@ontosdk/next 1.2.0 → 1.3.1

package/dist/OntoProvider.js

@@ -0,0 +1,2 @@
+"use strict";"use client";var s=Object.defineProperty;var y=Object.getOwnPropertyDescriptor;var S=Object.getOwnPropertyNames;var x=Object.prototype.hasOwnProperty;var O=(t,n)=>{for(var e in n)s(t,e,{get:n[e],enumerable:!0})},z=(t,n,e,o)=>{if(n&&typeof n=="object"||typeof n=="function")for(let i of S(n))!x.call(t,i)&&i!==e&&s(t,i,{get:()=>n[i],enumerable:!(o=y(n,i))||o.enumerable});return t};var A=t=>z(s({},"__esModule",{value:!0}),t);var H={};O(H,{OntoProvider:()=>b});module.exports=A(H);var h=require("next/navigation"),c=require("react");function T(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 P(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 w(t,n){let e=P(t,n),o={"@context":"https://schema.org","@type":"AboutPage",name:`About ${t.name}`,url:n};return t.summary&&(o.description=t.summary),e&&(o.mainEntity=e),o}function u(t,n,e){switch(t){case"scoring":return T(n,e);case"about":return w(n,e);default:return null}}function l(t){return t?JSON.stringify(t,null,2):null}var a=require("react/jsx-runtime");function b({baseUrl:t,children:n,config:e}){let o=(0,h.usePathname)(),i=t.endsWith("/")?t.slice(0,-1):t,d=`${i}${o}?format=md`,p=`${i}${o}`,r=(0,c.useMemo)(()=>e?.routes&&e.routes.find(f=>f.path===o)?.pageType||"default",[e,o]),g=(0,c.useMemo)(()=>{if(!e||r==="default")return null;let m=u(r,e,p);return l(m)},[e,r,p]);return(0,a.jsxs)(a.Fragment,{children:[(0,a.jsx)("link",{rel:"alternate",type:"text/markdown",href:d,title:"AI-optimized Markdown version"}),g&&(0,a.jsx)("script",{type:"application/ld+json",dangerouslySetInnerHTML:{__html:g}}),n]})}0&&(module.exports={OntoProvider});
+//# sourceMappingURL=OntoProvider.js.map

package/dist/OntoProvider.js.map

@@ -0,0 +1 @@
+{"version":3,"sources":["../src/OntoProvider.tsx","../src/schemas.ts"],"sourcesContent":["'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":"sbAAA,IAAAA,EAAA,GAAAC,EAAAD,EAAA,kBAAAE,IAAA,eAAAC,EAAAH,GAEA,IAAAI,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":["OntoProvider_exports","__export","OntoProvider","__toCommonJS","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/OntoProvider.mjs

@@ -0,0 +1,2 @@
+"use client";import{usePathname as S}from"next/navigation";import{useMemo as m}from"react";function d(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 f(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 y(t,n){let e=f(t,n),o={"@context":"https://schema.org","@type":"AboutPage",name:`About ${t.name}`,url:n};return t.summary&&(o.description=t.summary),e&&(o.mainEntity=e),o}function p(t,n,e){switch(t){case"scoring":return d(n,e);case"about":return y(n,e);default:return null}}function g(t){return t?JSON.stringify(t,null,2):null}import{Fragment as x,jsx as u,jsxs as O}from"react/jsx-runtime";function b({baseUrl:t,children:n,config:e}){let o=S(),a=t.endsWith("/")?t.slice(0,-1):t,l=`${a}${o}?format=md`,r=`${a}${o}`,i=m(()=>e?.routes&&e.routes.find(h=>h.path===o)?.pageType||"default",[e,o]),s=m(()=>{if(!e||i==="default")return null;let c=p(i,e,r);return g(c)},[e,i,r]);return O(x,{children:[u("link",{rel:"alternate",type:"text/markdown",href:l,title:"AI-optimized Markdown version"}),s&&u("script",{type:"application/ld+json",dangerouslySetInnerHTML:{__html:s}}),n]})}export{b as OntoProvider};
+//# sourceMappingURL=OntoProvider.mjs.map

package/dist/OntoProvider.mjs.map

@@ -0,0 +1 @@
+{"version":3,"sources":["../src/OntoProvider.tsx","../src/schemas.ts"],"sourcesContent":["'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":"aAEA,OAAS,eAAAA,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":["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/dist/config.d.mts

@@ -0,0 +1,85 @@
+/**
+ * Configuration schema for onto.config.ts
+ * Used to dynamically generate llms.txt and other AI discovery files
+ */
+type PageType = 'scoring' | 'about' | 'default';
+interface OntoRoute {
+    /**
+     * The URL path (e.g., '/docs', '/api/reference')
+     */
+    path: string;
+    /**
+     * Description of what this route contains
+     */
+    description: string;
+    /**
+     * Optional: Page type for automatic JSON-LD schema injection
+     * - 'scoring': Injects Methodology schema with AIO scoring weights (40/35/25)
+     * - 'about': Injects Organization/AboutPage schema
+     * - 'default': No automatic schema injection
+     */
+    pageType?: PageType;
+}
+interface OntoConfig {
+    /**
+     * The name of your project or site (required)
+     * Used as the H1 heading in llms.txt
+     */
+    name: string;
+    /**
+     * A short summary of your project (required)
+     * Displayed as a blockquote in llms.txt
+     * Should contain key information necessary for understanding the rest of the file
+     */
+    summary: string;
+    /**
+     * The base URL of your site (e.g., 'https://example.com')
+     */
+    baseUrl: string;
+    /**
+     * Optional: Additional sections to include in llms.txt
+     * Each section can contain any markdown content
+     */
+    sections?: {
+        heading: string;
+        content: string;
+    }[];
+    /**
+     * Key routes that AI agents should know about
+     * These will be formatted as a markdown list in llms.txt
+     */
+    routes?: OntoRoute[];
+    /**
+     * Optional: Links to external resources (documentation, API references, etc.)
+     */
+    externalLinks?: {
+        title: string;
+        url: string;
+        description?: string;
+    }[];
+    /**
+     * Optional: Organization information for JSON-LD schemas
+     */
+    organization?: {
+        name: string;
+        description?: string;
+        url?: string;
+        logo?: string;
+        foundingDate?: string;
+    };
+}
+/**
+ * Load the onto.config.ts file from the user's project
+ * This is used by the middleware to dynamically generate llms.txt
+ */
+declare function loadOntoConfig(): Promise<OntoConfig | null>;
+/**
+ * Generate llms.txt content from OntoConfig
+ * Follows the llms.txt specification:
+ * - H1 with project name
+ * - Blockquote with summary
+ * - Additional markdown sections
+ */
+declare function generateLlmsTxt(config: OntoConfig): string;
+
+export { type OntoConfig, type OntoRoute, type PageType, generateLlmsTxt, loadOntoConfig };

package/dist/config.d.ts

@@ -0,0 +1,85 @@
+/**
+ * Configuration schema for onto.config.ts
+ * Used to dynamically generate llms.txt and other AI discovery files
+ */
+type PageType = 'scoring' | 'about' | 'default';
+interface OntoRoute {
+    /**
+     * The URL path (e.g., '/docs', '/api/reference')
+     */
+    path: string;
+    /**
+     * Description of what this route contains
+     */
+    description: string;
+    /**
+     * Optional: Page type for automatic JSON-LD schema injection
+     * - 'scoring': Injects Methodology schema with AIO scoring weights (40/35/25)
+     * - 'about': Injects Organization/AboutPage schema
+     * - 'default': No automatic schema injection
+     */
+    pageType?: PageType;
+}
+interface OntoConfig {
+    /**
+     * The name of your project or site (required)
+     * Used as the H1 heading in llms.txt
+     */
+    name: string;
+    /**
+     * A short summary of your project (required)
+     * Displayed as a blockquote in llms.txt
+     * Should contain key information necessary for understanding the rest of the file
+     */
+    summary: string;
+    /**
+     * The base URL of your site (e.g., 'https://example.com')
+     */
+    baseUrl: string;
+    /**
+     * Optional: Additional sections to include in llms.txt
+     * Each section can contain any markdown content
+     */
+    sections?: {
+        heading: string;
+        content: string;
+    }[];
+    /**
+     * Key routes that AI agents should know about
+     * These will be formatted as a markdown list in llms.txt
+     */
+    routes?: OntoRoute[];
+    /**
+     * Optional: Links to external resources (documentation, API references, etc.)
+     */
+    externalLinks?: {
+        title: string;
+        url: string;
+        description?: string;
+    }[];
+    /**
+     * Optional: Organization information for JSON-LD schemas
+     */
+    organization?: {
+        name: string;
+        description?: string;
+        url?: string;
+        logo?: string;
+        foundingDate?: string;
+    };
+}
+/**
+ * Load the onto.config.ts file from the user's project
+ * This is used by the middleware to dynamically generate llms.txt
+ */
+declare function loadOntoConfig(): Promise<OntoConfig | null>;
+/**
+ * Generate llms.txt content from OntoConfig
+ * Follows the llms.txt specification:
+ * - H1 with project name
+ * - Blockquote with summary
+ * - Additional markdown sections
+ */
+declare function generateLlmsTxt(config: OntoConfig): string;
+
+export { type OntoConfig, type OntoRoute, type PageType, generateLlmsTxt, loadOntoConfig };

package/dist/config.js

@@ -0,0 +1,4 @@
+"use strict";var i=Object.defineProperty;var o=Object.getOwnPropertyDescriptor;var u=Object.getOwnPropertyNames;var p=Object.prototype.hasOwnProperty;var a=(n,t)=>{for(var s in t)i(n,s,{get:t[s],enumerable:!0})},l=(n,t,s,r)=>{if(t&&typeof t=="object"||typeof t=="function")for(let e of u(t))!p.call(n,e)&&e!==s&&i(n,e,{get:()=>t[e],enumerable:!(r=o(t,e))||r.enumerable});return n};var h=n=>l(i({},"__esModule",{value:!0}),n);var d={};a(d,{generateLlmsTxt:()=>c,loadOntoConfig:()=>g});module.exports=h(d);async function g(){try{let n=await import(process.cwd()+"/onto.config");return n.default||n}catch{return null}}function c(n){let t=[];if(t.push(`# ${n.name}`),t.push(""),t.push(`> ${n.summary}`),t.push(""),n.routes&&n.routes.length>0){t.push("## Key Routes"),t.push("");for(let s of n.routes){let r=`${n.baseUrl}${s.path}`;t.push(`- [${s.path}](${r}): ${s.description}`)}t.push("")}if(n.externalLinks&&n.externalLinks.length>0){t.push("## Resources"),t.push("");for(let s of n.externalLinks)s.description?t.push(`- [${s.title}](${s.url}): ${s.description}`):t.push(`- [${s.title}](${s.url})`);t.push("")}if(n.sections&&n.sections.length>0)for(let s of n.sections)t.push(`## ${s.heading}`),t.push(""),t.push(s.content),t.push("");return t.join(`
+`).trim()+`
+`}0&&(module.exports={generateLlmsTxt,loadOntoConfig});
+//# sourceMappingURL=config.js.map

package/dist/config.js.map

@@ -0,0 +1 @@
+{"version":3,"sources":["../src/config.ts"],"sourcesContent":["/**\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 * Load the onto.config.ts file from the user's project\r\n * This is used by the middleware to dynamically generate llms.txt\r\n */\r\nexport async function loadOntoConfig(): Promise<OntoConfig | null> {\r\n  try {\r\n    // Try to dynamically import the config file from the user's project root\r\n    // This runs in the middleware context, so we look in the project root\r\n    const config = await import(process.cwd() + '/onto.config');\r\n    return config.default || config;\r\n  } catch (error) {\r\n    // Config file doesn't exist or failed to load\r\n    return null;\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"],"mappings":"yaAAA,IAAAA,EAAA,GAAAC,EAAAD,EAAA,qBAAAE,EAAA,mBAAAC,IAAA,eAAAC,EAAAJ,GAoFA,eAAsBG,GAA6C,CACjE,GAAI,CAGF,IAAME,EAAS,MAAM,OAAO,QAAQ,IAAI,EAAI,gBAC5C,OAAOA,EAAO,SAAWA,CAC3B,MAAgB,CAEd,OAAO,IACT,CACF,CASO,SAASH,EAAgBG,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","names":["config_exports","__export","generateLlmsTxt","loadOntoConfig","__toCommonJS","config","lines","route","fullUrl","link","section"]}

package/dist/config.mjs

@@ -0,0 +1,4 @@
+async function r(){try{let n=await import(process.cwd()+"/onto.config");return n.default||n}catch{return null}}function i(n){let t=[];if(t.push(`# ${n.name}`),t.push(""),t.push(`> ${n.summary}`),t.push(""),n.routes&&n.routes.length>0){t.push("## Key Routes"),t.push("");for(let s of n.routes){let e=`${n.baseUrl}${s.path}`;t.push(`- [${s.path}](${e}): ${s.description}`)}t.push("")}if(n.externalLinks&&n.externalLinks.length>0){t.push("## Resources"),t.push("");for(let s of n.externalLinks)s.description?t.push(`- [${s.title}](${s.url}): ${s.description}`):t.push(`- [${s.title}](${s.url})`);t.push("")}if(n.sections&&n.sections.length>0)for(let s of n.sections)t.push(`## ${s.heading}`),t.push(""),t.push(s.content),t.push("");return t.join(`
+`).trim()+`
+`}export{i as generateLlmsTxt,r as loadOntoConfig};
+//# sourceMappingURL=config.mjs.map

package/dist/config.mjs.map

@@ -0,0 +1 @@
+{"version":3,"sources":["../src/config.ts"],"sourcesContent":["/**\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 * Load the onto.config.ts file from the user's project\r\n * This is used by the middleware to dynamically generate llms.txt\r\n */\r\nexport async function loadOntoConfig(): Promise<OntoConfig | null> {\r\n  try {\r\n    // Try to dynamically import the config file from the user's project root\r\n    // This runs in the middleware context, so we look in the project root\r\n    const config = await import(process.cwd() + '/onto.config');\r\n    return config.default || config;\r\n  } catch (error) {\r\n    // Config file doesn't exist or failed to load\r\n    return null;\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"],"mappings":"AAoFA,eAAsBA,GAA6C,CACjE,GAAI,CAGF,IAAMC,EAAS,MAAM,OAAO,QAAQ,IAAI,EAAI,gBAC5C,OAAOA,EAAO,SAAWA,CAC3B,MAAgB,CAEd,OAAO,IACT,CACF,CASO,SAASC,EAAgBD,EAA4B,CAC1D,IAAME,EAAkB,CAAC,EAWzB,GARAA,EAAM,KAAK,KAAKF,EAAO,IAAI,EAAE,EAC7BE,EAAM,KAAK,EAAE,EAGbA,EAAM,KAAK,KAAKF,EAAO,OAAO,EAAE,EAChCE,EAAM,KAAK,EAAE,EAGTF,EAAO,QAAUA,EAAO,OAAO,OAAS,EAAG,CAC7CE,EAAM,KAAK,eAAe,EAC1BA,EAAM,KAAK,EAAE,EACb,QAAWC,KAASH,EAAO,OAAQ,CACjC,IAAMI,EAAU,GAAGJ,EAAO,OAAO,GAAGG,EAAM,IAAI,GAC9CD,EAAM,KAAK,MAAMC,EAAM,IAAI,KAAKC,CAAO,MAAMD,EAAM,WAAW,EAAE,CAClE,CACAD,EAAM,KAAK,EAAE,CACf,CAGA,GAAIF,EAAO,eAAiBA,EAAO,cAAc,OAAS,EAAG,CAC3DE,EAAM,KAAK,cAAc,EACzBA,EAAM,KAAK,EAAE,EACb,QAAWG,KAAQL,EAAO,cACpBK,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,GAAIF,EAAO,UAAYA,EAAO,SAAS,OAAS,EAC9C,QAAWM,KAAWN,EAAO,SAC3BE,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","names":["loadOntoConfig","config","generateLlmsTxt","lines","route","fullUrl","link","section"]}

package/dist/index.d.mts

@@ -1,3 +1,6 @@
+export { OntoConfig, OntoConfig as OntoConfigType, OntoRoute, OntoRoute as OntoRouteType, PageType, generateLlmsTxt, loadOntoConfig } from './config.mjs';
+export { AIOMethodologySchema, AboutPageSchema, OrganizationSchema, generateAIOMethodologySchema, generateAboutPageSchema, generateOrganizationSchema, generateSchemaForPageType, serializeSchema } from './schemas.mjs';
+
 interface ExtractionResult {
     markdown: string;
     metadata: {

package/dist/index.d.ts

@@ -1,3 +1,6 @@
+export { OntoConfig, OntoConfig as OntoConfigType, OntoRoute, OntoRoute as OntoRouteType, PageType, generateLlmsTxt, loadOntoConfig } from './config.js';
+export { AIOMethodologySchema, AboutPageSchema, OrganizationSchema, generateAIOMethodologySchema, generateAboutPageSchema, generateOrganizationSchema, generateSchemaForPageType, serializeSchema } from './schemas.js';
+
 interface ExtractionResult {
     markdown: string;
     metadata: {

package/dist/index.js

@@ -1,9 +1,11 @@
-"use strict";var $=Object.create;var c=Object.defineProperty;var j=Object.getOwnPropertyDescriptor;var v=Object.getOwnPropertyNames;var O=Object.getPrototypeOf,R=Object.prototype.hasOwnProperty;var L=(t,n)=>{for(var o in n)c(t,o,{get:n[o],enumerable:!0})},u=(t,n,o,e)=>{if(n&&typeof n=="object"||typeof n=="function")for(let r of v(n))!R.call(t,r)&&r!==o&&c(t,r,{get:()=>n[r],enumerable:!(e=j(n,r))||e.enumerable});return t};var f=(t,n,o)=>(o=t!=null?$(O(t)):{},u(n||!t||!t.__esModule?c(o,"default",{value:t,enumerable:!0}):o,t)),z=t=>u(c({},"__esModule",{value:!0}),t);var E={};L(E,{extractContent:()=>g});module.exports=z(E);var h=f(require("cheerio")),p=f(require("turndown")),N=new p.default({headingStyle:"atx",codeBlockStyle:"fenced"});function g(t,n="Generated Output"){let o=t.length,e=h.load(t),r=e("title").text()||e("h1").first().text()||"Untitled Page",l=e('meta[name="description"]').attr("content")||"No description found.",a=[];e('script[type="application/ld+json"]').each((m,w)=>{try{let k=e(w).html()||"",x=JSON.parse(k);a.push(x)}catch{}}),e("script, style, noscript, iframe, svg, nav, footer, meta, link, header").remove();let s="";e("main").length>0?s=e("main").html()||"":e("article").length>0?s=e("article").html()||"":s=e("body").html()||"";let S=N.turndown(s),i=[`# ${r}`,`> ${l}`,"",`**Source:** ${n}`,`**Extracted:** ${new Date().toISOString()}`,"","---",""].join(`
-`)+S;a.length>0&&(i+=`
+"use strict";var C=Object.create;var c=Object.defineProperty;var L=Object.getOwnPropertyDescriptor;var H=Object.getOwnPropertyNames;var R=Object.getPrototypeOf,v=Object.prototype.hasOwnProperty;var I=(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 H(e))!v.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?C(R(t)):{},d(e||!t||!t.__esModule?c(n,"default",{value:t,enumerable:!0}):n,t)),M=t=>d(c({},"__esModule",{value:!0}),t);var E={};I(E,{extractContent:()=>x,generateAIOMethodologySchema:()=>l,generateAboutPageSchema:()=>p,generateLlmsTxt:()=>w,generateOrganizationSchema:()=>u,generateSchemaForPageType:()=>z,loadOntoConfig:()=>O,serializeSchema:()=>k});module.exports=M(E);var y=f(require("cheerio")),S=f(require("turndown")),j=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,b)=>{try{let A=o(b).html()||"",P=JSON.parse(A);a.push(P)}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 T=j.turndown(s),r=[`# ${i}`,`> ${g}`,"",`**Source:** ${e}`,`**Extracted:** ${new Date().toISOString()}`,"","---",""].join(`
+`)+T;a.length>0&&(r+=`
 
 ---
 ## Structured Data (JSON-LD)
 \`\`\`json
-`,a.forEach(m=>{i+=JSON.stringify(m,null,2)+`
-`}),i+="```\n");let d=i.length,y=o>0?(o-d)/o*100:0;return{markdown:i,metadata:{title:r,description:l,jsonLd:a},stats:{originalHtmlSize:o,markdownSize:d,tokenReductionRatio:y}}}0&&(module.exports={extractContent});
+`,a.forEach(h=>{r+=JSON.stringify(h,null,2)+`
+`}),r+="```\n");let m=r.length,$=n>0?(n-m)/n*100:0;return{markdown:r,metadata:{title:i,description:g,jsonLd:a},stats:{originalHtmlSize:n,markdownSize:m,tokenReductionRatio:$}}}async function O(){try{let t=await import(process.cwd()+"/onto.config");return t.default||t}catch{return null}}function w(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 l(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 u(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=u(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 z(t,e,n){switch(t){case"scoring":return l(e,n);case"about":return p(e,n);default:return null}}function k(t){return t?JSON.stringify(t,null,2):null}0&&(module.exports={extractContent,generateAIOMethodologySchema,generateAboutPageSchema,generateLlmsTxt,generateOrganizationSchema,generateSchemaForPageType,loadOntoConfig,serializeSchema});
 //# sourceMappingURL=index.js.map

package/dist/index.js.map

@@ -1 +1 @@
-{"version":3,"sources":["../src/index.ts","../src/extractor.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\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 ${totalFilesProcessed} semantic markdown endpoints.`);\r\n}\r\n"],"mappings":"0jBAAA,IAAAA,EAAA,GAAAC,EAAAD,EAAA,oBAAAE,IAAA,eAAAC,EAAAH,GCAA,IAAAI,EAAyB,sBACzBC,EAA4B,uBAEtBC,EAAkB,IAAI,EAAAC,QAAgB,CACxC,aAAc,MACd,eAAgB,QACpB,CAAC,EAwBM,SAASC,EAAeC,EAAcC,EAAoB,mBAAsC,CACnG,IAAMC,EAAeF,EAAK,OAEpBG,EAAY,OAAKH,CAAI,EAGrBI,EAAQD,EAAE,OAAO,EAAE,KAAK,GAAKA,EAAE,IAAI,EAAE,MAAM,EAAE,KAAK,GAAK,gBACvDE,EAAcF,EAAE,0BAA0B,EAAE,KAAK,SAAS,GAAK,wBAE/DG,EAAuB,CAAC,EAC9BH,EAAE,oCAAoC,EAAE,KAAK,CAACI,EAAGC,IAAO,CACpD,GAAI,CACA,IAAMC,EAAMN,EAAEK,CAAE,EAAE,KAAK,GAAK,GACtBE,EAAS,KAAK,MAAMD,CAAG,EAC7BH,EAAc,KAAKI,CAAM,CAC7B,MAAQ,CAER,CACJ,CAAC,EAGDP,EAAE,uEAAuE,EAAE,OAAO,EAOlF,IAAIQ,EAAc,GACdR,EAAE,MAAM,EAAE,OAAS,EACnBQ,EAAcR,EAAE,MAAM,EAAE,KAAK,GAAK,GAC3BA,EAAE,SAAS,EAAE,OAAS,EAC7BQ,EAAcR,EAAE,SAAS,EAAE,KAAK,GAAK,GAErCQ,EAAcR,EAAE,MAAM,EAAE,KAAK,GAAK,GAItC,IAAIS,EAAWf,EAAgB,SAASc,CAAW,EAc/CE,EAXgB,CAChB,KAAKT,CAAK,GACV,KAAKC,CAAW,GAChB,GACA,eAAeJ,CAAS,GACxB,kBAAkB,IAAI,KAAK,EAAE,YAAY,CAAC,GAC1C,GACA,MACA,EACJ,EAEgC,KAAK;AAAA,CAAI,EAAIW,EAGzCN,EAAc,OAAS,IACvBO,GAAiB;AAAA;AAAA;AAAA;AAAA;AAAA,EACjBP,EAAc,QAAQQ,GAAK,CACvBD,GAAiB,KAAK,UAAUC,EAAG,KAAM,CAAC,EAAI;AAAA,CAClD,CAAC,EACDD,GAAiB,SAGrB,IAAME,EAAeF,EAAc,OAC7BG,EAAsBd,EAAe,GAAMA,EAAea,GAAgBb,EAAgB,IAAM,EAEtG,MAAO,CACH,SAAUW,EACV,SAAU,CACN,MAAAT,EACA,YAAAC,EACA,OAAQC,CACZ,EACA,MAAO,CACH,iBAAkBJ,EAClB,aAAAa,EACA,oBAAAC,CACJ,CACJ,CACJ","names":["index_exports","__export","extractContent","__toCommonJS","cheerio","import_turndown","turndownService","TurndownService","extractContent","html","sourceUrl","originalSize","$","title","description","jsonLdScripts","_","el","raw","parsed","contentHtml","markdown","finalMarkdown","j","markdownSize","tokenReductionRatio"]}
+{"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, loadOntoConfig, 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 * Load the onto.config.ts file from the user's project\r\n * This is used by the middleware to dynamically generate llms.txt\r\n */\r\nexport async function loadOntoConfig(): Promise<OntoConfig | null> {\r\n  try {\r\n    // Try to dynamically import the config file from the user's project root\r\n    // This runs in the middleware context, so we look in the project root\r\n    const config = await import(process.cwd() + '/onto.config');\r\n    return config.default || config;\r\n  } catch (error) {\r\n    // Config file doesn't exist or failed to load\r\n    return null;\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,mBAAAC,EAAA,oBAAAC,IAAA,eAAAC,EAAAV,GCAA,IAAAW,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,CCzBA,eAAsBC,GAA6C,CACjE,GAAI,CAGF,IAAMC,EAAS,MAAM,OAAO,QAAQ,IAAI,EAAI,gBAC5C,OAAOA,EAAO,SAAWA,CAC3B,MAAgB,CAEd,OAAO,IACT,CACF,CASO,SAASC,EAAgBD,EAA4B,CAC1D,IAAME,EAAkB,CAAC,EAWzB,GARAA,EAAM,KAAK,KAAKF,EAAO,IAAI,EAAE,EAC7BE,EAAM,KAAK,EAAE,EAGbA,EAAM,KAAK,KAAKF,EAAO,OAAO,EAAE,EAChCE,EAAM,KAAK,EAAE,EAGTF,EAAO,QAAUA,EAAO,OAAO,OAAS,EAAG,CAC7CE,EAAM,KAAK,eAAe,EAC1BA,EAAM,KAAK,EAAE,EACb,QAAWC,KAASH,EAAO,OAAQ,CACjC,IAAMI,EAAU,GAAGJ,EAAO,OAAO,GAAGG,EAAM,IAAI,GAC9CD,EAAM,KAAK,MAAMC,EAAM,IAAI,KAAKC,CAAO,MAAMD,EAAM,WAAW,EAAE,CAClE,CACAD,EAAM,KAAK,EAAE,CACf,CAGA,GAAIF,EAAO,eAAiBA,EAAO,cAAc,OAAS,EAAG,CAC3DE,EAAM,KAAK,cAAc,EACzBA,EAAM,KAAK,EAAE,EACb,QAAWG,KAAQL,EAAO,cACpBK,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,GAAIF,EAAO,UAAYA,EAAO,SAAS,OAAS,EAC9C,QAAWM,KAAWN,EAAO,SAC3BE,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,CCtHO,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","loadOntoConfig","serializeSchema","__toCommonJS","cheerio","import_turndown","turndownService","TurndownService","extractContent","html","sourceUrl","originalSize","$","title","description","jsonLdScripts","_","el","raw","parsed","contentHtml","markdown","finalMarkdown","j","markdownSize","tokenReductionRatio","loadOntoConfig","config","generateLlmsTxt","lines","route","fullUrl","link","section","generateAIOMethodologySchema","config","pageUrl","generateOrganizationSchema","schema","generateAboutPageSchema","orgSchema","generateSchemaForPageType","pageType","serializeSchema"]}

package/dist/index.mjs

@@ -1,9 +1,11 @@
-import*as d from"cheerio";import S from"turndown";var y=new S({headingStyle:"atx",codeBlockStyle:"fenced"});function w(i,m="Generated Output"){let n=i.length,t=d.load(i),a=t("title").text()||t("h1").first().text()||"Untitled Page",s=t('meta[name="description"]').attr("content")||"No description found.",o=[];t('script[type="application/ld+json"]').each((l,h)=>{try{let p=t(h).html()||"",g=JSON.parse(p);o.push(g)}catch{}}),t("script, style, noscript, iframe, svg, nav, footer, meta, link, header").remove();let r="";t("main").length>0?r=t("main").html()||"":t("article").length>0?r=t("article").html()||"":r=t("body").html()||"";let u=y.turndown(r),e=[`# ${a}`,`> ${s}`,"",`**Source:** ${m}`,`**Extracted:** ${new Date().toISOString()}`,"","---",""].join(`
-`)+u;o.length>0&&(e+=`
+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((u,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+=`
 
 ---
 ## Structured Data (JSON-LD)
 \`\`\`json
-`,o.forEach(l=>{e+=JSON.stringify(l,null,2)+`
-`}),e+="```\n");let c=e.length,f=n>0?(n-c)/n*100:0;return{markdown:e,metadata:{title:a,description:s,jsonLd:o},stats:{originalHtmlSize:n,markdownSize:c,tokenReductionRatio:f}}}export{w as extractContent};
+`,r.forEach(u=>{i+=JSON.stringify(u,null,2)+`
+`}),i+="```\n");let l=i.length,f=e>0?(e-l)/e*100:0;return{markdown:i,metadata:{title:s,description:c,jsonLd:r},stats:{originalHtmlSize:e,markdownSize:l,tokenReductionRatio:f}}}async function k(){try{let t=await import(process.cwd()+"/onto.config");return t.default||t}catch{return null}}function T(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(`
+`).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 $(t,n,e){switch(t){case"scoring":return g(n,e);case"about":return h(n,e);default:return null}}function b(t){return t?JSON.stringify(t,null,2):null}export{z as extractContent,g as generateAIOMethodologySchema,h as generateAboutPageSchema,T as generateLlmsTxt,m as generateOrganizationSchema,$ as generateSchemaForPageType,k as loadOntoConfig,b as serializeSchema};
 //# sourceMappingURL=index.mjs.map

package/dist/index.mjs.map

@@ -1 +1 @@
-{"version":3,"sources":["../src/extractor.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 ${totalFilesProcessed} semantic markdown endpoints.`);\r\n}\r\n"],"mappings":"AAAA,UAAYA,MAAa,UACzB,OAAOC,MAAqB,WAE5B,IAAMC,EAAkB,IAAID,EAAgB,CACxC,aAAc,MACd,eAAgB,QACpB,CAAC,EAwBM,SAASE,EAAeC,EAAcC,EAAoB,mBAAsC,CACnG,IAAMC,EAAeF,EAAK,OAEpBG,EAAY,OAAKH,CAAI,EAGrBI,EAAQD,EAAE,OAAO,EAAE,KAAK,GAAKA,EAAE,IAAI,EAAE,MAAM,EAAE,KAAK,GAAK,gBACvDE,EAAcF,EAAE,0BAA0B,EAAE,KAAK,SAAS,GAAK,wBAE/DG,EAAuB,CAAC,EAC9BH,EAAE,oCAAoC,EAAE,KAAK,CAACI,EAAGC,IAAO,CACpD,GAAI,CACA,IAAMC,EAAMN,EAAEK,CAAE,EAAE,KAAK,GAAK,GACtBE,EAAS,KAAK,MAAMD,CAAG,EAC7BH,EAAc,KAAKI,CAAM,CAC7B,MAAQ,CAER,CACJ,CAAC,EAGDP,EAAE,uEAAuE,EAAE,OAAO,EAOlF,IAAIQ,EAAc,GACdR,EAAE,MAAM,EAAE,OAAS,EACnBQ,EAAcR,EAAE,MAAM,EAAE,KAAK,GAAK,GAC3BA,EAAE,SAAS,EAAE,OAAS,EAC7BQ,EAAcR,EAAE,SAAS,EAAE,KAAK,GAAK,GAErCQ,EAAcR,EAAE,MAAM,EAAE,KAAK,GAAK,GAItC,IAAIS,EAAWd,EAAgB,SAASa,CAAW,EAc/CE,EAXgB,CAChB,KAAKT,CAAK,GACV,KAAKC,CAAW,GAChB,GACA,eAAeJ,CAAS,GACxB,kBAAkB,IAAI,KAAK,EAAE,YAAY,CAAC,GAC1C,GACA,MACA,EACJ,EAEgC,KAAK;AAAA,CAAI,EAAIW,EAGzCN,EAAc,OAAS,IACvBO,GAAiB;AAAA;AAAA;AAAA;AAAA;AAAA,EACjBP,EAAc,QAAQQ,GAAK,CACvBD,GAAiB,KAAK,UAAUC,EAAG,KAAM,CAAC,EAAI;AAAA,CAClD,CAAC,EACDD,GAAiB,SAGrB,IAAME,EAAeF,EAAc,OAC7BG,EAAsBd,EAAe,GAAMA,EAAea,GAAgBb,EAAgB,IAAM,EAEtG,MAAO,CACH,SAAUW,EACV,SAAU,CACN,MAAAT,EACA,YAAAC,EACA,OAAQC,CACZ,EACA,MAAO,CACH,iBAAkBJ,EAClB,aAAAa,EACA,oBAAAC,CACJ,CACJ,CACJ","names":["cheerio","TurndownService","turndownService","extractContent","html","sourceUrl","originalSize","$","title","description","jsonLdScripts","_","el","raw","parsed","contentHtml","markdown","finalMarkdown","j","markdownSize","tokenReductionRatio"]}
+{"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 * Load the onto.config.ts file from the user's project\r\n * This is used by the middleware to dynamically generate llms.txt\r\n */\r\nexport async function loadOntoConfig(): Promise<OntoConfig | null> {\r\n  try {\r\n    // Try to dynamically import the config file from the user's project root\r\n    // This runs in the middleware context, so we look in the project root\r\n    const config = await import(process.cwd() + '/onto.config');\r\n    return config.default || config;\r\n  } catch (error) {\r\n    // Config file doesn't exist or failed to load\r\n    return null;\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,CCzBA,eAAsBC,GAA6C,CACjE,GAAI,CAGF,IAAMC,EAAS,MAAM,OAAO,QAAQ,IAAI,EAAI,gBAC5C,OAAOA,EAAO,SAAWA,CAC3B,MAAgB,CAEd,OAAO,IACT,CACF,CASO,SAASC,EAAgBD,EAA4B,CAC1D,IAAME,EAAkB,CAAC,EAWzB,GARAA,EAAM,KAAK,KAAKF,EAAO,IAAI,EAAE,EAC7BE,EAAM,KAAK,EAAE,EAGbA,EAAM,KAAK,KAAKF,EAAO,OAAO,EAAE,EAChCE,EAAM,KAAK,EAAE,EAGTF,EAAO,QAAUA,EAAO,OAAO,OAAS,EAAG,CAC7CE,EAAM,KAAK,eAAe,EAC1BA,EAAM,KAAK,EAAE,EACb,QAAWC,KAASH,EAAO,OAAQ,CACjC,IAAMI,EAAU,GAAGJ,EAAO,OAAO,GAAGG,EAAM,IAAI,GAC9CD,EAAM,KAAK,MAAMC,EAAM,IAAI,KAAKC,CAAO,MAAMD,EAAM,WAAW,EAAE,CAClE,CACAD,EAAM,KAAK,EAAE,CACf,CAGA,GAAIF,EAAO,eAAiBA,EAAO,cAAc,OAAS,EAAG,CAC3DE,EAAM,KAAK,cAAc,EACzBA,EAAM,KAAK,EAAE,EACb,QAAWG,KAAQL,EAAO,cACpBK,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,GAAIF,EAAO,UAAYA,EAAO,SAAS,OAAS,EAC9C,QAAWM,KAAWN,EAAO,SAC3BE,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,CCtHO,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","loadOntoConfig","config","generateLlmsTxt","lines","route","fullUrl","link","section","generateAIOMethodologySchema","config","pageUrl","generateOrganizationSchema","schema","generateAboutPageSchema","orgSchema","generateSchemaForPageType","pageType","serializeSchema"]}

package/dist/middleware.d.mts

@@ -1,4 +1,5 @@
 import { NextRequest, NextResponse } from 'next/server';
+import { OntoConfig } from './config.mjs';
 
 /**
  * Comprehensive registry of AI bot user-agent strings.
@@ -18,8 +19,8 @@ declare const AI_BOT_USER_AGENTS: string[];
  * Given a raw user-agent string, returns the matched AiBot entry or undefined.
  * Comparison is case-insensitive to handle inconsistent agent casing.
  */
-declare function matchBot(userAgent: string): AiBot | undefined;
+declare function matchBot(userAgent: string | null): AiBot | undefined;
 
-declare function ontoMiddleware(request: NextRequest): Promise<NextResponse<unknown>>;
+declare function ontoMiddleware(request: NextRequest, config?: OntoConfig): Promise<NextResponse<unknown>>;
 
 export { AI_BOT_USER_AGENTS, type AiBot, matchBot, ontoMiddleware };

package/dist/middleware.d.ts

@@ -1,4 +1,5 @@
 import { NextRequest, NextResponse } from 'next/server';
+import { OntoConfig } from './config.js';
 
 /**
  * Comprehensive registry of AI bot user-agent strings.
@@ -18,8 +19,8 @@ declare const AI_BOT_USER_AGENTS: string[];
  * Given a raw user-agent string, returns the matched AiBot entry or undefined.
  * Comparison is case-insensitive to handle inconsistent agent casing.
  */
-declare function matchBot(userAgent: string): AiBot | undefined;
+declare function matchBot(userAgent: string | null): AiBot | undefined;
 
-declare function ontoMiddleware(request: NextRequest): Promise<NextResponse<unknown>>;
+declare function ontoMiddleware(request: NextRequest, config?: OntoConfig): Promise<NextResponse<unknown>>;
 
 export { AI_BOT_USER_AGENTS, type AiBot, matchBot, ontoMiddleware };

package/dist/middleware.js

@@ -1,6 +1,8 @@
-"use strict";var l=Object.defineProperty;var _=Object.getOwnPropertyDescriptor;var w=Object.getOwnPropertyNames;var C=Object.prototype.hasOwnProperty;var T=(e,t)=>{for(var a in t)l(e,a,{get:t[a],enumerable:!0})},R=(e,t,a,n)=>{if(t&&typeof t=="object"||typeof t=="function")for(let c of w(t))!C.call(e,c)&&c!==a&&l(e,c,{get:()=>t[c],enumerable:!(n=_(t,c))||n.enumerable});return e};var k=e=>R(l({},"__esModule",{value:!0}),e);var S={};T(S,{AI_BOT_USER_AGENTS:()=>f,matchBot:()=>s,ontoMiddleware:()=>G});module.exports=k(S);var m=require("next/server");var A=[{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"}],f=A.map(e=>e.name);function s(e){let t=e.toLowerCase();return A.find(a=>t.includes(a.name.toLowerCase()))}async function G(e){let t=e.headers.get("user-agent")||"",a=e.headers.get("accept")||"",n=s(t),c=!!n,g=a.includes("text/markdown");if(c||g){let r=e.nextUrl.clone();if(r.pathname.startsWith("/_next")||r.pathname.includes("."))return m.NextResponse.next();let o=r.pathname;(o==="/"||o==="")&&(o="/index"),o.endsWith("/")&&o!=="/"&&(o=o.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"};n&&(d["X-Onto-Bot"]=`${n.name} (${n.company})`);let p=process.env.ONTO_API_KEY,y=process.env.ONTO_DASHBOARD_URL||"https://app.buildonto.dev";if(p){fetch(`${y}/api/track`,{method:"POST",headers:{"x-onto-key":p,"Content-Type":"application/json"},body:JSON.stringify({route:r.pathname,userAgent:t,bot:n?n.name:null,company:n?n.company:null})}).catch(()=>{});try{let i=await fetch(`${y}/api/sdk/inject?route=${r.pathname}`,{headers:{"x-onto-key":p},signal:AbortSignal.timeout(1500)});if(i.ok){let{injection:u}=await i.json();if(u){let B=`${r.origin}/.onto${o}.md`,x=await fetch(B);if(x.ok){let O=`${await x.text()}
+"use strict";var u=Object.defineProperty;var T=Object.getOwnPropertyDescriptor;var k=Object.getOwnPropertyNames;var b=Object.prototype.hasOwnProperty;var R=(e,t)=>{for(var o in t)u(e,o,{get:t[o],enumerable:!0})},_=(e,t,o,a)=>{if(t&&typeof t=="object"||typeof t=="function")for(let n of k(t))!b.call(e,n)&&n!==o&&u(e,n,{get:()=>t[n],enumerable:!(a=T(t,n))||a.enumerable});return e};var P=e=>_(u({},"__esModule",{value:!0}),e);var G={};R(G,{AI_BOT_USER_AGENTS:()=>x,matchBot:()=>m,ontoMiddleware:()=>U});module.exports=P(G);var c=require("next/server");var f=[{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"}],x=f.map(e=>e.name);function m(e){if(!e)return;let t=e.toLowerCase();return f.find(o=>t.includes(o.name.toLowerCase()))}async function O(){try{let e=await import(process.cwd()+"/onto.config");return e.default||e}catch{return null}}function C(e){let t=[];if(t.push(`# ${e.name}`),t.push(""),t.push(`> ${e.summary}`),t.push(""),e.routes&&e.routes.length>0){t.push("## Key Routes"),t.push("");for(let o of e.routes){let a=`${e.baseUrl}${o.path}`;t.push(`- [${o.path}](${a}): ${o.description}`)}t.push("")}if(e.externalLinks&&e.externalLinks.length>0){t.push("## Resources"),t.push("");for(let o of e.externalLinks)o.description?t.push(`- [${o.title}](${o.url}): ${o.description}`):t.push(`- [${o.title}](${o.url})`);t.push("")}if(e.sections&&e.sections.length>0)for(let o of e.sections)t.push(`## ${o.heading}`),t.push(""),t.push(o.content),t.push("");return t.join(`
+`).trim()+`
+`}async function U(e,t){let o=e.headers.get("user-agent"),a=e.nextUrl.clone(),n=m(o),$=e.headers.get("accept")||"",A=!!n,B=$.includes("text/markdown");if(A||B){if(a.pathname.startsWith("/_next"))return c.NextResponse.next();if(a.pathname==="/llms.txt")try{let i=t||await O();if(i){let r=C(i),p=new c.NextResponse(r,{headers:{"Content-Type":"text/plain; charset=utf-8","Cache-Control":"public, max-age=3600, s-maxage=3600, stale-while-revalidate=86400"}});return n&&p.headers.set("X-Onto-Bot",`${n.name} (${n.company})`),p}else{a.pathname="/llms.txt";let r=c.NextResponse.rewrite(a);return r.headers.set("Content-Type","text/plain; charset=utf-8"),r.headers.set("Cache-Control","public, max-age=3600, s-maxage=3600, stale-while-revalidate=86400"),n&&r.headers.set("X-Onto-Bot",`${n.name} (${n.company})`),r}}catch(i){console.error("[Onto] Failed to generate llms.txt:",i),a.pathname="/llms.txt";let r=c.NextResponse.rewrite(a);return r.headers.set("Content-Type","text/plain; charset=utf-8"),n&&r.headers.set("X-Onto-Bot",`${n.name} (${n.company})`),r}if(a.pathname.includes("."))return c.NextResponse.next();let s=a.pathname;(s==="/"||s==="")&&(s="/index"),s.endsWith("/")&&s!=="/"&&(s=s.slice(0,-1));let h={"Content-Type":"text/markdown; charset=utf-8","Cache-Control":"public, max-age=3600, s-maxage=3600, stale-while-revalidate=86400"};n&&(h["X-Onto-Bot"]=`${n.name} (${n.company})`);let l=process.env.ONTO_API_KEY,d=process.env.ONTO_DASHBOARD_URL||"https://app.buildonto.dev";if(l){fetch(`${d}/api/track`,{method:"POST",headers:{"x-onto-key":l,"Content-Type":"application/json"},body:JSON.stringify({route:a.pathname,userAgent:o,bot:n?n.name:null,company:n?n.company:null})}).catch(()=>{});try{let i=await fetch(`${d}/api/sdk/inject?route=${a.pathname}`,{headers:{"x-onto-key":l},signal:AbortSignal.timeout(1500)});if(i.ok){let{injection:r}=await i.json();if(r){let p=`${a.origin}/.onto${s}.md`,y=await fetch(p);if(y.ok){let w=`${await y.text()}
 
 ---
 
-${u}`;return new m.NextResponse(O,{headers:{...d,"X-Onto-Injected":"true"}})}}}}catch(i){console.error("[Onto] Injection failed",i)}}r.pathname=`/.onto${o}.md`;let h=m.NextResponse.rewrite(r);return n&&h.headers.set("X-Onto-Bot",`${n.name} (${n.company})`),h}return m.NextResponse.next()}0&&(module.exports={AI_BOT_USER_AGENTS,matchBot,ontoMiddleware});
+${r}`;return new c.NextResponse(w,{headers:{...h,"X-Onto-Injected":"true"}})}}}}catch(i){console.error("[Onto] Injection failed",i)}}a.pathname=`/.onto${s}.md`;let g=c.NextResponse.rewrite(a);return n&&g.headers.set("X-Onto-Bot",`${n.name} (${n.company})`),g}return c.NextResponse.next()}0&&(module.exports={AI_BOT_USER_AGENTS,matchBot,ontoMiddleware});
 //# sourceMappingURL=middleware.js.map