"});function z(t,n="Generated Output"){let e=t.length,o=p.load(t),s=o("title").text()||o("h1").first().text()||"Untitled Page",c=o('meta[name="description"]').attr("content")||"No description found.",r=[];o('script[type="application/ld+json"]').each((l,y)=>{try{let S=o(y).html()||"",x=JSON.parse(S);r.push(x)}catch{}}),o("script, style, noscript, iframe, svg, nav, footer, meta, link, header").remove();let a="";o("main").length>0?a=o("main").html()||"":o("article").length>0?a=o("article").html()||"":a=o("body").html()||"";let d=w.turndown(a),i=[`# ${s}`,`> ${c}`,"",`**Source:** ${n}`,`**Extracted:** ${new Date().toISOString()}`,"","---",""].join(`
-`)+d;r.length>0&&(i+=`
+import*as A from"cheerio";import R from"turndown";var M=new R({headingStyle:"atx",codeBlockStyle:"fenced"});function I(t,e="Generated Output"){let n=t.length,o=A.load(t),a=o("title").text()||o("h1").first().text()||"Untitled Page",u=o('meta[name="description"]').attr("content")||"No description found.",m=[];o('script[type="application/ld+json"]').each((r,c)=>{try{let y=o(c).html()||"",f=JSON.parse(y);m.push(f)}catch{}}),o("script, style, noscript, iframe, svg, nav, footer, meta, link, header").remove();let p="";o("main").length>0?p=o("main").html()||"":o("article").length>0?p=o("article").html()||"":p=o("body").html()||"";let i=M.turndown(p),s=[`# ${a}`,`> ${u}`,"",`**Source:** ${e}`,`**Extracted:** ${new Date().toISOString()}`,"","---",""].join(`
+`)+i;m.length>0&&(s+=`
 
 ---
 ## Structured Data (JSON-LD)
 \`\`\`json
-`,r.forEach(l=>{i+=JSON.stringify(l,null,2)+`
-`}),i+="```\n");let u=i.length,f=e>0?(e-u)/e*100:0;return{markdown:i,metadata:{title:s,description:c,jsonLd:r},stats:{originalHtmlSize:e,markdownSize:u,tokenReductionRatio:f}}}function k(t){let n=[];if(n.push(`# ${t.name}`),n.push(""),n.push(`> ${t.summary}`),n.push(""),t.routes&&t.routes.length>0){n.push("## Key Routes"),n.push("");for(let e of t.routes){let o=`${t.baseUrl}${e.path}`;n.push(`- [${e.path}](${o}): ${e.description}`)}n.push("")}if(t.externalLinks&&t.externalLinks.length>0){n.push("## Resources"),n.push("");for(let e of t.externalLinks)e.description?n.push(`- [${e.title}](${e.url}): ${e.description}`):n.push(`- [${e.title}](${e.url})`);n.push("")}if(t.sections&&t.sections.length>0)for(let e of t.sections)n.push(`## ${e.heading}`),n.push(""),n.push(e.content),n.push("");return n.join(`
+`,m.forEach(r=>{s+=JSON.stringify(r,null,2)+`
+`}),s+="```\n");let h=s.length,g=n>0?(n-h)/n*100:0;return{markdown:s,metadata:{title:a,description:u,jsonLd:m},stats:{originalHtmlSize:n,markdownSize:h,tokenReductionRatio:g}}}function x(t){let e=[];if(e.push(`# ${t.name}`),e.push(""),e.push(`> ${t.summary}`),e.push(""),t.routes&&t.routes.length>0){e.push("## Key Routes"),e.push("");for(let n of t.routes){let o=`${t.baseUrl}${n.path}`;e.push(`- [${n.path}](${o}): ${n.description}`)}e.push("")}if(t.externalLinks&&t.externalLinks.length>0){e.push("## Resources"),e.push("");for(let n of t.externalLinks)n.description?e.push(`- [${n.title}](${n.url}): ${n.description}`):e.push(`- [${n.title}](${n.url})`);e.push("")}if(t.sections&&t.sections.length>0)for(let n of t.sections)e.push(`## ${n.heading}`),e.push(""),e.push(n.content),e.push("");return e.join(`
 `).trim()+`
-`}function g(t,n){return{"@context":"https://schema.org","@type":"HowTo",name:"AIO Score Calculation Methodology",description:"AI Optimization (AIO) Score measures how well a website is optimized for AI agents and LLM crawlers. Scored out of 100 points based on four key metrics.",step:[{"@type":"HowToStep",name:"Content Negotiation",text:"Check if the site responds to Accept: text/markdown header. Weight: 40%. Penalty: -30 points if missing. This ensures AI bots receive optimized content instead of heavy HTML.",position:1},{"@type":"HowToStep",name:"Token Efficiency (React Tax)",text:"Measure the ratio of visible text to total HTML size. Weight: 35%. Penalty: -30 points if HTML > 100KB but text < 1KB. Detects JavaScript-heavy sites that are difficult for AI to parse.",position:2},{"@type":"HowToStep",name:"Structured Data",text:"Verify presence of JSON-LD structured data (Schema.org). Weight: 25%. Penalty: -25 points if missing. Enables AI to confidently extract pricing, products, and entities.",position:3},{"@type":"HowToStep",name:"Semantic HTML",text:"Check for semantic tags like <main> and <article>. Bonus: +15 points if present. Helps AI agents separate navigation from core content.",position:4}]}}function m(t,n){if(!t.organization)return null;let e={"@context":"https://schema.org","@type":"Organization",name:t.organization.name};return t.organization.url&&(e.url=t.organization.url),t.organization.description&&(e.description=t.organization.description),t.organization.logo&&(e.logo=t.organization.logo),t.organization.foundingDate&&(e.foundingDate=t.organization.foundingDate),e}function h(t,n){let e=m(t,n),o={"@context":"https://schema.org","@type":"AboutPage",name:`About ${t.name}`,url:n};return t.summary&&(o.description=t.summary),e&&(o.mainEntity=e),o}function T(t,n,e){switch(t){case"scoring":return g(n,e);case"about":return h(n,e);default:return null}}function $(t){return t?JSON.stringify(t,null,2):null}export{z as extractContent,g as generateAIOMethodologySchema,h as generateAboutPageSchema,k as generateLlmsTxt,m as generateOrganizationSchema,T as generateSchemaForPageType,$ as serializeSchema};
+`}import{NextResponse as l}from"next/server";var T=[{name:"GPTBot",company:"OpenAI"},{name:"ChatGPT-User",company:"OpenAI"},{name:"OAI-SearchBot",company:"OpenAI"},{name:"Googlebot",company:"Google"},{name:"Google-CloudVertexBot",company:"Google"},{name:"Google-Extended",company:"Google"},{name:"GoogleOther",company:"Google"},{name:"ClaudeBot",company:"Anthropic"},{name:"Claude-User",company:"Anthropic"},{name:"anthropic-ai",company:"Anthropic"},{name:"PerplexityBot",company:"Perplexity"},{name:"Perplexity-User",company:"Perplexity"},{name:"Meta-ExternalAgent",company:"Meta"},{name:"Meta-ExternalFetcher",company:"Meta"},{name:"FacebookBot",company:"Meta"},{name:"CCBot",company:"Common Crawl"},{name:"Bytespider",company:"ByteDance"},{name:"Applebot-Extended",company:"Apple"},{name:"cohere-ai",company:"Cohere"},{name:"YouBot",company:"You.com"}],L=T.map(t=>t.name);function S(t){if(!t)return;let e=t.toLowerCase();return T.find(n=>e.includes(n.name.toLowerCase()))}async function v(t,e){let n=t.headers.get("user-agent"),o=t.nextUrl.clone(),a=S(n),u=t.headers.get("accept")||"",m=!!a,p=u.includes("text/markdown");if(m||p){if(o.pathname.startsWith("/_next"))return l.next();if(o.pathname==="/llms.txt")try{if(e){let r=x(e),c=new l(r,{headers:{"Content-Type":"text/plain; charset=utf-8","Cache-Control":"public, max-age=3600, s-maxage=3600, stale-while-revalidate=86400"}});return a&&c.headers.set("X-Onto-Bot",`${a.name} (${a.company})`),c}else{o.pathname="/llms.txt";let r=l.rewrite(o);return r.headers.set("Content-Type","text/plain; charset=utf-8"),r.headers.set("Cache-Control","public, max-age=3600, s-maxage=3600, stale-while-revalidate=86400"),a&&r.headers.set("X-Onto-Bot",`${a.name} (${a.company})`),r}}catch(r){console.error("[Onto] Failed to generate llms.txt:",r),o.pathname="/llms.txt";let c=l.rewrite(o);return c.headers.set("Content-Type","text/plain; charset=utf-8"),a&&c.headers.set("X-Onto-Bot",`${a.name} (${a.company})`),c}if(o.pathname.includes("."))return l.next();let i=o.pathname;(i==="/"||i==="")&&(i="/index"),i.endsWith("/")&&i!=="/"&&(i=i.slice(0,-1));let d={"Content-Type":"text/markdown; charset=utf-8","Cache-Control":"public, max-age=3600, s-maxage=3600, stale-while-revalidate=86400"};a&&(d["X-Onto-Bot"]=`${a.name} (${a.company})`);let s=process.env.ONTO_API_KEY,h=process.env.ONTO_DASHBOARD_URL||"https://app.buildonto.dev";if(s){fetch(`${h}/api/track`,{method:"POST",headers:{"x-onto-key":s,"Content-Type":"application/json"},body:JSON.stringify({route:o.pathname,userAgent:n,bot:a?a.name:null,company:a?a.company:null})}).catch(()=>{});try{let r=await fetch(`${h}/api/sdk/inject?route=${o.pathname}`,{headers:{"x-onto-key":s},signal:AbortSignal.timeout(1500)});if(r.ok){let{injection:c}=await r.json();if(c){let y=`${o.origin}/.onto${i}.md`,f=await fetch(y);if(f.ok){let B=`${await f.text()}
+
+---
+
+${c}`;return new l(B,{headers:{...d,"X-Onto-Injected":"true"}})}}}}catch(r){console.error("[Onto] Injection failed",r)}}o.pathname=`/.onto${i}.md`;let g=l.rewrite(o);return a&&g.headers.set("X-Onto-Bot",`${a.name} (${a.company})`),g}return l.next()}import{usePathname as H}from"next/navigation";import{Fragment as N,jsx as $,jsxs as _}from"react/jsx-runtime";function E(){let e=H();(e==="/"||e==="")&&(e="/index"),e.endsWith("/")&&e!=="/"&&(e=e.slice(0,-1));let n=`/.onto${e}.md`;return _(N,{children:[$("link",{rel:"alternate",type:"text/markdown",href:n,title:"AI-optimized Markdown version"}),$("link",{rel:"alternate",type:"text/plain",href:"/llms.txt",title:"LLM-readable site manifest"})]})}import{usePathname as j}from"next/navigation";import{useMemo as b}from"react";function k(t,e){return{"@context":"https://schema.org","@type":"HowTo",name:"AIO Score Calculation Methodology",description:"AI Optimization (AIO) Score measures how well a website is optimized for AI agents and LLM crawlers. Scored out of 100 points based on four key metrics.",step:[{"@type":"HowToStep",name:"Content Negotiation",text:"Check if the site responds to Accept: text/markdown header. Weight: 40%. Penalty: -30 points if missing. This ensures AI bots receive optimized content instead of heavy HTML.",position:1},{"@type":"HowToStep",name:"Token Efficiency (React Tax)",text:"Measure the ratio of visible text to total HTML size. Weight: 35%. Penalty: -30 points if HTML > 100KB but text < 1KB. Detects JavaScript-heavy sites that are difficult for AI to parse.",position:2},{"@type":"HowToStep",name:"Structured Data",text:"Verify presence of JSON-LD structured data (Schema.org). Weight: 25%. Penalty: -25 points if missing. Enables AI to confidently extract pricing, products, and entities.",position:3},{"@type":"HowToStep",name:"Semantic HTML",text:"Check for semantic tags like <main> and <article>. Bonus: +15 points if present. Helps AI agents separate navigation from core content.",position:4}]}}function C(t,e){if(!t.organization)return null;let n={"@context":"https://schema.org","@type":"Organization",name:t.organization.name};return t.organization.url&&(n.url=t.organization.url),t.organization.description&&(n.description=t.organization.description),t.organization.logo&&(n.logo=t.organization.logo),t.organization.foundingDate&&(n.foundingDate=t.organization.foundingDate),n}function P(t,e){let n=C(t,e),o={"@context":"https://schema.org","@type":"AboutPage",name:`About ${t.name}`,url:e};return t.summary&&(o.description=t.summary),n&&(o.mainEntity=n),o}function O(t,e,n){switch(t){case"scoring":return k(e,n);case"about":return P(e,n);default:return null}}function w(t){return t?JSON.stringify(t,null,2):null}import{Fragment as G,jsx as z,jsxs as D}from"react/jsx-runtime";function U({baseUrl:t,children:e,config:n}){let o=j(),a=t.endsWith("/")?t.slice(0,-1):t,u=`${a}${o}?format=md`,m=`${a}${o}`,p=b(()=>n?.routes&&n.routes.find(s=>s.path===o)?.pageType||"default",[n,o]),i=b(()=>{if(!n||p==="default")return null;let d=O(p,n,m);return w(d)},[n,p,m]);return D(G,{children:[z("link",{rel:"alternate",type:"text/markdown",href:u,title:"AI-optimized Markdown version"}),i&&z("script",{type:"application/ld+json",dangerouslySetInnerHTML:{__html:i}}),e]})}export{E as OntoHead,U as OntoProvider,I as extractContent,k as generateAIOMethodologySchema,P as generateAboutPageSchema,x as generateLlmsTxt,C as generateOrganizationSchema,O as generateSchemaForPageType,v as ontoMiddleware,w as serializeSchema};
 //# sourceMappingURL=index.mjs.map

package/dist/index.mjs.map

@@ -1 +1 @@
-{"version":3,"sources":["../src/extractor.ts","../src/config.ts","../src/schemas.ts"],"sourcesContent":["import * as cheerio from 'cheerio';\r\nimport TurndownService from 'turndown';\r\n\r\nconst turndownService = new TurndownService({\r\n    headingStyle: 'atx',\r\n    codeBlockStyle: 'fenced',\r\n});\r\n\r\n// Configure turndown to keep some layout or handle semantic tags differently if needed\r\n\r\nexport interface ExtractionResult {\r\n    markdown: string;\r\n    metadata: {\r\n        title: string;\r\n        description: string;\r\n        jsonLd: any[];\r\n    };\r\n    stats: {\r\n        originalHtmlSize: number;\r\n        markdownSize: number;\r\n        tokenReductionRatio: number;\r\n    };\r\n}\r\n\r\n/**\r\n * Extracts pure semantic markdown and metadata from rendered Next.js HTML strings.\r\n * @param html The raw HTML string.\r\n * @param sourceUrl (Optional) the URL this was generated from, to attach as metadata.\r\n * @returns {ExtractionResult} The extracted payload.\r\n */\r\nexport function extractContent(html: string, sourceUrl: string = 'Generated Output'): ExtractionResult {\r\n    const originalSize = html.length;\r\n\r\n    const $ = cheerio.load(html);\r\n\r\n    // 1. Extract Metadata BEFORE removing structure\r\n    const title = $('title').text() || $('h1').first().text() || 'Untitled Page';\r\n    const description = $('meta[name=\"description\"]').attr('content') || 'No description found.';\r\n\r\n    const jsonLdScripts: any[] = [];\r\n    $('script[type=\"application/ld+json\"]').each((_, el) => {\r\n        try {\r\n            const raw = $(el).html() || '';\r\n            const parsed = JSON.parse(raw);\r\n            jsonLdScripts.push(parsed);\r\n        } catch {\r\n            // ignore bad json\r\n        }\r\n    });\r\n\r\n    // 2. Strip noise (React boilerplate, styles, unnecessary tags)\r\n    $('script, style, noscript, iframe, svg, nav, footer, meta, link, header').remove();\r\n\r\n    // Optionally remove typical Next.js hidden wrappers if they don't contain real content.\r\n    // Next.js uses <div id=\"__next\"> but we mostly just want semantic content.\r\n\r\n    // 3. Find the entry point for content\r\n    // Prefer <main> or <article> over <body>\r\n    let contentHtml = '';\r\n    if ($('main').length > 0) {\r\n        contentHtml = $('main').html() || '';\r\n    } else if ($('article').length > 0) {\r\n        contentHtml = $('article').html() || '';\r\n    } else {\r\n        contentHtml = $('body').html() || '';\r\n    }\r\n\r\n    // 4. Convert to Markdown\r\n    let markdown = turndownService.turndown(contentHtml);\r\n\r\n    // 5. Optionally inject Metadata header\r\n    const headerLines = [\r\n        `# ${title}`,\r\n        `> ${description}`,\r\n        ``,\r\n        `**Source:** ${sourceUrl}`,\r\n        `**Extracted:** ${new Date().toISOString()}`,\r\n        ``,\r\n        `---`,\r\n        ``\r\n    ];\r\n\r\n    let finalMarkdown = headerLines.join('\\n') + markdown;\r\n\r\n    // Add JSON-LD section if exists\r\n    if (jsonLdScripts.length > 0) {\r\n        finalMarkdown += '\\n\\n---\\n## Structured Data (JSON-LD)\\n```json\\n';\r\n        jsonLdScripts.forEach(j => {\r\n            finalMarkdown += JSON.stringify(j, null, 2) + '\\n';\r\n        });\r\n        finalMarkdown += '```\\n';\r\n    }\r\n\r\n    const markdownSize = finalMarkdown.length;\r\n    const tokenReductionRatio = originalSize > 0 ? ((originalSize - markdownSize) / originalSize) * 100 : 0;\r\n\r\n    return {\r\n        markdown: finalMarkdown,\r\n        metadata: {\r\n            title,\r\n            description,\r\n            jsonLd: jsonLdScripts\r\n        },\r\n        stats: {\r\n            originalHtmlSize: originalSize,\r\n            markdownSize,\r\n            tokenReductionRatio\r\n        }\r\n    };\r\n}\r\n\r\nexport async function generateStaticPayloads(nextAppDirDir: string, ontoPublicDir: string) {\r\n    const fs = await import('fs');\r\n    const path = await import('path');\r\n    const { glob } = await import('glob');\r\n\r\n    if (!fs.existsSync(nextAppDirDir)) {\r\n        return;\r\n    }\r\n\r\n    const files = await glob('**/*.html', { cwd: nextAppDirDir });\r\n    if (files.length === 0) return;\r\n\r\n    if (!fs.existsSync(ontoPublicDir)) {\r\n        fs.mkdirSync(ontoPublicDir, { recursive: true });\r\n    }\r\n\r\n    let totalFilesProcessed = 0;\r\n\r\n    for (const file of files) {\r\n        const inputPath = path.join(nextAppDirDir, file);\r\n        const outputPathRelative = file.replace(/\\.html$/, '.md');\r\n        const outputPath = path.join(ontoPublicDir, outputPathRelative);\r\n\r\n        try {\r\n            const htmlContent = fs.readFileSync(inputPath, 'utf8');\r\n\r\n            let routeName = file.replace(/\\.html$/, '');\r\n            if (routeName === 'index') routeName = '/';\r\n            else routeName = `/${routeName}`;\r\n\r\n            const result = extractContent(htmlContent, routeName);\r\n\r\n            const outputDir = path.dirname(outputPath);\r\n            if (!fs.existsSync(outputDir)) {\r\n                fs.mkdirSync(outputDir, { recursive: true });\r\n            }\r\n\r\n            fs.writeFileSync(outputPath, result.markdown, 'utf8');\r\n            totalFilesProcessed++;\r\n        } catch (e: any) {\r\n            console.error(`[Onto] Failed to process ${file}: ${e.message}`);\r\n        }\r\n    }\r\n    console.log(`[Onto] Successfully generated ${totalFilesProcessed} semantic markdown endpoints.`);\r\n}\r\n","/**\r\n * Configuration schema for onto.config.ts\r\n * Used to dynamically generate llms.txt and other AI discovery files\r\n */\r\n\r\nexport type PageType = 'scoring' | 'about' | 'default';\r\n\r\nexport interface OntoRoute {\r\n  /**\r\n   * The URL path (e.g., '/docs', '/api/reference')\r\n   */\r\n  path: string;\r\n  /**\r\n   * Description of what this route contains\r\n   */\r\n  description: string;\r\n  /**\r\n   * Optional: Page type for automatic JSON-LD schema injection\r\n   * - 'scoring': Injects Methodology schema with AIO scoring weights (40/35/25)\r\n   * - 'about': Injects Organization/AboutPage schema\r\n   * - 'default': No automatic schema injection\r\n   */\r\n  pageType?: PageType;\r\n}\r\n\r\nexport interface OntoConfig {\r\n  /**\r\n   * The name of your project or site (required)\r\n   * Used as the H1 heading in llms.txt\r\n   */\r\n  name: string;\r\n\r\n  /**\r\n   * A short summary of your project (required)\r\n   * Displayed as a blockquote in llms.txt\r\n   * Should contain key information necessary for understanding the rest of the file\r\n   */\r\n  summary: string;\r\n\r\n  /**\r\n   * The base URL of your site (e.g., 'https://example.com')\r\n   */\r\n  baseUrl: string;\r\n\r\n  /**\r\n   * Optional: Additional sections to include in llms.txt\r\n   * Each section can contain any markdown content\r\n   */\r\n  sections?: {\r\n    heading: string;\r\n    content: string;\r\n  }[];\r\n\r\n  /**\r\n   * Key routes that AI agents should know about\r\n   * These will be formatted as a markdown list in llms.txt\r\n   */\r\n  routes?: OntoRoute[];\r\n\r\n  /**\r\n   * Optional: Links to external resources (documentation, API references, etc.)\r\n   */\r\n  externalLinks?: {\r\n    title: string;\r\n    url: string;\r\n    description?: string;\r\n  }[];\r\n\r\n  /**\r\n   * Optional: Organization information for JSON-LD schemas\r\n   */\r\n  organization?: {\r\n    name: string;\r\n    description?: string;\r\n    url?: string;\r\n    logo?: string;\r\n    foundingDate?: string;\r\n  };\r\n}\r\n\r\n/**\r\n * Generate llms.txt content from OntoConfig\r\n * Follows the llms.txt specification:\r\n * - H1 with project name\r\n * - Blockquote with summary\r\n * - Additional markdown sections\r\n */\r\nexport function generateLlmsTxt(config: OntoConfig): string {\r\n  const lines: string[] = [];\r\n\r\n  // H1: Project name (required)\r\n  lines.push(`# ${config.name}`);\r\n  lines.push('');\r\n\r\n  // Blockquote: Summary (required)\r\n  lines.push(`> ${config.summary}`);\r\n  lines.push('');\r\n\r\n  // Key Routes section (if provided)\r\n  if (config.routes && config.routes.length > 0) {\r\n    lines.push('## Key Routes');\r\n    lines.push('');\r\n    for (const route of config.routes) {\r\n      const fullUrl = `${config.baseUrl}${route.path}`;\r\n      lines.push(`- [${route.path}](${fullUrl}): ${route.description}`);\r\n    }\r\n    lines.push('');\r\n  }\r\n\r\n  // External Links section (if provided)\r\n  if (config.externalLinks && config.externalLinks.length > 0) {\r\n    lines.push('## Resources');\r\n    lines.push('');\r\n    for (const link of config.externalLinks) {\r\n      if (link.description) {\r\n        lines.push(`- [${link.title}](${link.url}): ${link.description}`);\r\n      } else {\r\n        lines.push(`- [${link.title}](${link.url})`);\r\n      }\r\n    }\r\n    lines.push('');\r\n  }\r\n\r\n  // Custom sections (if provided)\r\n  if (config.sections && config.sections.length > 0) {\r\n    for (const section of config.sections) {\r\n      lines.push(`## ${section.heading}`);\r\n      lines.push('');\r\n      lines.push(section.content);\r\n      lines.push('');\r\n    }\r\n  }\r\n\r\n  return lines.join('\\n').trim() + '\\n';\r\n}\r\n","/**\r\n * JSON-LD Schema generators for automatic structured data injection\r\n * Follows Schema.org standards for AI-friendly metadata\r\n */\r\n\r\nimport { OntoConfig } from './config';\r\n\r\n/**\r\n * Standard AIO (AI Optimization) scoring methodology\r\n * Based on the Onto scoring algorithm:\r\n * - Content Negotiation: 40% (30 points)\r\n * - React Tax / Token Efficiency: 35% (30 points)\r\n * - Structured Data: 25% (25 points)\r\n * - Semantic HTML: Bonus (15 points)\r\n */\r\nexport interface AIOMethodologySchema {\r\n  '@context': 'https://schema.org';\r\n  '@type': 'HowTo';\r\n  name: string;\r\n  description: string;\r\n  step: Array<{\r\n    '@type': 'HowToStep';\r\n    name: string;\r\n    text: string;\r\n    position: number;\r\n  }>;\r\n}\r\n\r\n/**\r\n * Generate AIO Scoring Methodology JSON-LD schema\r\n * This explains to AI agents how the scoring system works\r\n */\r\nexport function generateAIOMethodologySchema(\r\n  config: OntoConfig,\r\n  pageUrl: string\r\n): AIOMethodologySchema {\r\n  return {\r\n    '@context': 'https://schema.org',\r\n    '@type': 'HowTo',\r\n    name: 'AIO Score Calculation Methodology',\r\n    description: 'AI Optimization (AIO) Score measures how well a website is optimized for AI agents and LLM crawlers. Scored out of 100 points based on four key metrics.',\r\n    step: [\r\n      {\r\n        '@type': 'HowToStep',\r\n        name: 'Content Negotiation',\r\n        text: 'Check if the site responds to Accept: text/markdown header. Weight: 40%. Penalty: -30 points if missing. This ensures AI bots receive optimized content instead of heavy HTML.',\r\n        position: 1\r\n      },\r\n      {\r\n        '@type': 'HowToStep',\r\n        name: 'Token Efficiency (React Tax)',\r\n        text: 'Measure the ratio of visible text to total HTML size. Weight: 35%. Penalty: -30 points if HTML > 100KB but text < 1KB. Detects JavaScript-heavy sites that are difficult for AI to parse.',\r\n        position: 2\r\n      },\r\n      {\r\n        '@type': 'HowToStep',\r\n        name: 'Structured Data',\r\n        text: 'Verify presence of JSON-LD structured data (Schema.org). Weight: 25%. Penalty: -25 points if missing. Enables AI to confidently extract pricing, products, and entities.',\r\n        position: 3\r\n      },\r\n      {\r\n        '@type': 'HowToStep',\r\n        name: 'Semantic HTML',\r\n        text: 'Check for semantic tags like <main> and <article>. Bonus: +15 points if present. Helps AI agents separate navigation from core content.',\r\n        position: 4\r\n      }\r\n    ]\r\n  };\r\n}\r\n\r\n/**\r\n * Organization schema for About pages\r\n */\r\nexport interface OrganizationSchema {\r\n  '@context': 'https://schema.org';\r\n  '@type': 'Organization';\r\n  name: string;\r\n  url?: string;\r\n  description?: string;\r\n  logo?: string;\r\n  foundingDate?: string;\r\n}\r\n\r\n/**\r\n * Generate Organization JSON-LD schema for About pages\r\n */\r\nexport function generateOrganizationSchema(\r\n  config: OntoConfig,\r\n  pageUrl: string\r\n): OrganizationSchema | null {\r\n  if (!config.organization) {\r\n    return null;\r\n  }\r\n\r\n  const schema: OrganizationSchema = {\r\n    '@context': 'https://schema.org',\r\n    '@type': 'Organization',\r\n    name: config.organization.name\r\n  };\r\n\r\n  if (config.organization.url) {\r\n    schema.url = config.organization.url;\r\n  }\r\n\r\n  if (config.organization.description) {\r\n    schema.description = config.organization.description;\r\n  }\r\n\r\n  if (config.organization.logo) {\r\n    schema.logo = config.organization.logo;\r\n  }\r\n\r\n  if (config.organization.foundingDate) {\r\n    schema.foundingDate = config.organization.foundingDate;\r\n  }\r\n\r\n  return schema;\r\n}\r\n\r\n/**\r\n * AboutPage schema combining Organization and WebPage\r\n */\r\nexport interface AboutPageSchema {\r\n  '@context': 'https://schema.org';\r\n  '@type': 'AboutPage';\r\n  name: string;\r\n  url: string;\r\n  description?: string;\r\n  mainEntity?: OrganizationSchema;\r\n}\r\n\r\n/**\r\n * Generate AboutPage JSON-LD schema\r\n */\r\nexport function generateAboutPageSchema(\r\n  config: OntoConfig,\r\n  pageUrl: string\r\n): AboutPageSchema {\r\n  const orgSchema = generateOrganizationSchema(config, pageUrl);\r\n\r\n  const schema: AboutPageSchema = {\r\n    '@context': 'https://schema.org',\r\n    '@type': 'AboutPage',\r\n    name: `About ${config.name}`,\r\n    url: pageUrl\r\n  };\r\n\r\n  if (config.summary) {\r\n    schema.description = config.summary;\r\n  }\r\n\r\n  if (orgSchema) {\r\n    schema.mainEntity = orgSchema;\r\n  }\r\n\r\n  return schema;\r\n}\r\n\r\n/**\r\n * Determine which schema to generate based on page type\r\n */\r\nexport function generateSchemaForPageType(\r\n  pageType: 'scoring' | 'about' | 'default',\r\n  config: OntoConfig,\r\n  pageUrl: string\r\n): any | null {\r\n  switch (pageType) {\r\n    case 'scoring':\r\n      return generateAIOMethodologySchema(config, pageUrl);\r\n    case 'about':\r\n      return generateAboutPageSchema(config, pageUrl);\r\n    case 'default':\r\n    default:\r\n      return null;\r\n  }\r\n}\r\n\r\n/**\r\n * Serialize schema to JSON-LD script tag content\r\n */\r\nexport function serializeSchema(schema: any | null): string | null {\r\n  if (!schema) {\r\n    return null;\r\n  }\r\n  return JSON.stringify(schema, null, 2);\r\n}\r\n"],"mappings":"AAAA,UAAYA,MAAa,UACzB,OAAOC,MAAqB,WAE5B,IAAMC,EAAkB,IAAID,EAAgB,CACxC,aAAc,MACd,eAAgB,QACpB,CAAC,EAwBM,SAASE,EAAeC,EAAcC,EAAoB,mBAAsC,CACnG,IAAMC,EAAeF,EAAK,OAEpBG,EAAY,OAAKH,CAAI,EAGrBI,EAAQD,EAAE,OAAO,EAAE,KAAK,GAAKA,EAAE,IAAI,EAAE,MAAM,EAAE,KAAK,GAAK,gBACvDE,EAAcF,EAAE,0BAA0B,EAAE,KAAK,SAAS,GAAK,wBAE/DG,EAAuB,CAAC,EAC9BH,EAAE,oCAAoC,EAAE,KAAK,CAACI,EAAGC,IAAO,CACpD,GAAI,CACA,IAAMC,EAAMN,EAAEK,CAAE,EAAE,KAAK,GAAK,GACtBE,EAAS,KAAK,MAAMD,CAAG,EAC7BH,EAAc,KAAKI,CAAM,CAC7B,MAAQ,CAER,CACJ,CAAC,EAGDP,EAAE,uEAAuE,EAAE,OAAO,EAOlF,IAAIQ,EAAc,GACdR,EAAE,MAAM,EAAE,OAAS,EACnBQ,EAAcR,EAAE,MAAM,EAAE,KAAK,GAAK,GAC3BA,EAAE,SAAS,EAAE,OAAS,EAC7BQ,EAAcR,EAAE,SAAS,EAAE,KAAK,GAAK,GAErCQ,EAAcR,EAAE,MAAM,EAAE,KAAK,GAAK,GAItC,IAAIS,EAAWd,EAAgB,SAASa,CAAW,EAc/CE,EAXgB,CAChB,KAAKT,CAAK,GACV,KAAKC,CAAW,GAChB,GACA,eAAeJ,CAAS,GACxB,kBAAkB,IAAI,KAAK,EAAE,YAAY,CAAC,GAC1C,GACA,MACA,EACJ,EAEgC,KAAK;AAAA,CAAI,EAAIW,EAGzCN,EAAc,OAAS,IACvBO,GAAiB;AAAA;AAAA;AAAA;AAAA;AAAA,EACjBP,EAAc,QAAQQ,GAAK,CACvBD,GAAiB,KAAK,UAAUC,EAAG,KAAM,CAAC,EAAI;AAAA,CAClD,CAAC,EACDD,GAAiB,SAGrB,IAAME,EAAeF,EAAc,OAC7BG,EAAsBd,EAAe,GAAMA,EAAea,GAAgBb,EAAgB,IAAM,EAEtG,MAAO,CACH,SAAUW,EACV,SAAU,CACN,MAAAT,EACA,YAAAC,EACA,OAAQC,CACZ,EACA,MAAO,CACH,iBAAkBJ,EAClB,aAAAa,EACA,oBAAAC,CACJ,CACJ,CACJ,CCtBO,SAASC,EAAgBC,EAA4B,CAC1D,IAAMC,EAAkB,CAAC,EAWzB,GARAA,EAAM,KAAK,KAAKD,EAAO,IAAI,EAAE,EAC7BC,EAAM,KAAK,EAAE,EAGbA,EAAM,KAAK,KAAKD,EAAO,OAAO,EAAE,EAChCC,EAAM,KAAK,EAAE,EAGTD,EAAO,QAAUA,EAAO,OAAO,OAAS,EAAG,CAC7CC,EAAM,KAAK,eAAe,EAC1BA,EAAM,KAAK,EAAE,EACb,QAAWC,KAASF,EAAO,OAAQ,CACjC,IAAMG,EAAU,GAAGH,EAAO,OAAO,GAAGE,EAAM,IAAI,GAC9CD,EAAM,KAAK,MAAMC,EAAM,IAAI,KAAKC,CAAO,MAAMD,EAAM,WAAW,EAAE,CAClE,CACAD,EAAM,KAAK,EAAE,CACf,CAGA,GAAID,EAAO,eAAiBA,EAAO,cAAc,OAAS,EAAG,CAC3DC,EAAM,KAAK,cAAc,EACzBA,EAAM,KAAK,EAAE,EACb,QAAWG,KAAQJ,EAAO,cACpBI,EAAK,YACPH,EAAM,KAAK,MAAMG,EAAK,KAAK,KAAKA,EAAK,GAAG,MAAMA,EAAK,WAAW,EAAE,EAEhEH,EAAM,KAAK,MAAMG,EAAK,KAAK,KAAKA,EAAK,GAAG,GAAG,EAG/CH,EAAM,KAAK,EAAE,CACf,CAGA,GAAID,EAAO,UAAYA,EAAO,SAAS,OAAS,EAC9C,QAAWK,KAAWL,EAAO,SAC3BC,EAAM,KAAK,MAAMI,EAAQ,OAAO,EAAE,EAClCJ,EAAM,KAAK,EAAE,EACbA,EAAM,KAAKI,EAAQ,OAAO,EAC1BJ,EAAM,KAAK,EAAE,EAIjB,OAAOA,EAAM,KAAK;AAAA,CAAI,EAAE,KAAK,EAAI;AAAA,CACnC,CCtGO,SAASK,EACdC,EACAC,EACsB,CACtB,MAAO,CACL,WAAY,qBACZ,QAAS,QACT,KAAM,oCACN,YAAa,2JACb,KAAM,CACJ,CACE,QAAS,YACT,KAAM,sBACN,KAAM,iLACN,SAAU,CACZ,EACA,CACE,QAAS,YACT,KAAM,+BACN,KAAM,4LACN,SAAU,CACZ,EACA,CACE,QAAS,YACT,KAAM,kBACN,KAAM,2KACN,SAAU,CACZ,EACA,CACE,QAAS,YACT,KAAM,gBACN,KAAM,0IACN,SAAU,CACZ,CACF,CACF,CACF,CAkBO,SAASC,EACdF,EACAC,EAC2B,CAC3B,GAAI,CAACD,EAAO,aACV,OAAO,KAGT,IAAMG,EAA6B,CACjC,WAAY,qBACZ,QAAS,eACT,KAAMH,EAAO,aAAa,IAC5B,EAEA,OAAIA,EAAO,aAAa,MACtBG,EAAO,IAAMH,EAAO,aAAa,KAG/BA,EAAO,aAAa,cACtBG,EAAO,YAAcH,EAAO,aAAa,aAGvCA,EAAO,aAAa,OACtBG,EAAO,KAAOH,EAAO,aAAa,MAGhCA,EAAO,aAAa,eACtBG,EAAO,aAAeH,EAAO,aAAa,cAGrCG,CACT,CAiBO,SAASC,EACdJ,EACAC,EACiB,CACjB,IAAMI,EAAYH,EAA2BF,EAAQC,CAAO,EAEtDE,EAA0B,CAC9B,WAAY,qBACZ,QAAS,YACT,KAAM,SAASH,EAAO,IAAI,GAC1B,IAAKC,CACP,EAEA,OAAID,EAAO,UACTG,EAAO,YAAcH,EAAO,SAG1BK,IACFF,EAAO,WAAaE,GAGfF,CACT,CAKO,SAASG,EACdC,EACAP,EACAC,EACY,CACZ,OAAQM,EAAU,CAChB,IAAK,UACH,OAAOR,EAA6BC,EAAQC,CAAO,EACrD,IAAK,QACH,OAAOG,EAAwBJ,EAAQC,CAAO,EAEhD,QACE,OAAO,IACX,CACF,CAKO,SAASO,EAAgBL,EAAmC,CACjE,OAAKA,EAGE,KAAK,UAAUA,EAAQ,KAAM,CAAC,EAF5B,IAGX","names":["cheerio","TurndownService","turndownService","extractContent","html","sourceUrl","originalSize","$","title","description","jsonLdScripts","_","el","raw","parsed","contentHtml","markdown","finalMarkdown","j","markdownSize","tokenReductionRatio","generateLlmsTxt","config","lines","route","fullUrl","link","section","generateAIOMethodologySchema","config","pageUrl","generateOrganizationSchema","schema","generateAboutPageSchema","orgSchema","generateSchemaForPageType","pageType","serializeSchema"]}
+{"version":3,"sources":["../src/extractor.ts","../src/config.ts","../src/middleware.ts","../src/bots.ts","../src/OntoHead.tsx","../src/OntoProvider.tsx","../src/schemas.ts"],"sourcesContent":["import * as cheerio from 'cheerio';\r\nimport TurndownService from 'turndown';\r\n\r\nconst turndownService = new TurndownService({\r\n    headingStyle: 'atx',\r\n    codeBlockStyle: 'fenced',\r\n});\r\n\r\n// Configure turndown to keep some layout or handle semantic tags differently if needed\r\n\r\nexport interface ExtractionResult {\r\n    markdown: string;\r\n    metadata: {\r\n        title: string;\r\n        description: string;\r\n        jsonLd: any[];\r\n    };\r\n    stats: {\r\n        originalHtmlSize: number;\r\n        markdownSize: number;\r\n        tokenReductionRatio: number;\r\n    };\r\n}\r\n\r\n/**\r\n * Extracts pure semantic markdown and metadata from rendered Next.js HTML strings.\r\n * @param html The raw HTML string.\r\n * @param sourceUrl (Optional) the URL this was generated from, to attach as metadata.\r\n * @returns {ExtractionResult} The extracted payload.\r\n */\r\nexport function extractContent(html: string, sourceUrl: string = 'Generated Output'): ExtractionResult {\r\n    const originalSize = html.length;\r\n\r\n    const $ = cheerio.load(html);\r\n\r\n    // 1. Extract Metadata BEFORE removing structure\r\n    const title = $('title').text() || $('h1').first().text() || 'Untitled Page';\r\n    const description = $('meta[name=\"description\"]').attr('content') || 'No description found.';\r\n\r\n    const jsonLdScripts: any[] = [];\r\n    $('script[type=\"application/ld+json\"]').each((_, el) => {\r\n        try {\r\n            const raw = $(el).html() || '';\r\n            const parsed = JSON.parse(raw);\r\n            jsonLdScripts.push(parsed);\r\n        } catch {\r\n            // ignore bad json\r\n        }\r\n    });\r\n\r\n    // 2. Strip noise (React boilerplate, styles, unnecessary tags)\r\n    $('script, style, noscript, iframe, svg, nav, footer, meta, link, header').remove();\r\n\r\n    // Optionally remove typical Next.js hidden wrappers if they don't contain real content.\r\n    // Next.js uses <div id=\"__next\"> but we mostly just want semantic content.\r\n\r\n    // 3. Find the entry point for content\r\n    // Prefer <main> or <article> over <body>\r\n    let contentHtml = '';\r\n    if ($('main').length > 0) {\r\n        contentHtml = $('main').html() || '';\r\n    } else if ($('article').length > 0) {\r\n        contentHtml = $('article').html() || '';\r\n    } else {\r\n        contentHtml = $('body').html() || '';\r\n    }\r\n\r\n    // 4. Convert to Markdown\r\n    let markdown = turndownService.turndown(contentHtml);\r\n\r\n    // 5. Optionally inject Metadata header\r\n    const headerLines = [\r\n        `# ${title}`,\r\n        `> ${description}`,\r\n        ``,\r\n        `**Source:** ${sourceUrl}`,\r\n        `**Extracted:** ${new Date().toISOString()}`,\r\n        ``,\r\n        `---`,\r\n        ``\r\n    ];\r\n\r\n    let finalMarkdown = headerLines.join('\\n') + markdown;\r\n\r\n    // Add JSON-LD section if exists\r\n    if (jsonLdScripts.length > 0) {\r\n        finalMarkdown += '\\n\\n---\\n## Structured Data (JSON-LD)\\n```json\\n';\r\n        jsonLdScripts.forEach(j => {\r\n            finalMarkdown += JSON.stringify(j, null, 2) + '\\n';\r\n        });\r\n        finalMarkdown += '```\\n';\r\n    }\r\n\r\n    const markdownSize = finalMarkdown.length;\r\n    const tokenReductionRatio = originalSize > 0 ? ((originalSize - markdownSize) / originalSize) * 100 : 0;\r\n\r\n    return {\r\n        markdown: finalMarkdown,\r\n        metadata: {\r\n            title,\r\n            description,\r\n            jsonLd: jsonLdScripts\r\n        },\r\n        stats: {\r\n            originalHtmlSize: originalSize,\r\n            markdownSize,\r\n            tokenReductionRatio\r\n        }\r\n    };\r\n}\r\n\r\nexport async function generateStaticPayloads(nextAppDirDir: string, ontoPublicDir: string) {\r\n    const fs = await import('fs');\r\n    const path = await import('path');\r\n    const { glob } = await import('glob');\r\n\r\n    if (!fs.existsSync(nextAppDirDir)) {\r\n        return;\r\n    }\r\n\r\n    const files = await glob('**/*.html', { cwd: nextAppDirDir });\r\n    if (files.length === 0) return;\r\n\r\n    if (!fs.existsSync(ontoPublicDir)) {\r\n        fs.mkdirSync(ontoPublicDir, { recursive: true });\r\n    }\r\n\r\n    let totalFilesProcessed = 0;\r\n\r\n    for (const file of files) {\r\n        const inputPath = path.join(nextAppDirDir, file);\r\n        const outputPathRelative = file.replace(/\\.html$/, '.md');\r\n        const outputPath = path.join(ontoPublicDir, outputPathRelative);\r\n\r\n        try {\r\n            const htmlContent = fs.readFileSync(inputPath, 'utf8');\r\n\r\n            let routeName = file.replace(/\\.html$/, '');\r\n            if (routeName === 'index') routeName = '/';\r\n            else routeName = `/${routeName}`;\r\n\r\n            const result = extractContent(htmlContent, routeName);\r\n\r\n            const outputDir = path.dirname(outputPath);\r\n            if (!fs.existsSync(outputDir)) {\r\n                fs.mkdirSync(outputDir, { recursive: true });\r\n            }\r\n\r\n            fs.writeFileSync(outputPath, result.markdown, 'utf8');\r\n            totalFilesProcessed++;\r\n        } catch (e: any) {\r\n            console.error(`[Onto] Failed to process ${file}: ${e.message}`);\r\n        }\r\n    }\r\n    console.log(`[Onto] Successfully generated ${totalFilesProcessed} semantic markdown endpoints.`);\r\n}\r\n","/**\r\n * Configuration schema for onto.config.ts\r\n * Used to dynamically generate llms.txt and other AI discovery files\r\n */\r\n\r\nexport type PageType = 'scoring' | 'about' | 'default';\r\n\r\nexport interface OntoRoute {\r\n  /**\r\n   * The URL path (e.g., '/docs', '/api/reference')\r\n   */\r\n  path: string;\r\n  /**\r\n   * Description of what this route contains\r\n   */\r\n  description: string;\r\n  /**\r\n   * Optional: Page type for automatic JSON-LD schema injection\r\n   * - 'scoring': Injects Methodology schema with AIO scoring weights (40/35/25)\r\n   * - 'about': Injects Organization/AboutPage schema\r\n   * - 'default': No automatic schema injection\r\n   */\r\n  pageType?: PageType;\r\n}\r\n\r\nexport interface OntoConfig {\r\n  /**\r\n   * The name of your project or site (required)\r\n   * Used as the H1 heading in llms.txt\r\n   */\r\n  name: string;\r\n\r\n  /**\r\n   * A short summary of your project (required)\r\n   * Displayed as a blockquote in llms.txt\r\n   * Should contain key information necessary for understanding the rest of the file\r\n   */\r\n  summary: string;\r\n\r\n  /**\r\n   * The base URL of your site (e.g., 'https://example.com')\r\n   */\r\n  baseUrl: string;\r\n\r\n  /**\r\n   * Optional: Additional sections to include in llms.txt\r\n   * Each section can contain any markdown content\r\n   */\r\n  sections?: {\r\n    heading: string;\r\n    content: string;\r\n  }[];\r\n\r\n  /**\r\n   * Key routes that AI agents should know about\r\n   * These will be formatted as a markdown list in llms.txt\r\n   */\r\n  routes?: OntoRoute[];\r\n\r\n  /**\r\n   * Optional: Links to external resources (documentation, API references, etc.)\r\n   */\r\n  externalLinks?: {\r\n    title: string;\r\n    url: string;\r\n    description?: string;\r\n  }[];\r\n\r\n  /**\r\n   * Optional: Organization information for JSON-LD schemas\r\n   */\r\n  organization?: {\r\n    name: string;\r\n    description?: string;\r\n    url?: string;\r\n    logo?: string;\r\n    foundingDate?: string;\r\n  };\r\n}\r\n\r\n/**\r\n * Generate llms.txt content from OntoConfig\r\n * Follows the llms.txt specification:\r\n * - H1 with project name\r\n * - Blockquote with summary\r\n * - Additional markdown sections\r\n */\r\nexport function generateLlmsTxt(config: OntoConfig): string {\r\n  const lines: string[] = [];\r\n\r\n  // H1: Project name (required)\r\n  lines.push(`# ${config.name}`);\r\n  lines.push('');\r\n\r\n  // Blockquote: Summary (required)\r\n  lines.push(`> ${config.summary}`);\r\n  lines.push('');\r\n\r\n  // Key Routes section (if provided)\r\n  if (config.routes && config.routes.length > 0) {\r\n    lines.push('## Key Routes');\r\n    lines.push('');\r\n    for (const route of config.routes) {\r\n      const fullUrl = `${config.baseUrl}${route.path}`;\r\n      lines.push(`- [${route.path}](${fullUrl}): ${route.description}`);\r\n    }\r\n    lines.push('');\r\n  }\r\n\r\n  // External Links section (if provided)\r\n  if (config.externalLinks && config.externalLinks.length > 0) {\r\n    lines.push('## Resources');\r\n    lines.push('');\r\n    for (const link of config.externalLinks) {\r\n      if (link.description) {\r\n        lines.push(`- [${link.title}](${link.url}): ${link.description}`);\r\n      } else {\r\n        lines.push(`- [${link.title}](${link.url})`);\r\n      }\r\n    }\r\n    lines.push('');\r\n  }\r\n\r\n  // Custom sections (if provided)\r\n  if (config.sections && config.sections.length > 0) {\r\n    for (const section of config.sections) {\r\n      lines.push(`## ${section.heading}`);\r\n      lines.push('');\r\n      lines.push(section.content);\r\n      lines.push('');\r\n    }\r\n  }\r\n\r\n  return lines.join('\\n').trim() + '\\n';\r\n}\r\n","import { NextRequest, NextResponse } from 'next/server';\r\nimport { matchBot } from './bots';\r\nimport { generateLlmsTxt, OntoConfig } from './config';\r\n\r\nexport async function ontoMiddleware(request: any, config?: OntoConfig) {\r\n    const userAgent = request.headers.get('user-agent');\r\n    const url = request.nextUrl.clone();\r\n    const matched = matchBot(userAgent);\r\n\r\n    const accept = request.headers.get('accept') || '';\r\n\r\n    const isAiBot = !!matched;\r\n    const isMarkdownRequested = accept.includes('text/markdown');\r\n\r\n    // If traffic is identified as an AI Bot or markdown is requested\r\n    if (isAiBot || isMarkdownRequested) {\r\n\r\n        // Ignore internal next.js requests & static assets (but not llms.txt)\r\n        if (url.pathname.startsWith('/_next')) {\r\n            return NextResponse.next();\r\n        }\r\n\r\n        // --- llms.txt Auto-Discovery ---\r\n        // Serve the llms.txt manifest to AI agents\r\n        if (url.pathname === '/llms.txt') {\r\n            try {\r\n                if (config) {\r\n                    // Generate llms.txt dynamically from config\r\n                    const llmsTxtContent = generateLlmsTxt(config);\r\n                    const response = new NextResponse(llmsTxtContent, {\r\n                        headers: {\r\n                            'Content-Type': 'text/plain; charset=utf-8',\r\n                            'Cache-Control': 'public, max-age=3600, s-maxage=3600, stale-while-revalidate=86400',\r\n                        }\r\n                    });\r\n\r\n                    if (matched) {\r\n                        response.headers.set('X-Onto-Bot', `${matched.name} (${matched.company})`);\r\n                    }\r\n\r\n                    return response;\r\n                } else {\r\n                    // Fallback: try to serve static llms.txt from public folder\r\n                    url.pathname = '/llms.txt';\r\n                    const response = NextResponse.rewrite(url);\r\n                    response.headers.set('Content-Type', 'text/plain; charset=utf-8');\r\n                    response.headers.set('Cache-Control', 'public, max-age=3600, s-maxage=3600, stale-while-revalidate=86400');\r\n                    if (matched) {\r\n                        response.headers.set('X-Onto-Bot', `${matched.name} (${matched.company})`);\r\n                    }\r\n                    return response;\r\n                }\r\n            } catch (error) {\r\n                console.error('[Onto] Failed to generate llms.txt:', error);\r\n                // Fallback to static file on error\r\n                url.pathname = '/llms.txt';\r\n                const response = NextResponse.rewrite(url);\r\n                response.headers.set('Content-Type', 'text/plain; charset=utf-8');\r\n                if (matched) {\r\n                    response.headers.set('X-Onto-Bot', `${matched.name} (${matched.company})`);\r\n                }\r\n                return response;\r\n            }\r\n        }\r\n\r\n        // Skip other static assets\r\n        if (url.pathname.includes('.')) {\r\n            return NextResponse.next();\r\n        }\r\n\r\n        // Determine the corresponding payload path\r\n        let payloadPath = url.pathname;\r\n        if (payloadPath === '/' || payloadPath === '') {\r\n            payloadPath = '/index';\r\n        }\r\n\r\n        // Strip trailing slash if present\r\n        if (payloadPath.endsWith('/') && payloadPath !== '/') {\r\n            payloadPath = payloadPath.slice(0, -1);\r\n        }\r\n\r\n        // Common response headers for all bot responses\r\n        const botHeaders: Record<string, string> = {\r\n            'Content-Type': 'text/markdown; charset=utf-8',\r\n            'Cache-Control': 'public, max-age=3600, s-maxage=3600, stale-while-revalidate=86400',\r\n        };\r\n        if (matched) {\r\n            botHeaders['X-Onto-Bot'] = `${matched.name} (${matched.company})`;\r\n        }\r\n\r\n        // --- Onto Control Plane Integration (Premium) ---\r\n        const ONTO_API_KEY = process.env.ONTO_API_KEY;\r\n        const DASHBOARD_URL = process.env.ONTO_DASHBOARD_URL || 'https://app.buildonto.dev';\r\n\r\n        if (ONTO_API_KEY) {\r\n            // 1. Fire-and-forget tracking — includes structured bot info\r\n            fetch(`${DASHBOARD_URL}/api/track`, {\r\n                method: 'POST',\r\n                headers: {\r\n                    'x-onto-key': ONTO_API_KEY,\r\n                    'Content-Type': 'application/json'\r\n                },\r\n                body: JSON.stringify({\r\n                    route: url.pathname,\r\n                    userAgent: userAgent,\r\n                    bot: matched ? matched.name : null,\r\n                    company: matched ? matched.company : null,\r\n                })\r\n            }).catch(() => {});\r\n\r\n            // 2. Dynamic Context Injection\r\n            try {\r\n                const injectRes = await fetch(`${DASHBOARD_URL}/api/sdk/inject?route=${url.pathname}`, {\r\n                    headers: { 'x-onto-key': ONTO_API_KEY },\r\n                    signal: AbortSignal.timeout(1500)\r\n                });\r\n\r\n                if (injectRes.ok) {\r\n                    const { injection } = await injectRes.json();\r\n                    \r\n                    if (injection) {\r\n                        const localMdUrl = `${url.origin}/.onto${payloadPath}.md`;\r\n                        const mdRes = await fetch(localMdUrl);\r\n                        \r\n                        if (mdRes.ok) {\r\n                            const baseMarkdown = await mdRes.text();\r\n                            const finalMarkdown = `${baseMarkdown}\\n\\n---\\n\\n${injection}`;\r\n                            \r\n                            return new NextResponse(finalMarkdown, {\r\n                                headers: {\r\n                                    ...botHeaders,\r\n                                    'X-Onto-Injected': 'true'\r\n                                }\r\n                            });\r\n                        }\r\n                    }\r\n                }\r\n            } catch (err) {\r\n                console.error('[Onto] Injection failed', err);\r\n            }\r\n        }\r\n        // ------------------------------------------------\r\n\r\n        url.pathname = `/.onto${payloadPath}.md`;\r\n\r\n        // Rewrite implicitly serves the target URL transparently to the client.\r\n        const response = NextResponse.rewrite(url);\r\n        if (matched) {\r\n            response.headers.set('X-Onto-Bot', `${matched.name} (${matched.company})`);\r\n        }\r\n        return response;\r\n    }\r\n\r\n    return NextResponse.next();\r\n}\r\n\r\n// Re-export the bot registry for consumers who want to extend or inspect it\r\nexport { AI_BOT_USER_AGENTS, matchBot } from './bots';\r\nexport type { AiBot } from './bots';\r\n\r\n\r\n","/**\n * Comprehensive registry of AI bot user-agent strings.\n * The middleware uses this list to detect AI crawlers and serve optimized markdown.\n */\n\nexport interface AiBot {\n    /** The user-agent substring to match against */\n    name: string;\n    /** The company operating this bot */\n    company: string;\n}\n\n/**\n * Structured registry of all known AI bots, grouped by company.\n * Useful for analytics and the Control Plane dashboard.\n */\nexport const AI_BOTS: AiBot[] = [\n    // OpenAI\n    { name: 'GPTBot',            company: 'OpenAI' },\n    { name: 'ChatGPT-User',     company: 'OpenAI' },\n    { name: 'OAI-SearchBot',    company: 'OpenAI' },\n\n    // Google\n    { name: 'Googlebot',              company: 'Google' },\n    { name: 'Google-CloudVertexBot',   company: 'Google' },\n    { name: 'Google-Extended',         company: 'Google' },\n    { name: 'GoogleOther',             company: 'Google' },\n\n    // Anthropic\n    { name: 'ClaudeBot',       company: 'Anthropic' },\n    { name: 'Claude-User',    company: 'Anthropic' },\n    { name: 'anthropic-ai',   company: 'Anthropic' },\n\n    // Perplexity\n    { name: 'PerplexityBot',   company: 'Perplexity' },\n    { name: 'Perplexity-User', company: 'Perplexity' },\n\n    // Meta\n    { name: 'Meta-ExternalAgent',   company: 'Meta' },\n    { name: 'Meta-ExternalFetcher', company: 'Meta' },\n    { name: 'FacebookBot',         company: 'Meta' },\n\n    // Common Crawl (used by most smaller AI companies)\n    { name: 'CCBot', company: 'Common Crawl' },\n\n    // Other notable AI crawlers\n    { name: 'Bytespider',        company: 'ByteDance' },\n    { name: 'Applebot-Extended', company: 'Apple' },\n    { name: 'cohere-ai',        company: 'Cohere' },\n    { name: 'YouBot',           company: 'You.com' },\n];\n\n/**\n * Flat list of user-agent substrings for fast matching in the middleware.\n */\nexport const AI_BOT_USER_AGENTS: string[] = AI_BOTS.map(bot => bot.name);\n\n/**\n * Given a raw user-agent string, returns the matched AiBot entry or undefined.\n * Comparison is case-insensitive to handle inconsistent agent casing.\n */\nexport function matchBot(userAgent: string | null): AiBot | undefined {\n    if (!userAgent) return undefined;\n    const lowerUA = userAgent.toLowerCase();\n    return AI_BOTS.find(bot => \n        lowerUA.includes(bot.name.toLowerCase())\n    );\n}\n","'use client';\n\nimport { usePathname } from 'next/navigation';\n\n/**\n * OntoHead — Auto-Discovery component for AI agents.\n * \n * Injects `<link rel=\"alternate\">` tags into the page `<head>` so AI crawlers\n * can discover the optimized markdown endpoint for the current route.\n * \n * Usage in a Next.js App Router layout:\n * ```tsx\n * import { OntoHead } from '@ontosdk/next/components';\n * \n * export default function RootLayout({ children }) {\n *   return (\n *     <html>\n *       <head>\n *         <OntoHead />\n *       </head>\n *       <body>{children}</body>\n *     </html>\n *   );\n * }\n * ```\n */\nexport function OntoHead() {\n    const pathname = usePathname();\n\n    // Map current route to its .onto markdown payload path\n    let payloadPath = pathname;\n    if (payloadPath === '/' || payloadPath === '') {\n        payloadPath = '/index';\n    }\n    if (payloadPath.endsWith('/') && payloadPath !== '/') {\n        payloadPath = payloadPath.slice(0, -1);\n    }\n\n    const markdownHref = `/.onto${payloadPath}.md`;\n\n    return (\n        <>\n            {/* Per-page markdown alternate for AI agents */}\n            <link\n                rel=\"alternate\"\n                type=\"text/markdown\"\n                href={markdownHref}\n                title=\"AI-optimized Markdown version\"\n            />\n            {/* Site-wide llms.txt manifest */}\n            <link\n                rel=\"alternate\"\n                type=\"text/plain\"\n                href=\"/llms.txt\"\n                title=\"LLM-readable site manifest\"\n            />\n        </>\n    );\n}\n","'use client';\r\n\r\nimport { usePathname } from 'next/navigation';\r\nimport { ReactNode, useMemo } from 'react';\r\nimport type { OntoConfig, PageType } from './config';\r\nimport { generateSchemaForPageType, serializeSchema } from './schemas';\r\n\r\nexport interface OntoProviderProps {\r\n  /**\r\n   * The base URL of your site (e.g., 'https://example.com')\r\n   * Used to construct the full href for the AI discovery link tag.\r\n   */\r\n  baseUrl: string;\r\n  /**\r\n   * Child components to render\r\n   */\r\n  children: ReactNode;\r\n  /**\r\n   * Optional: Onto configuration for automatic JSON-LD schema injection\r\n   * If provided, the provider will automatically inject JSON-LD schemas\r\n   * based on the page type configuration\r\n   */\r\n  config?: OntoConfig;\r\n}\r\n\r\n/**\r\n * OntoProvider — Automatic AI Discovery Provider\r\n *\r\n * Wraps your application and automatically injects:\r\n * 1. `<link rel=\"alternate\">` tags for AI discovery\r\n * 2. JSON-LD structured data schemas based on page type\r\n *\r\n * With config, automatically generates JSON-LD schemas:\r\n * - 'scoring' pages get Methodology schema with AIO weights (40/35/25)\r\n * - 'about' pages get Organization/AboutPage schema\r\n *\r\n * Usage in a Next.js App Router layout:\r\n * ```tsx\r\n * import { OntoProvider } from '@ontosdk/next/provider';\r\n * import config from '../onto.config';\r\n *\r\n * export default function RootLayout({ children }) {\r\n *   return (\r\n *     <OntoProvider baseUrl=\"https://example.com\" config={config}>\r\n *       <html>\r\n *         <head />\r\n *         <body>{children}</body>\r\n *       </html>\r\n *     </OntoProvider>\r\n *   );\r\n * }\r\n * ```\r\n */\r\nexport function OntoProvider({ baseUrl, children, config }: OntoProviderProps) {\r\n  const pathname = usePathname();\r\n\r\n  // Construct the full URL with the current path and ?format=md query string\r\n  const cleanBaseUrl = baseUrl.endsWith('/') ? baseUrl.slice(0, -1) : baseUrl;\r\n  const markdownHref = `${cleanBaseUrl}${pathname}?format=md`;\r\n  const fullPageUrl = `${cleanBaseUrl}${pathname}`;\r\n\r\n  // Determine page type from config routes\r\n  const pageType: PageType = useMemo(() => {\r\n    if (!config?.routes) return 'default';\r\n\r\n    const matchingRoute = config.routes.find(route => route.path === pathname);\r\n    return matchingRoute?.pageType || 'default';\r\n  }, [config, pathname]);\r\n\r\n  // Generate JSON-LD schema based on page type\r\n  const jsonLdSchema = useMemo(() => {\r\n    if (!config || pageType === 'default') return null;\r\n\r\n    const schema = generateSchemaForPageType(pageType, config, fullPageUrl);\r\n    return serializeSchema(schema);\r\n  }, [config, pageType, fullPageUrl]);\r\n\r\n  return (\r\n    <>\r\n      <link\r\n        rel=\"alternate\"\r\n        type=\"text/markdown\"\r\n        href={markdownHref}\r\n        title=\"AI-optimized Markdown version\"\r\n      />\r\n      {jsonLdSchema && (\r\n        <script\r\n          type=\"application/ld+json\"\r\n          dangerouslySetInnerHTML={{ __html: jsonLdSchema }}\r\n        />\r\n      )}\r\n      {children}\r\n    </>\r\n  );\r\n}\r\n","/**\r\n * JSON-LD Schema generators for automatic structured data injection\r\n * Follows Schema.org standards for AI-friendly metadata\r\n */\r\n\r\nimport { OntoConfig } from './config';\r\n\r\n/**\r\n * Standard AIO (AI Optimization) scoring methodology\r\n * Based on the Onto scoring algorithm:\r\n * - Content Negotiation: 40% (30 points)\r\n * - React Tax / Token Efficiency: 35% (30 points)\r\n * - Structured Data: 25% (25 points)\r\n * - Semantic HTML: Bonus (15 points)\r\n */\r\nexport interface AIOMethodologySchema {\r\n  '@context': 'https://schema.org';\r\n  '@type': 'HowTo';\r\n  name: string;\r\n  description: string;\r\n  step: Array<{\r\n    '@type': 'HowToStep';\r\n    name: string;\r\n    text: string;\r\n    position: number;\r\n  }>;\r\n}\r\n\r\n/**\r\n * Generate AIO Scoring Methodology JSON-LD schema\r\n * This explains to AI agents how the scoring system works\r\n */\r\nexport function generateAIOMethodologySchema(\r\n  config: OntoConfig,\r\n  pageUrl: string\r\n): AIOMethodologySchema {\r\n  return {\r\n    '@context': 'https://schema.org',\r\n    '@type': 'HowTo',\r\n    name: 'AIO Score Calculation Methodology',\r\n    description: 'AI Optimization (AIO) Score measures how well a website is optimized for AI agents and LLM crawlers. Scored out of 100 points based on four key metrics.',\r\n    step: [\r\n      {\r\n        '@type': 'HowToStep',\r\n        name: 'Content Negotiation',\r\n        text: 'Check if the site responds to Accept: text/markdown header. Weight: 40%. Penalty: -30 points if missing. This ensures AI bots receive optimized content instead of heavy HTML.',\r\n        position: 1\r\n      },\r\n      {\r\n        '@type': 'HowToStep',\r\n        name: 'Token Efficiency (React Tax)',\r\n        text: 'Measure the ratio of visible text to total HTML size. Weight: 35%. Penalty: -30 points if HTML > 100KB but text < 1KB. Detects JavaScript-heavy sites that are difficult for AI to parse.',\r\n        position: 2\r\n      },\r\n      {\r\n        '@type': 'HowToStep',\r\n        name: 'Structured Data',\r\n        text: 'Verify presence of JSON-LD structured data (Schema.org). Weight: 25%. Penalty: -25 points if missing. Enables AI to confidently extract pricing, products, and entities.',\r\n        position: 3\r\n      },\r\n      {\r\n        '@type': 'HowToStep',\r\n        name: 'Semantic HTML',\r\n        text: 'Check for semantic tags like <main> and <article>. Bonus: +15 points if present. Helps AI agents separate navigation from core content.',\r\n        position: 4\r\n      }\r\n    ]\r\n  };\r\n}\r\n\r\n/**\r\n * Organization schema for About pages\r\n */\r\nexport interface OrganizationSchema {\r\n  '@context': 'https://schema.org';\r\n  '@type': 'Organization';\r\n  name: string;\r\n  url?: string;\r\n  description?: string;\r\n  logo?: string;\r\n  foundingDate?: string;\r\n}\r\n\r\n/**\r\n * Generate Organization JSON-LD schema for About pages\r\n */\r\nexport function generateOrganizationSchema(\r\n  config: OntoConfig,\r\n  pageUrl: string\r\n): OrganizationSchema | null {\r\n  if (!config.organization) {\r\n    return null;\r\n  }\r\n\r\n  const schema: OrganizationSchema = {\r\n    '@context': 'https://schema.org',\r\n    '@type': 'Organization',\r\n    name: config.organization.name\r\n  };\r\n\r\n  if (config.organization.url) {\r\n    schema.url = config.organization.url;\r\n  }\r\n\r\n  if (config.organization.description) {\r\n    schema.description = config.organization.description;\r\n  }\r\n\r\n  if (config.organization.logo) {\r\n    schema.logo = config.organization.logo;\r\n  }\r\n\r\n  if (config.organization.foundingDate) {\r\n    schema.foundingDate = config.organization.foundingDate;\r\n  }\r\n\r\n  return schema;\r\n}\r\n\r\n/**\r\n * AboutPage schema combining Organization and WebPage\r\n */\r\nexport interface AboutPageSchema {\r\n  '@context': 'https://schema.org';\r\n  '@type': 'AboutPage';\r\n  name: string;\r\n  url: string;\r\n  description?: string;\r\n  mainEntity?: OrganizationSchema;\r\n}\r\n\r\n/**\r\n * Generate AboutPage JSON-LD schema\r\n */\r\nexport function generateAboutPageSchema(\r\n  config: OntoConfig,\r\n  pageUrl: string\r\n): AboutPageSchema {\r\n  const orgSchema = generateOrganizationSchema(config, pageUrl);\r\n\r\n  const schema: AboutPageSchema = {\r\n    '@context': 'https://schema.org',\r\n    '@type': 'AboutPage',\r\n    name: `About ${config.name}`,\r\n    url: pageUrl\r\n  };\r\n\r\n  if (config.summary) {\r\n    schema.description = config.summary;\r\n  }\r\n\r\n  if (orgSchema) {\r\n    schema.mainEntity = orgSchema;\r\n  }\r\n\r\n  return schema;\r\n}\r\n\r\n/**\r\n * Determine which schema to generate based on page type\r\n */\r\nexport function generateSchemaForPageType(\r\n  pageType: 'scoring' | 'about' | 'default',\r\n  config: OntoConfig,\r\n  pageUrl: string\r\n): any | null {\r\n  switch (pageType) {\r\n    case 'scoring':\r\n      return generateAIOMethodologySchema(config, pageUrl);\r\n    case 'about':\r\n      return generateAboutPageSchema(config, pageUrl);\r\n    case 'default':\r\n    default:\r\n      return null;\r\n  }\r\n}\r\n\r\n/**\r\n * Serialize schema to JSON-LD script tag content\r\n */\r\nexport function serializeSchema(schema: any | null): string | null {\r\n  if (!schema) {\r\n    return null;\r\n  }\r\n  return JSON.stringify(schema, null, 2);\r\n}\r\n"],"mappings":"AAAA,UAAYA,MAAa,UACzB,OAAOC,MAAqB,WAE5B,IAAMC,EAAkB,IAAID,EAAgB,CACxC,aAAc,MACd,eAAgB,QACpB,CAAC,EAwBM,SAASE,EAAeC,EAAcC,EAAoB,mBAAsC,CACnG,IAAMC,EAAeF,EAAK,OAEpBG,EAAY,OAAKH,CAAI,EAGrBI,EAAQD,EAAE,OAAO,EAAE,KAAK,GAAKA,EAAE,IAAI,EAAE,MAAM,EAAE,KAAK,GAAK,gBACvDE,EAAcF,EAAE,0BAA0B,EAAE,KAAK,SAAS,GAAK,wBAE/DG,EAAuB,CAAC,EAC9BH,EAAE,oCAAoC,EAAE,KAAK,CAACI,EAAGC,IAAO,CACpD,GAAI,CACA,IAAMC,EAAMN,EAAEK,CAAE,EAAE,KAAK,GAAK,GACtBE,EAAS,KAAK,MAAMD,CAAG,EAC7BH,EAAc,KAAKI,CAAM,CAC7B,MAAQ,CAER,CACJ,CAAC,EAGDP,EAAE,uEAAuE,EAAE,OAAO,EAOlF,IAAIQ,EAAc,GACdR,EAAE,MAAM,EAAE,OAAS,EACnBQ,EAAcR,EAAE,MAAM,EAAE,KAAK,GAAK,GAC3BA,EAAE,SAAS,EAAE,OAAS,EAC7BQ,EAAcR,EAAE,SAAS,EAAE,KAAK,GAAK,GAErCQ,EAAcR,EAAE,MAAM,EAAE,KAAK,GAAK,GAItC,IAAIS,EAAWd,EAAgB,SAASa,CAAW,EAc/CE,EAXgB,CAChB,KAAKT,CAAK,GACV,KAAKC,CAAW,GAChB,GACA,eAAeJ,CAAS,GACxB,kBAAkB,IAAI,KAAK,EAAE,YAAY,CAAC,GAC1C,GACA,MACA,EACJ,EAEgC,KAAK;AAAA,CAAI,EAAIW,EAGzCN,EAAc,OAAS,IACvBO,GAAiB;AAAA;AAAA;AAAA;AAAA;AAAA,EACjBP,EAAc,QAAQQ,GAAK,CACvBD,GAAiB,KAAK,UAAUC,EAAG,KAAM,CAAC,EAAI;AAAA,CAClD,CAAC,EACDD,GAAiB,SAGrB,IAAME,EAAeF,EAAc,OAC7BG,EAAsBd,EAAe,GAAMA,EAAea,GAAgBb,EAAgB,IAAM,EAEtG,MAAO,CACH,SAAUW,EACV,SAAU,CACN,MAAAT,EACA,YAAAC,EACA,OAAQC,CACZ,EACA,MAAO,CACH,iBAAkBJ,EAClB,aAAAa,EACA,oBAAAC,CACJ,CACJ,CACJ,CCtBO,SAASC,EAAgBC,EAA4B,CAC1D,IAAMC,EAAkB,CAAC,EAWzB,GARAA,EAAM,KAAK,KAAKD,EAAO,IAAI,EAAE,EAC7BC,EAAM,KAAK,EAAE,EAGbA,EAAM,KAAK,KAAKD,EAAO,OAAO,EAAE,EAChCC,EAAM,KAAK,EAAE,EAGTD,EAAO,QAAUA,EAAO,OAAO,OAAS,EAAG,CAC7CC,EAAM,KAAK,eAAe,EAC1BA,EAAM,KAAK,EAAE,EACb,QAAWC,KAASF,EAAO,OAAQ,CACjC,IAAMG,EAAU,GAAGH,EAAO,OAAO,GAAGE,EAAM,IAAI,GAC9CD,EAAM,KAAK,MAAMC,EAAM,IAAI,KAAKC,CAAO,MAAMD,EAAM,WAAW,EAAE,CAClE,CACAD,EAAM,KAAK,EAAE,CACf,CAGA,GAAID,EAAO,eAAiBA,EAAO,cAAc,OAAS,EAAG,CAC3DC,EAAM,KAAK,cAAc,EACzBA,EAAM,KAAK,EAAE,EACb,QAAWG,KAAQJ,EAAO,cACpBI,EAAK,YACPH,EAAM,KAAK,MAAMG,EAAK,KAAK,KAAKA,EAAK,GAAG,MAAMA,EAAK,WAAW,EAAE,EAEhEH,EAAM,KAAK,MAAMG,EAAK,KAAK,KAAKA,EAAK,GAAG,GAAG,EAG/CH,EAAM,KAAK,EAAE,CACf,CAGA,GAAID,EAAO,UAAYA,EAAO,SAAS,OAAS,EAC9C,QAAWK,KAAWL,EAAO,SAC3BC,EAAM,KAAK,MAAMI,EAAQ,OAAO,EAAE,EAClCJ,EAAM,KAAK,EAAE,EACbA,EAAM,KAAKI,EAAQ,OAAO,EAC1BJ,EAAM,KAAK,EAAE,EAIjB,OAAOA,EAAM,KAAK;AAAA,CAAI,EAAE,KAAK,EAAI;AAAA,CACnC,CCtIA,OAAsB,gBAAAK,MAAoB,cCgBnC,IAAMC,EAAmB,CAE5B,CAAE,KAAM,SAAqB,QAAS,QAAS,EAC/C,CAAE,KAAM,eAAoB,QAAS,QAAS,EAC9C,CAAE,KAAM,gBAAoB,QAAS,QAAS,EAG9C,CAAE,KAAM,YAA0B,QAAS,QAAS,EACpD,CAAE,KAAM,wBAA2B,QAAS,QAAS,EACrD,CAAE,KAAM,kBAA2B,QAAS,QAAS,EACrD,CAAE,KAAM,cAA2B,QAAS,QAAS,EAGrD,CAAE,KAAM,YAAmB,QAAS,WAAY,EAChD,CAAE,KAAM,cAAkB,QAAS,WAAY,EAC/C,CAAE,KAAM,eAAkB,QAAS,WAAY,EAG/C,CAAE,KAAM,gBAAmB,QAAS,YAAa,EACjD,CAAE,KAAM,kBAAmB,QAAS,YAAa,EAGjD,CAAE,KAAM,qBAAwB,QAAS,MAAO,EAChD,CAAE,KAAM,uBAAwB,QAAS,MAAO,EAChD,CAAE,KAAM,cAAuB,QAAS,MAAO,EAG/C,CAAE,KAAM,QAAS,QAAS,cAAe,EAGzC,CAAE,KAAM,aAAqB,QAAS,WAAY,EAClD,CAAE,KAAM,oBAAqB,QAAS,OAAQ,EAC9C,CAAE,KAAM,YAAoB,QAAS,QAAS,EAC9C,CAAE,KAAM,SAAoB,QAAS,SAAU,CACnD,EAKaC,EAA+BD,EAAQ,IAAIE,GAAOA,EAAI,IAAI,EAMhE,SAASC,EAASC,EAA6C,CAClE,GAAI,CAACA,EAAW,OAChB,IAAMC,EAAUD,EAAU,YAAY,EACtC,OAAOJ,EAAQ,KAAKE,GAChBG,EAAQ,SAASH,EAAI,KAAK,YAAY,CAAC,CAC3C,CACJ,CD/DA,eAAsBI,EAAeC,EAAcC,EAAqB,CACpE,IAAMC,EAAYF,EAAQ,QAAQ,IAAI,YAAY,EAC5CG,EAAMH,EAAQ,QAAQ,MAAM,EAC5BI,EAAUC,EAASH,CAAS,EAE5BI,EAASN,EAAQ,QAAQ,IAAI,QAAQ,GAAK,GAE1CO,EAAU,CAAC,CAACH,EACZI,EAAsBF,EAAO,SAAS,eAAe,EAG3D,GAAIC,GAAWC,EAAqB,CAGhC,GAAIL,EAAI,SAAS,WAAW,QAAQ,EAChC,OAAOM,EAAa,KAAK,EAK7B,GAAIN,EAAI,WAAa,YACjB,GAAI,CACA,GAAIF,EAAQ,CAER,IAAMS,EAAiBC,EAAgBV,CAAM,EACvCW,EAAW,IAAIH,EAAaC,EAAgB,CAC9C,QAAS,CACL,eAAgB,4BAChB,gBAAiB,mEACrB,CACJ,CAAC,EAED,OAAIN,GACAQ,EAAS,QAAQ,IAAI,aAAc,GAAGR,EAAQ,IAAI,KAAKA,EAAQ,OAAO,GAAG,EAGtEQ,CACX,KAAO,CAEHT,EAAI,SAAW,YACf,IAAMS,EAAWH,EAAa,QAAQN,CAAG,EACzC,OAAAS,EAAS,QAAQ,IAAI,eAAgB,2BAA2B,EAChEA,EAAS,QAAQ,IAAI,gBAAiB,mEAAmE,EACrGR,GACAQ,EAAS,QAAQ,IAAI,aAAc,GAAGR,EAAQ,IAAI,KAAKA,EAAQ,OAAO,GAAG,EAEtEQ,CACX,CACJ,OAASC,EAAO,CACZ,QAAQ,MAAM,sCAAuCA,CAAK,EAE1DV,EAAI,SAAW,YACf,IAAMS,EAAWH,EAAa,QAAQN,CAAG,EACzC,OAAAS,EAAS,QAAQ,IAAI,eAAgB,2BAA2B,EAC5DR,GACAQ,EAAS,QAAQ,IAAI,aAAc,GAAGR,EAAQ,IAAI,KAAKA,EAAQ,OAAO,GAAG,EAEtEQ,CACX,CAIJ,GAAIT,EAAI,SAAS,SAAS,GAAG,EACzB,OAAOM,EAAa,KAAK,EAI7B,IAAIK,EAAcX,EAAI,UAClBW,IAAgB,KAAOA,IAAgB,MACvCA,EAAc,UAIdA,EAAY,SAAS,GAAG,GAAKA,IAAgB,MAC7CA,EAAcA,EAAY,MAAM,EAAG,EAAE,GAIzC,IAAMC,EAAqC,CACvC,eAAgB,+BAChB,gBAAiB,mEACrB,EACIX,IACAW,EAAW,YAAY,EAAI,GAAGX,EAAQ,IAAI,KAAKA,EAAQ,OAAO,KAIlE,IAAMY,EAAe,QAAQ,IAAI,aAC3BC,EAAgB,QAAQ,IAAI,oBAAsB,4BAExD,GAAID,EAAc,CAEd,MAAM,GAAGC,CAAa,aAAc,CAChC,OAAQ,OACR,QAAS,CACL,aAAcD,EACd,eAAgB,kBACpB,EACA,KAAM,KAAK,UAAU,CACjB,MAAOb,EAAI,SACX,UAAWD,EACX,IAAKE,EAAUA,EAAQ,KAAO,KAC9B,QAASA,EAAUA,EAAQ,QAAU,IACzC,CAAC,CACL,CAAC,EAAE,MAAM,IAAM,CAAC,CAAC,EAGjB,GAAI,CACA,IAAMc,EAAY,MAAM,MAAM,GAAGD,CAAa,yBAAyBd,EAAI,QAAQ,GAAI,CACnF,QAAS,CAAE,aAAca,CAAa,EACtC,OAAQ,YAAY,QAAQ,IAAI,CACpC,CAAC,EAED,GAAIE,EAAU,GAAI,CACd,GAAM,CAAE,UAAAC,CAAU,EAAI,MAAMD,EAAU,KAAK,EAE3C,GAAIC,EAAW,CACX,IAAMC,EAAa,GAAGjB,EAAI,MAAM,SAASW,CAAW,MAC9CO,EAAQ,MAAM,MAAMD,CAAU,EAEpC,GAAIC,EAAM,GAAI,CAEV,IAAMC,EAAgB,GADD,MAAMD,EAAM,KAAK,CACD;AAAA;AAAA;AAAA;AAAA,EAAcF,CAAS,GAE5D,OAAO,IAAIV,EAAaa,EAAe,CACnC,QAAS,CACL,GAAGP,EACH,kBAAmB,MACvB,CACJ,CAAC,CACL,CACJ,CACJ,CACJ,OAASQ,EAAK,CACV,QAAQ,MAAM,0BAA2BA,CAAG,CAChD,CACJ,CAGApB,EAAI,SAAW,SAASW,CAAW,MAGnC,IAAMF,EAAWH,EAAa,QAAQN,CAAG,EACzC,OAAIC,GACAQ,EAAS,QAAQ,IAAI,aAAc,GAAGR,EAAQ,IAAI,KAAKA,EAAQ,OAAO,GAAG,EAEtEQ,CACX,CAEA,OAAOH,EAAa,KAAK,CAC7B,CExJA,OAAS,eAAAe,MAAmB,kBAuCpB,mBAAAC,EAEI,OAAAC,EAFJ,QAAAC,MAAA,oBAfD,SAASC,GAAW,CAIvB,IAAIC,EAHaL,EAAY,GAIzBK,IAAgB,KAAOA,IAAgB,MACvCA,EAAc,UAEdA,EAAY,SAAS,GAAG,GAAKA,IAAgB,MAC7CA,EAAcA,EAAY,MAAM,EAAG,EAAE,GAGzC,IAAMC,EAAe,SAASD,CAAW,MAEzC,OACIF,EAAAF,EAAA,CAEI,UAAAC,EAAC,QACG,IAAI,YACJ,KAAK,gBACL,KAAMI,EACN,MAAM,gCACV,EAEAJ,EAAC,QACG,IAAI,YACJ,KAAK,aACL,KAAK,YACL,MAAM,6BACV,GACJ,CAER,CCxDA,OAAS,eAAAK,MAAmB,kBAC5B,OAAoB,WAAAC,MAAe,QC6B5B,SAASC,EACdC,EACAC,EACsB,CACtB,MAAO,CACL,WAAY,qBACZ,QAAS,QACT,KAAM,oCACN,YAAa,2JACb,KAAM,CACJ,CACE,QAAS,YACT,KAAM,sBACN,KAAM,iLACN,SAAU,CACZ,EACA,CACE,QAAS,YACT,KAAM,+BACN,KAAM,4LACN,SAAU,CACZ,EACA,CACE,QAAS,YACT,KAAM,kBACN,KAAM,2KACN,SAAU,CACZ,EACA,CACE,QAAS,YACT,KAAM,gBACN,KAAM,0IACN,SAAU,CACZ,CACF,CACF,CACF,CAkBO,SAASC,EACdF,EACAC,EAC2B,CAC3B,GAAI,CAACD,EAAO,aACV,OAAO,KAGT,IAAMG,EAA6B,CACjC,WAAY,qBACZ,QAAS,eACT,KAAMH,EAAO,aAAa,IAC5B,EAEA,OAAIA,EAAO,aAAa,MACtBG,EAAO,IAAMH,EAAO,aAAa,KAG/BA,EAAO,aAAa,cACtBG,EAAO,YAAcH,EAAO,aAAa,aAGvCA,EAAO,aAAa,OACtBG,EAAO,KAAOH,EAAO,aAAa,MAGhCA,EAAO,aAAa,eACtBG,EAAO,aAAeH,EAAO,aAAa,cAGrCG,CACT,CAiBO,SAASC,EACdJ,EACAC,EACiB,CACjB,IAAMI,EAAYH,EAA2BF,EAAQC,CAAO,EAEtDE,EAA0B,CAC9B,WAAY,qBACZ,QAAS,YACT,KAAM,SAASH,EAAO,IAAI,GAC1B,IAAKC,CACP,EAEA,OAAID,EAAO,UACTG,EAAO,YAAcH,EAAO,SAG1BK,IACFF,EAAO,WAAaE,GAGfF,CACT,CAKO,SAASG,EACdC,EACAP,EACAC,EACY,CACZ,OAAQM,EAAU,CAChB,IAAK,UACH,OAAOR,EAA6BC,EAAQC,CAAO,EACrD,IAAK,QACH,OAAOG,EAAwBJ,EAAQC,CAAO,EAEhD,QACE,OAAO,IACX,CACF,CAKO,SAASO,EAAgBL,EAAmC,CACjE,OAAKA,EAGE,KAAK,UAAUA,EAAQ,KAAM,CAAC,EAF5B,IAGX,CD3GI,mBAAAM,EACE,OAAAC,EADF,QAAAC,MAAA,oBAzBG,SAASC,EAAa,CAAE,QAAAC,EAAS,SAAAC,EAAU,OAAAC,CAAO,EAAsB,CAC7E,IAAMC,EAAWC,EAAY,EAGvBC,EAAeL,EAAQ,SAAS,GAAG,EAAIA,EAAQ,MAAM,EAAG,EAAE,EAAIA,EAC9DM,EAAe,GAAGD,CAAY,GAAGF,CAAQ,aACzCI,EAAc,GAAGF,CAAY,GAAGF,CAAQ,GAGxCK,EAAqBC,EAAQ,IAC5BP,GAAQ,QAESA,EAAO,OAAO,KAAKQ,GAASA,EAAM,OAASP,CAAQ,GACnD,UAAY,UACjC,CAACD,EAAQC,CAAQ,CAAC,EAGfQ,EAAeF,EAAQ,IAAM,CACjC,GAAI,CAACP,GAAUM,IAAa,UAAW,OAAO,KAE9C,IAAMI,EAASC,EAA0BL,EAAUN,EAAQK,CAAW,EACtE,OAAOO,EAAgBF,CAAM,CAC/B,EAAG,CAACV,EAAQM,EAAUD,CAAW,CAAC,EAElC,OACET,EAAAF,EAAA,CACE,UAAAC,EAAC,QACC,IAAI,YACJ,KAAK,gBACL,KAAMS,EACN,MAAM,gCACR,EACCK,GACCd,EAAC,UACC,KAAK,sBACL,wBAAyB,CAAE,OAAQc,CAAa,EAClD,EAEDV,GACH,CAEJ","names":["cheerio","TurndownService","turndownService","extractContent","html","sourceUrl","originalSize","$","title","description","jsonLdScripts","_","el","raw","parsed","contentHtml","markdown","finalMarkdown","j","markdownSize","tokenReductionRatio","generateLlmsTxt","config","lines","route","fullUrl","link","section","NextResponse","AI_BOTS","AI_BOT_USER_AGENTS","bot","matchBot","userAgent","lowerUA","ontoMiddleware","request","config","userAgent","url","matched","matchBot","accept","isAiBot","isMarkdownRequested","NextResponse","llmsTxtContent","generateLlmsTxt","response","error","payloadPath","botHeaders","ONTO_API_KEY","DASHBOARD_URL","injectRes","injection","localMdUrl","mdRes","finalMarkdown","err","usePathname","Fragment","jsx","jsxs","OntoHead","payloadPath","markdownHref","usePathname","useMemo","generateAIOMethodologySchema","config","pageUrl","generateOrganizationSchema","schema","generateAboutPageSchema","orgSchema","generateSchemaForPageType","pageType","serializeSchema","Fragment","jsx","jsxs","OntoProvider","baseUrl","children","config","pathname","usePathname","cleanBaseUrl","markdownHref","fullPageUrl","pageType","useMemo","route","jsonLdSchema","schema","generateSchemaForPageType","serializeSchema"]}

package/package.json

@@ -1,6 +1,6 @@
 {
     "name": "@ontosdk/next",
-    "version": "1.3.9",
+    "version": "1.4.0",
     "description": "Extracts semantic Markdown from React/Next.js pages for AI Agents",
     "main": "dist/index.js",
     "module": "dist/index.mjs",

package/src/index.ts

@@ -3,6 +3,9 @@
 export { extractContent } from './extractor';
 export { OntoConfig, OntoRoute, generateLlmsTxt } from './config';
 export type { OntoConfig as OntoConfigType, OntoRoute as OntoRouteType, PageType } from './config';
+export { ontoMiddleware } from './middleware';
+export { OntoHead } from './OntoHead';
+export { OntoProvider } from './OntoProvider';
 export {
   generateAIOMethodologySchema,
   generateOrganizationSchema,