@ontosdk/next 1.3.1 → 1.3.2

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Files changed (34) hide show
  1. package/dist/cli.js +8 -6
  2. package/dist/cli.js.map +1 -1
  3. package/dist/cli.mjs +8 -6
  4. package/dist/cli.mjs.map +1 -1
  5. package/dist/config.js +2 -2
  6. package/dist/config.js.map +1 -1
  7. package/dist/config.mjs +2 -2
  8. package/dist/config.mjs.map +1 -1
  9. package/dist/index.js +4 -4
  10. package/dist/index.js.map +1 -1
  11. package/dist/index.mjs +2 -2
  12. package/dist/index.mjs.map +1 -1
  13. package/dist/middleware.js +3 -3
  14. package/dist/middleware.js.map +1 -1
  15. package/dist/middleware.mjs +3 -3
  16. package/dist/middleware.mjs.map +1 -1
  17. package/package.json +1 -1
  18. package/src/cli.ts +34 -0
  19. package/src/config.ts +0 -16
  20. package/src/middleware.ts +3 -6
  21. package/dist/OntoHead.d.mts +0 -27
  22. package/dist/OntoHead.d.ts +0 -27
  23. package/dist/OntoProvider.d.mts +0 -52
  24. package/dist/OntoProvider.d.ts +0 -52
  25. package/dist/cli.d.mts +0 -1
  26. package/dist/cli.d.ts +0 -1
  27. package/dist/config.d.mts +0 -85
  28. package/dist/config.d.ts +0 -85
  29. package/dist/index.d.mts +0 -25
  30. package/dist/index.d.ts +0 -25
  31. package/dist/middleware.d.mts +0 -26
  32. package/dist/middleware.d.ts +0 -26
  33. package/dist/schemas.d.mts +0 -73
  34. package/dist/schemas.d.ts +0 -73
package/dist/cli.js CHANGED
@@ -1,12 +1,14 @@
1
1
  #!/usr/bin/env node
2
- "use strict";var F=Object.create;var O=Object.defineProperty;var z=Object.getOwnPropertyDescriptor;var N=Object.getOwnPropertyNames;var R=Object.getPrototypeOf,C=Object.prototype.hasOwnProperty;var E=(e,i,n,t)=>{if(i&&typeof i=="object"||typeof i=="function")for(let s of N(i))!C.call(e,s)&&s!==n&&O(e,s,{get:()=>i[s],enumerable:!(t=z(i,s))||t.enumerable});return e};var S=(e,i,n)=>(n=e!=null?F(R(e)):{},E(i||!e||!e.__esModule?O(n,"default",{value:e,enumerable:!0}):n,e));var j=require("glob"),a=S(require("fs")),g=S(require("path")),o=S(require("picocolors"));var k=S(require("cheerio")),v=S(require("turndown")),L=new v.default({headingStyle:"atx",codeBlockStyle:"fenced"});function P(e,i="Generated Output"){let n=e.length,t=k.load(e),s=t("title").text()||t("h1").first().text()||"Untitled Page",m=t('meta[name="description"]').attr("content")||"No description found.",u=[];t('script[type="application/ld+json"]').each((f,d)=>{try{let $=t(d).html()||"",x=JSON.parse($);u.push(x)}catch{}}),t("script, style, noscript, iframe, svg, nav, footer, meta, link, header").remove();let h="";t("main").length>0?h=t("main").html()||"":t("article").length>0?h=t("article").html()||"":h=t("body").html()||"";let w=L.turndown(h),r=[`# ${s}`,`> ${m}`,"",`**Source:** ${i}`,`**Extracted:** ${new Date().toISOString()}`,"","---",""].join(`
3
- `)+w;u.length>0&&(r+=`
2
+ "use strict";var F=Object.create;var O=Object.defineProperty;var L=Object.getOwnPropertyDescriptor;var T=Object.getOwnPropertyNames;var z=Object.getPrototypeOf,N=Object.prototype.hasOwnProperty;var E=(e,t,n,o)=>{if(t&&typeof t=="object"||typeof t=="function")for(let r of T(t))!N.call(e,r)&&r!==n&&O(e,r,{get:()=>t[r],enumerable:!(o=L(t,r))||o.enumerable});return e};var S=(e,t,n)=>(n=e!=null?F(z(e)):{},E(t||!e||!e.__esModule?O(n,"default",{value:e,enumerable:!0}):n,e));var b=require("glob"),a=S(require("fs")),l=S(require("path")),s=S(require("picocolors"));var k=S(require("cheerio")),P=S(require("turndown")),D=new P.default({headingStyle:"atx",codeBlockStyle:"fenced"});function j(e,t="Generated Output"){let n=e.length,o=k.load(e),r=o("title").text()||o("h1").first().text()||"Untitled Page",m=o('meta[name="description"]').attr("content")||"No description found.",f=[];o('script[type="application/ld+json"]').each((p,h)=>{try{let d=o(h).html()||"",x=JSON.parse(d);f.push(x)}catch{}}),o("script, style, noscript, iframe, svg, nav, footer, meta, link, header").remove();let g="";o("main").length>0?g=o("main").html()||"":o("article").length>0?g=o("article").html()||"":g=o("body").html()||"";let $=D.turndown(g),i=[`# ${r}`,`> ${m}`,"",`**Source:** ${t}`,`**Extracted:** ${new Date().toISOString()}`,"","---",""].join(`
3
+ `)+$;f.length>0&&(i+=`
4
4
 
5
5
  ---
6
6
  ## Structured Data (JSON-LD)
7
7
  \`\`\`json
8
- `,u.forEach(f=>{r+=JSON.stringify(f,null,2)+`
9
- `}),r+="```\n");let c=r.length,p=n>0?(n-c)/n*100:0;return{markdown:r,metadata:{title:s,description:m,jsonLd:u},stats:{originalHtmlSize:n,markdownSize:c,tokenReductionRatio:p}}}function T(){let e=g.default.join(process.cwd(),".env.local");a.default.existsSync(e)&&a.default.readFileSync(e,"utf8").split(/\r?\n/).forEach(n=>{let t=n.trim();if(!t||t.startsWith("#"))return;let[s,...m]=t.split("=");s&&m.length>0&&(process.env[s.trim()]=m.join("=").trim().replace(/^["']|["']$/g,""))})}async function H(){T(),console.log(o.default.cyan(`
10
- [Onto] Starting Semantic Output Generation...`));let e=process.cwd(),i=g.default.join(e,".next/server/app"),n=g.default.join(e,"public/.onto");if(!a.default.existsSync(i)){console.log(o.default.yellow(`[Onto] Could not find Next.js app output at ${i}`)),console.log(o.default.yellow('[Onto] Ensure this is run after "next build" and you are using the App Router.'));return}let t=await(0,j.glob)("**/*.html",{cwd:i});if(t.length===0){console.log(o.default.yellow("[Onto] No static HTML files found to process."));return}a.default.existsSync(n)||a.default.mkdirSync(n,{recursive:!0});let s=0,m=0,u=0;for(let l of t){let r=g.default.join(i,l),c=l.replace(/\.html$/,".md"),p=g.default.join(n,c);try{let f=a.default.readFileSync(r,"utf8"),d=P(f,`/${c.replace(/\.md$/,"")}`),$=g.default.dirname(p);a.default.existsSync($)||a.default.mkdirSync($,{recursive:!0}),a.default.writeFileSync(p,d.markdown,"utf8"),s+=d.stats.originalHtmlSize,m+=d.stats.markdownSize,u++;let x=(d.stats.originalHtmlSize/1024).toFixed(1),b=(d.stats.markdownSize/1024).toFixed(1),y=l.replace(/\.html$/,"");y==="index"?y="/":y=`/${y}`,console.log(o.default.green("\u2713 Optimized")+o.default.dim(` ${y} `)+o.default.blue(`[${x}KB -> ${b}KB]`))}catch(f){console.error(o.default.red(`\u2717 Failed to process ${l}: ${f.message}`))}}console.log(o.default.bold(o.default.magenta(`Processed ${u} pages. Total Size: ${(s/1024).toFixed(1)}KB -> ${(m/1024).toFixed(1)}KB`)));let h=process.env.ONTO_API_KEY,w=process.env.ONTO_DASHBOARD_URL||"https://app.buildonto.dev";if(h&&u>0){console.log(o.default.cyan(`[Onto] Syncing manifest with Control Plane [${w}]...`));try{let l=t.map(c=>{let p=c.replace(/\.html$/,""),f=p==="index"?"/":`/${p}`,d=g.default.join(n,c.replace(/\.html$/,".md"));return{route:f,filename:`${p}.md`,content:a.default.readFileSync(d,"utf8")}}),r=await fetch(`${w}/api/files`,{method:"POST",headers:{"x-onto-key":h,"Content-Type":"application/json"},body:JSON.stringify({files:l})});if(r.ok)console.log(o.default.green("\u2713 Control Plane sync successful"));else{let c=await r.json().catch(()=>({}));console.log(o.default.yellow(`\u26A0 Control Plane sync skipped: ${c.error||r.statusText}`))}}catch(l){console.log(o.default.yellow(`\u26A0 Control Plane sync failed: ${l.message}`))}}console.log(o.default.dim(`Edge payloads are ready at /public/.onto/*
11
- `))}H().catch(e=>{console.error(o.default.red(`[Onto] Fatal Error: ${e.message}`)),process.exit(1)});
8
+ `,f.forEach(p=>{i+=JSON.stringify(p,null,2)+`
9
+ `}),i+="```\n");let u=i.length,c=n>0?(n-u)/n*100:0;return{markdown:i,metadata:{title:r,description:m,jsonLd:f},stats:{originalHtmlSize:n,markdownSize:u,tokenReductionRatio:c}}}function v(e){let t=[];if(t.push(`# ${e.name}`),t.push(""),t.push(`> ${e.summary}`),t.push(""),e.routes&&e.routes.length>0){t.push("## Key Routes"),t.push("");for(let n of e.routes){let o=`${e.baseUrl}${n.path}`;t.push(`- [${n.path}](${o}): ${n.description}`)}t.push("")}if(e.externalLinks&&e.externalLinks.length>0){t.push("## Resources"),t.push("");for(let n of e.externalLinks)n.description?t.push(`- [${n.title}](${n.url}): ${n.description}`):t.push(`- [${n.title}](${n.url})`);t.push("")}if(e.sections&&e.sections.length>0)for(let n of e.sections)t.push(`## ${n.heading}`),t.push(""),t.push(n.content),t.push("");return t.join(`
10
+ `).trim()+`
11
+ `}async function K(){try{let t=await import("file://"+l.default.join(process.cwd(),"onto.config").replace(/\\/g,"/")+".ts");return t.default||t}catch{try{let n=await import("file://"+l.default.join(process.cwd(),"onto.config").replace(/\\/g,"/")+".js");return n.default||n}catch{return null}}}function H(){let e=l.default.join(process.cwd(),".env.local");a.default.existsSync(e)&&a.default.readFileSync(e,"utf8").split(/\r?\n/).forEach(n=>{let o=n.trim();if(!o||o.startsWith("#"))return;let[r,...m]=o.split("=");r&&m.length>0&&(process.env[r.trim()]=m.join("=").trim().replace(/^["']|["']$/g,""))})}async function _(){H(),console.log(s.default.cyan(`
12
+ [Onto] Starting Semantic Output Generation...`));let e=process.cwd(),t=l.default.join(e,".next/server/app"),n=l.default.join(e,"public/.onto");if(!a.default.existsSync(t)){console.log(s.default.yellow(`[Onto] Could not find Next.js app output at ${t}`)),console.log(s.default.yellow('[Onto] Ensure this is run after "next build" and you are using the App Router.'));return}let o=await(0,b.glob)("**/*.html",{cwd:t});if(o.length===0){console.log(s.default.yellow("[Onto] No static HTML files found to process."));return}a.default.existsSync(n)||a.default.mkdirSync(n,{recursive:!0});let r=0,m=0,f=0;for(let i of o){let u=l.default.join(t,i),c=i.replace(/\.html$/,".md"),p=l.default.join(n,c);try{let h=a.default.readFileSync(u,"utf8"),d=j(h,`/${c.replace(/\.md$/,"")}`),x=l.default.dirname(p);a.default.existsSync(x)||a.default.mkdirSync(x,{recursive:!0}),a.default.writeFileSync(p,d.markdown,"utf8"),r+=d.stats.originalHtmlSize,m+=d.stats.markdownSize,f++;let C=(d.stats.originalHtmlSize/1024).toFixed(1),R=(d.stats.markdownSize/1024).toFixed(1),y=i.replace(/\.html$/,"");y==="index"?y="/":y=`/${y}`,console.log(s.default.green("\u2713 Optimized")+s.default.dim(` ${y} `)+s.default.blue(`[${C}KB -> ${R}KB]`))}catch(h){console.error(s.default.red(`\u2717 Failed to process ${i}: ${h.message}`))}}console.log(s.default.bold(s.default.magenta(`Processed ${f} pages. Total Size: ${(r/1024).toFixed(1)}KB -> ${(m/1024).toFixed(1)}KB`)));let g=process.env.ONTO_API_KEY,$=process.env.ONTO_DASHBOARD_URL||"https://app.buildonto.dev";if(g&&f>0){console.log(s.default.cyan(`[Onto] Syncing manifest with Control Plane [${$}]...`));try{let i=o.map(c=>{let p=c.replace(/\.html$/,""),h=p==="index"?"/":`/${p}`,d=l.default.join(n,c.replace(/\.html$/,".md"));return{route:h,filename:`${p}.md`,content:a.default.readFileSync(d,"utf8")}}),u=await fetch(`${$}/api/files`,{method:"POST",headers:{"x-onto-key":g,"Content-Type":"application/json"},body:JSON.stringify({files:i})});if(u.ok)console.log(s.default.green("\u2713 Control Plane sync successful"));else{let c=await u.json().catch(()=>({}));console.log(s.default.yellow(`\u26A0 Control Plane sync skipped: ${c.error||u.statusText}`))}}catch(i){console.log(s.default.yellow(`\u26A0 Control Plane sync failed: ${i.message}`))}}let w=await K();if(w){let i=v(w),u=l.default.join(e,"public/llms.txt"),c=l.default.join(e,"public");a.default.existsSync(c)||a.default.mkdirSync(c,{recursive:!0}),a.default.writeFileSync(u,i,"utf8"),console.log(s.default.green("\u2713 Generated")+s.default.dim(" /llms.txt"))}console.log(s.default.dim(`Edge payloads are ready at /public/.onto/*
13
+ `))}_().catch(e=>{console.error(s.default.red(`[Onto] Fatal Error: ${e.message}`)),process.exit(1)});
12
14
  //# sourceMappingURL=cli.js.map
package/dist/cli.js.map CHANGED
@@ -1 +1 @@
1
- {"version":3,"sources":["../src/cli.ts","../src/extractor.ts"],"sourcesContent":["#!/usr/bin/env node\r\nimport { glob } from 'glob';\r\nimport fs from 'fs';\r\nimport path from 'path';\r\nimport pc from 'picocolors';\r\nimport { extractContent } from './extractor';\r\n\r\n// Simple helper to load .env.local from the current working directory\r\nfunction loadEnv() {\r\n const envPath = path.join(process.cwd(), '.env.local');\r\n if (fs.existsSync(envPath)) {\r\n const envContent = fs.readFileSync(envPath, 'utf8');\r\n envContent.split(/\\r?\\n/).forEach(line => {\r\n const trimmedLine = line.trim();\r\n if (!trimmedLine || trimmedLine.startsWith('#')) return;\r\n const [key, ...valueParts] = trimmedLine.split('=');\r\n if (key && valueParts.length > 0) {\r\n process.env[key.trim()] = valueParts.join('=').trim().replace(/^[\"']|[\"']$/g, '');\r\n }\r\n });\r\n }\r\n}\r\n\r\nasync function main() {\r\n loadEnv();\r\n console.log(pc.cyan('\\n[Onto] Starting Semantic Output Generation...'));\r\n\r\n const cwd = process.cwd();\r\n const nextAppDirDir = path.join(cwd, '.next/server/app');\r\n const ontoPublicDir = path.join(cwd, 'public/.onto');\r\n\r\n if (!fs.existsSync(nextAppDirDir)) {\r\n console.log(pc.yellow(`[Onto] Could not find Next.js app output at ${nextAppDirDir}`));\r\n console.log(pc.yellow(`[Onto] Ensure this is run after \"next build\" and you are using the App Router.`));\r\n return;\r\n }\r\n\r\n // Find all HTML files rendered by Next.js in the app directory\r\n const files = await glob('**/*.html', { cwd: nextAppDirDir });\r\n\r\n if (files.length === 0) {\r\n console.log(pc.yellow(`[Onto] No static HTML files found to process.`));\r\n return;\r\n }\r\n\r\n // Ensure output directory exists\r\n if (!fs.existsSync(ontoPublicDir)) {\r\n fs.mkdirSync(ontoPublicDir, { recursive: true });\r\n }\r\n\r\n let totalOriginalSize = 0;\r\n let totalMarkdownSize = 0;\r\n let totalFilesProcessed = 0;\r\n\r\n for (const file of files) {\r\n const inputPath = path.join(nextAppDirDir, file);\r\n\r\n // We map file path e.g. \"pricing.html\" to \"pricing.md\", or \"blog/post.html\" to \"blog/post.md\"\r\n let outputPathRelative = file.replace(/\\.html$/, '.md');\r\n // If it's a dynamic route page, or purely root index.html\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 const result = extractContent(htmlContent, `/${outputPathRelative.replace(/\\.md$/, '')}`);\r\n\r\n // Ensure specific sub-directory exists (e.g., for blog/post.md)\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\r\n totalOriginalSize += result.stats.originalHtmlSize;\r\n totalMarkdownSize += result.stats.markdownSize;\r\n totalFilesProcessed++;\r\n\r\n const origKb = (result.stats.originalHtmlSize / 1024).toFixed(1);\r\n const mdKb = (result.stats.markdownSize / 1024).toFixed(1);\r\n\r\n // /index.html -> /\r\n let routeName = file.replace(/\\.html$/, '');\r\n if (routeName === 'index') routeName = '/';\r\n else routeName = `/${routeName}`;\r\n\r\n console.log(\r\n pc.green(`✓ Optimized`) +\r\n pc.dim(` ${routeName} `) +\r\n pc.blue(`[${origKb}KB -> ${mdKb}KB]`)\r\n );\r\n } catch (e: any) {\r\n console.error(pc.red(`✗ Failed to process ${file}: ${e.message}`));\r\n }\r\n }\r\n\r\n console.log(\r\n pc.bold(\r\n pc.magenta(`Processed ${totalFilesProcessed} pages. Total Size: ${(totalOriginalSize / 1024).toFixed(1)}KB -> ${(totalMarkdownSize / 1024).toFixed(1)}KB`)\r\n )\r\n );\r\n\r\n // Sync with Onto Control Plane (Premium)\r\n const ONTO_API_KEY = process.env.ONTO_API_KEY;\r\n const DASHBOARD_URL = process.env.ONTO_DASHBOARD_URL || 'https://app.buildonto.dev';\r\n\r\n if (ONTO_API_KEY && totalFilesProcessed > 0) {\r\n console.log(pc.cyan(`[Onto] Syncing manifest with Control Plane [${DASHBOARD_URL}]...`));\r\n try {\r\n const manifest = files.map(file => {\r\n const routeName = file.replace(/\\.html$/, '');\r\n const route = routeName === 'index' ? '/' : `/${routeName}`;\r\n const mdPath = path.join(ontoPublicDir, file.replace(/\\.html$/, '.md'));\r\n return {\r\n route,\r\n filename: `${routeName}.md`,\r\n content: fs.readFileSync(mdPath, 'utf8')\r\n };\r\n });\r\n\r\n const res = await fetch(`${DASHBOARD_URL}/api/files`, {\r\n method: 'POST',\r\n headers: {\r\n 'x-onto-key': ONTO_API_KEY,\r\n 'Content-Type': 'application/json'\r\n },\r\n body: JSON.stringify({ files: manifest })\r\n });\r\n\r\n if (res.ok) {\r\n console.log(pc.green('✓ Control Plane sync successful'));\r\n } else {\r\n const errData = await res.json().catch(() => ({}));\r\n console.log(pc.yellow(`⚠ Control Plane sync skipped: ${errData.error || res.statusText}`));\r\n }\r\n } catch (e: any) {\r\n console.log(pc.yellow(`⚠ Control Plane sync failed: ${e.message}`));\r\n }\r\n }\r\n\r\n console.log(pc.dim(`Edge payloads are ready at /public/.onto/*\\n`));\r\n}\r\n\r\nmain().catch(e => {\r\n console.error(pc.red(`[Onto] Fatal Error: ${e.message}`));\r\n process.exit(1);\r\n});\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":";wdACA,IAAAA,EAAqB,gBACrBC,EAAe,iBACfC,EAAiB,mBACjBC,EAAe,yBCJf,IAAAC,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,CDrGA,SAASC,GAAU,CACf,IAAMC,EAAU,EAAAC,QAAK,KAAK,QAAQ,IAAI,EAAG,YAAY,EACjD,EAAAC,QAAG,WAAWF,CAAO,GACF,EAAAE,QAAG,aAAaF,EAAS,MAAM,EACvC,MAAM,OAAO,EAAE,QAAQG,GAAQ,CACtC,IAAMC,EAAcD,EAAK,KAAK,EAC9B,GAAI,CAACC,GAAeA,EAAY,WAAW,GAAG,EAAG,OACjD,GAAM,CAACC,EAAK,GAAGC,CAAU,EAAIF,EAAY,MAAM,GAAG,EAC9CC,GAAOC,EAAW,OAAS,IAC3B,QAAQ,IAAID,EAAI,KAAK,CAAC,EAAIC,EAAW,KAAK,GAAG,EAAE,KAAK,EAAE,QAAQ,eAAgB,EAAE,EAExF,CAAC,CAET,CAEA,eAAeC,GAAO,CAClBR,EAAQ,EACR,QAAQ,IAAI,EAAAS,QAAG,KAAK;AAAA,8CAAiD,CAAC,EAEtE,IAAMC,EAAM,QAAQ,IAAI,EAClBC,EAAgB,EAAAT,QAAK,KAAKQ,EAAK,kBAAkB,EACjDE,EAAgB,EAAAV,QAAK,KAAKQ,EAAK,cAAc,EAEnD,GAAI,CAAC,EAAAP,QAAG,WAAWQ,CAAa,EAAG,CAC/B,QAAQ,IAAI,EAAAF,QAAG,OAAO,+CAA+CE,CAAa,EAAE,CAAC,EACrF,QAAQ,IAAI,EAAAF,QAAG,OAAO,gFAAgF,CAAC,EACvG,MACJ,CAGA,IAAMI,EAAQ,QAAM,QAAK,YAAa,CAAE,IAAKF,CAAc,CAAC,EAE5D,GAAIE,EAAM,SAAW,EAAG,CACpB,QAAQ,IAAI,EAAAJ,QAAG,OAAO,+CAA+C,CAAC,EACtE,MACJ,CAGK,EAAAN,QAAG,WAAWS,CAAa,GAC5B,EAAAT,QAAG,UAAUS,EAAe,CAAE,UAAW,EAAK,CAAC,EAGnD,IAAIE,EAAoB,EACpBC,EAAoB,EACpBC,EAAsB,EAE1B,QAAWC,KAAQJ,EAAO,CACtB,IAAMK,EAAY,EAAAhB,QAAK,KAAKS,EAAeM,CAAI,EAG3CE,EAAqBF,EAAK,QAAQ,UAAW,KAAK,EAEhDG,EAAa,EAAAlB,QAAK,KAAKU,EAAeO,CAAkB,EAE9D,GAAI,CACA,IAAME,EAAc,EAAAlB,QAAG,aAAae,EAAW,MAAM,EAE/CI,EAASC,EAAeF,EAAa,IAAIF,EAAmB,QAAQ,QAAS,EAAE,CAAC,EAAE,EAGlFK,EAAY,EAAAtB,QAAK,QAAQkB,CAAU,EACpC,EAAAjB,QAAG,WAAWqB,CAAS,GACxB,EAAArB,QAAG,UAAUqB,EAAW,CAAE,UAAW,EAAK,CAAC,EAG/C,EAAArB,QAAG,cAAciB,EAAYE,EAAO,SAAU,MAAM,EAEpDR,GAAqBQ,EAAO,MAAM,iBAClCP,GAAqBO,EAAO,MAAM,aAClCN,IAEA,IAAMS,GAAUH,EAAO,MAAM,iBAAmB,MAAM,QAAQ,CAAC,EACzDI,GAAQJ,EAAO,MAAM,aAAe,MAAM,QAAQ,CAAC,EAGrDK,EAAYV,EAAK,QAAQ,UAAW,EAAE,EACtCU,IAAc,QAASA,EAAY,IAClCA,EAAY,IAAIA,CAAS,GAE9B,QAAQ,IACJ,EAAAlB,QAAG,MAAM,kBAAa,EACtB,EAAAA,QAAG,IAAI,IAAIkB,CAAS,GAAG,EACvB,EAAAlB,QAAG,KAAK,IAAIgB,CAAM,SAASC,CAAI,KAAK,CACxC,CACJ,OAASE,EAAQ,CACb,QAAQ,MAAM,EAAAnB,QAAG,IAAI,4BAAuBQ,CAAI,KAAKW,EAAE,OAAO,EAAE,CAAC,CACrE,CACJ,CAEA,QAAQ,IACJ,EAAAnB,QAAG,KACC,EAAAA,QAAG,QAAQ,aAAaO,CAAmB,wBAAwBF,EAAoB,MAAM,QAAQ,CAAC,CAAC,UAAUC,EAAoB,MAAM,QAAQ,CAAC,CAAC,IAAI,CAC7J,CACJ,EAGA,IAAMc,EAAe,QAAQ,IAAI,aAC3BC,EAAgB,QAAQ,IAAI,oBAAsB,4BAExD,GAAID,GAAgBb,EAAsB,EAAG,CACzC,QAAQ,IAAI,EAAAP,QAAG,KAAK,+CAA+CqB,CAAa,MAAM,CAAC,EACvF,GAAI,CACA,IAAMC,EAAWlB,EAAM,IAAII,GAAQ,CAC/B,IAAMU,EAAYV,EAAK,QAAQ,UAAW,EAAE,EACtCe,EAAQL,IAAc,QAAU,IAAM,IAAIA,CAAS,GACnDM,EAAS,EAAA/B,QAAK,KAAKU,EAAeK,EAAK,QAAQ,UAAW,KAAK,CAAC,EACtE,MAAO,CACH,MAAAe,EACA,SAAU,GAAGL,CAAS,MACtB,QAAS,EAAAxB,QAAG,aAAa8B,EAAQ,MAAM,CAC3C,CACJ,CAAC,EAEKC,EAAM,MAAM,MAAM,GAAGJ,CAAa,aAAc,CAClD,OAAQ,OACR,QAAS,CACL,aAAcD,EACd,eAAgB,kBACpB,EACA,KAAM,KAAK,UAAU,CAAE,MAAOE,CAAS,CAAC,CAC5C,CAAC,EAED,GAAIG,EAAI,GACJ,QAAQ,IAAI,EAAAzB,QAAG,MAAM,sCAAiC,CAAC,MACpD,CACH,IAAM0B,EAAU,MAAMD,EAAI,KAAK,EAAE,MAAM,KAAO,CAAC,EAAE,EACjD,QAAQ,IAAI,EAAAzB,QAAG,OAAO,sCAAiC0B,EAAQ,OAASD,EAAI,UAAU,EAAE,CAAC,CAC7F,CACJ,OAASN,EAAQ,CACb,QAAQ,IAAI,EAAAnB,QAAG,OAAO,qCAAgCmB,EAAE,OAAO,EAAE,CAAC,CACtE,CACJ,CAEA,QAAQ,IAAI,EAAAnB,QAAG,IAAI;AAAA,CAA8C,CAAC,CACtE,CAEAD,EAAK,EAAE,MAAM,GAAK,CACd,QAAQ,MAAM,EAAAC,QAAG,IAAI,uBAAuB,EAAE,OAAO,EAAE,CAAC,EACxD,QAAQ,KAAK,CAAC,CAClB,CAAC","names":["import_glob","import_fs","import_path","import_picocolors","cheerio","import_turndown","turndownService","TurndownService","extractContent","html","sourceUrl","originalSize","$","title","description","jsonLdScripts","_","el","raw","parsed","contentHtml","markdown","finalMarkdown","j","markdownSize","tokenReductionRatio","loadEnv","envPath","path","fs","line","trimmedLine","key","valueParts","main","pc","cwd","nextAppDirDir","ontoPublicDir","files","totalOriginalSize","totalMarkdownSize","totalFilesProcessed","file","inputPath","outputPathRelative","outputPath","htmlContent","result","extractContent","outputDir","origKb","mdKb","routeName","e","ONTO_API_KEY","DASHBOARD_URL","manifest","route","mdPath","res","errData"]}
1
+ {"version":3,"sources":["../src/cli.ts","../src/extractor.ts","../src/config.ts"],"sourcesContent":["#!/usr/bin/env node\r\nimport { glob } from 'glob';\r\nimport fs from 'fs';\r\nimport path from 'path';\r\nimport pc from 'picocolors';\r\nimport { extractContent } from './extractor';\r\nimport { generateLlmsTxt, OntoConfig } from './config';\r\n\r\nasync function loadOntoConfig(): Promise<OntoConfig | null> {\r\n try {\r\n const configPath = path.join(process.cwd(), 'onto.config');\r\n // Node.js dynamic import\r\n const config = await import('file://' + configPath.replace(/\\\\/g, '/') + '.ts');\r\n return config.default || config;\r\n } catch (error) {\r\n try {\r\n const configPath = path.join(process.cwd(), 'onto.config');\r\n const config = await import('file://' + configPath.replace(/\\\\/g, '/') + '.js');\r\n return config.default || config;\r\n } catch (e) {\r\n return null;\r\n }\r\n }\r\n}\r\n\r\n// Simple helper to load .env.local from the current working directory\r\nfunction loadEnv() {\r\n const envPath = path.join(process.cwd(), '.env.local');\r\n if (fs.existsSync(envPath)) {\r\n const envContent = fs.readFileSync(envPath, 'utf8');\r\n envContent.split(/\\r?\\n/).forEach(line => {\r\n const trimmedLine = line.trim();\r\n if (!trimmedLine || trimmedLine.startsWith('#')) return;\r\n const [key, ...valueParts] = trimmedLine.split('=');\r\n if (key && valueParts.length > 0) {\r\n process.env[key.trim()] = valueParts.join('=').trim().replace(/^[\"']|[\"']$/g, '');\r\n }\r\n });\r\n }\r\n}\r\n\r\nasync function main() {\r\n loadEnv();\r\n console.log(pc.cyan('\\n[Onto] Starting Semantic Output Generation...'));\r\n\r\n const cwd = process.cwd();\r\n const nextAppDirDir = path.join(cwd, '.next/server/app');\r\n const ontoPublicDir = path.join(cwd, 'public/.onto');\r\n\r\n if (!fs.existsSync(nextAppDirDir)) {\r\n console.log(pc.yellow(`[Onto] Could not find Next.js app output at ${nextAppDirDir}`));\r\n console.log(pc.yellow(`[Onto] Ensure this is run after \"next build\" and you are using the App Router.`));\r\n return;\r\n }\r\n\r\n // Find all HTML files rendered by Next.js in the app directory\r\n const files = await glob('**/*.html', { cwd: nextAppDirDir });\r\n\r\n if (files.length === 0) {\r\n console.log(pc.yellow(`[Onto] No static HTML files found to process.`));\r\n return;\r\n }\r\n\r\n // Ensure output directory exists\r\n if (!fs.existsSync(ontoPublicDir)) {\r\n fs.mkdirSync(ontoPublicDir, { recursive: true });\r\n }\r\n\r\n let totalOriginalSize = 0;\r\n let totalMarkdownSize = 0;\r\n let totalFilesProcessed = 0;\r\n\r\n for (const file of files) {\r\n const inputPath = path.join(nextAppDirDir, file);\r\n\r\n // We map file path e.g. \"pricing.html\" to \"pricing.md\", or \"blog/post.html\" to \"blog/post.md\"\r\n let outputPathRelative = file.replace(/\\.html$/, '.md');\r\n // If it's a dynamic route page, or purely root index.html\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 const result = extractContent(htmlContent, `/${outputPathRelative.replace(/\\.md$/, '')}`);\r\n\r\n // Ensure specific sub-directory exists (e.g., for blog/post.md)\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\r\n totalOriginalSize += result.stats.originalHtmlSize;\r\n totalMarkdownSize += result.stats.markdownSize;\r\n totalFilesProcessed++;\r\n\r\n const origKb = (result.stats.originalHtmlSize / 1024).toFixed(1);\r\n const mdKb = (result.stats.markdownSize / 1024).toFixed(1);\r\n\r\n // /index.html -> /\r\n let routeName = file.replace(/\\.html$/, '');\r\n if (routeName === 'index') routeName = '/';\r\n else routeName = `/${routeName}`;\r\n\r\n console.log(\r\n pc.green(`✓ Optimized`) +\r\n pc.dim(` ${routeName} `) +\r\n pc.blue(`[${origKb}KB -> ${mdKb}KB]`)\r\n );\r\n } catch (e: any) {\r\n console.error(pc.red(`✗ Failed to process ${file}: ${e.message}`));\r\n }\r\n }\r\n\r\n console.log(\r\n pc.bold(\r\n pc.magenta(`Processed ${totalFilesProcessed} pages. Total Size: ${(totalOriginalSize / 1024).toFixed(1)}KB -> ${(totalMarkdownSize / 1024).toFixed(1)}KB`)\r\n )\r\n );\r\n\r\n // Sync with Onto Control Plane (Premium)\r\n const ONTO_API_KEY = process.env.ONTO_API_KEY;\r\n const DASHBOARD_URL = process.env.ONTO_DASHBOARD_URL || 'https://app.buildonto.dev';\r\n\r\n if (ONTO_API_KEY && totalFilesProcessed > 0) {\r\n console.log(pc.cyan(`[Onto] Syncing manifest with Control Plane [${DASHBOARD_URL}]...`));\r\n try {\r\n const manifest = files.map(file => {\r\n const routeName = file.replace(/\\.html$/, '');\r\n const route = routeName === 'index' ? '/' : `/${routeName}`;\r\n const mdPath = path.join(ontoPublicDir, file.replace(/\\.html$/, '.md'));\r\n return {\r\n route,\r\n filename: `${routeName}.md`,\r\n content: fs.readFileSync(mdPath, 'utf8')\r\n };\r\n });\r\n\r\n const res = await fetch(`${DASHBOARD_URL}/api/files`, {\r\n method: 'POST',\r\n headers: {\r\n 'x-onto-key': ONTO_API_KEY,\r\n 'Content-Type': 'application/json'\r\n },\r\n body: JSON.stringify({ files: manifest })\r\n });\r\n\r\n if (res.ok) {\r\n console.log(pc.green('✓ Control Plane sync successful'));\r\n } else {\r\n const errData = await res.json().catch(() => ({}));\r\n console.log(pc.yellow(`⚠ Control Plane sync skipped: ${errData.error || res.statusText}`));\r\n }\r\n } catch (e: any) {\r\n console.log(pc.yellow(`⚠ Control Plane sync failed: ${e.message}`));\r\n }\r\n }\r\n\r\n // --- Generate llms.txt manifest ---\r\n const config = await loadOntoConfig();\r\n if (config) {\r\n const llmsTxtContent = generateLlmsTxt(config);\r\n const llmsTxtPath = path.join(cwd, 'public/llms.txt');\r\n \r\n // Ensure public dir exists\r\n const publicDir = path.join(cwd, 'public');\r\n if (!fs.existsSync(publicDir)) {\r\n fs.mkdirSync(publicDir, { recursive: true });\r\n }\r\n\r\n fs.writeFileSync(llmsTxtPath, llmsTxtContent, 'utf8');\r\n console.log(pc.green('✓ Generated') + pc.dim(' /llms.txt'));\r\n }\r\n\r\n console.log(pc.dim(`Edge payloads are ready at /public/.onto/*\\n`));\r\n}\r\n\r\nmain().catch(e => {\r\n console.error(pc.red(`[Onto] Fatal Error: ${e.message}`));\r\n process.exit(1);\r\n});\r\n","import * as cheerio from 'cheerio';\r\nimport TurndownService from 'turndown';\r\n\r\nconst turndownService = new TurndownService({\r\n headingStyle: 'atx',\r\n codeBlockStyle: 'fenced',\r\n});\r\n\r\n// Configure turndown to keep some layout or handle semantic tags differently if needed\r\n\r\nexport interface ExtractionResult {\r\n markdown: string;\r\n metadata: {\r\n title: string;\r\n description: string;\r\n jsonLd: any[];\r\n };\r\n stats: {\r\n originalHtmlSize: number;\r\n markdownSize: number;\r\n tokenReductionRatio: number;\r\n };\r\n}\r\n\r\n/**\r\n * Extracts pure semantic markdown and metadata from rendered Next.js HTML strings.\r\n * @param html The raw HTML string.\r\n * @param sourceUrl (Optional) the URL this was generated from, to attach as metadata.\r\n * @returns {ExtractionResult} The extracted payload.\r\n */\r\nexport function extractContent(html: string, sourceUrl: string = 'Generated Output'): ExtractionResult {\r\n const originalSize = html.length;\r\n\r\n const $ = cheerio.load(html);\r\n\r\n // 1. Extract Metadata BEFORE removing structure\r\n const title = $('title').text() || $('h1').first().text() || 'Untitled Page';\r\n const description = $('meta[name=\"description\"]').attr('content') || 'No description found.';\r\n\r\n const jsonLdScripts: any[] = [];\r\n $('script[type=\"application/ld+json\"]').each((_, el) => {\r\n try {\r\n const raw = $(el).html() || '';\r\n const parsed = JSON.parse(raw);\r\n jsonLdScripts.push(parsed);\r\n } catch {\r\n // ignore bad json\r\n }\r\n });\r\n\r\n // 2. Strip noise (React boilerplate, styles, unnecessary tags)\r\n $('script, style, noscript, iframe, svg, nav, footer, meta, link, header').remove();\r\n\r\n // Optionally remove typical Next.js hidden wrappers if they don't contain real content.\r\n // Next.js uses <div id=\"__next\"> but we mostly just want semantic content.\r\n\r\n // 3. Find the entry point for content\r\n // Prefer <main> or <article> over <body>\r\n let contentHtml = '';\r\n if ($('main').length > 0) {\r\n contentHtml = $('main').html() || '';\r\n } else if ($('article').length > 0) {\r\n contentHtml = $('article').html() || '';\r\n } else {\r\n contentHtml = $('body').html() || '';\r\n }\r\n\r\n // 4. Convert to Markdown\r\n let markdown = turndownService.turndown(contentHtml);\r\n\r\n // 5. Optionally inject Metadata header\r\n const headerLines = [\r\n `# ${title}`,\r\n `> ${description}`,\r\n ``,\r\n `**Source:** ${sourceUrl}`,\r\n `**Extracted:** ${new Date().toISOString()}`,\r\n ``,\r\n `---`,\r\n ``\r\n ];\r\n\r\n let finalMarkdown = headerLines.join('\\n') + markdown;\r\n\r\n // Add JSON-LD section if exists\r\n if (jsonLdScripts.length > 0) {\r\n finalMarkdown += '\\n\\n---\\n## Structured Data (JSON-LD)\\n```json\\n';\r\n jsonLdScripts.forEach(j => {\r\n finalMarkdown += JSON.stringify(j, null, 2) + '\\n';\r\n });\r\n finalMarkdown += '```\\n';\r\n }\r\n\r\n const markdownSize = finalMarkdown.length;\r\n const tokenReductionRatio = originalSize > 0 ? ((originalSize - markdownSize) / originalSize) * 100 : 0;\r\n\r\n return {\r\n markdown: finalMarkdown,\r\n metadata: {\r\n title,\r\n description,\r\n jsonLd: jsonLdScripts\r\n },\r\n stats: {\r\n originalHtmlSize: originalSize,\r\n markdownSize,\r\n tokenReductionRatio\r\n }\r\n };\r\n}\r\n\r\nexport async function generateStaticPayloads(nextAppDirDir: string, ontoPublicDir: string) {\r\n const fs = await import('fs');\r\n const path = await import('path');\r\n const { glob } = await import('glob');\r\n\r\n if (!fs.existsSync(nextAppDirDir)) {\r\n return;\r\n }\r\n\r\n const files = await glob('**/*.html', { cwd: nextAppDirDir });\r\n if (files.length === 0) return;\r\n\r\n if (!fs.existsSync(ontoPublicDir)) {\r\n fs.mkdirSync(ontoPublicDir, { recursive: true });\r\n }\r\n\r\n let totalFilesProcessed = 0;\r\n\r\n for (const file of files) {\r\n const inputPath = path.join(nextAppDirDir, file);\r\n const outputPathRelative = file.replace(/\\.html$/, '.md');\r\n const outputPath = path.join(ontoPublicDir, outputPathRelative);\r\n\r\n try {\r\n const htmlContent = fs.readFileSync(inputPath, 'utf8');\r\n\r\n let routeName = file.replace(/\\.html$/, '');\r\n if (routeName === 'index') routeName = '/';\r\n else routeName = `/${routeName}`;\r\n\r\n const result = extractContent(htmlContent, routeName);\r\n\r\n const outputDir = path.dirname(outputPath);\r\n if (!fs.existsSync(outputDir)) {\r\n fs.mkdirSync(outputDir, { recursive: true });\r\n }\r\n\r\n fs.writeFileSync(outputPath, result.markdown, 'utf8');\r\n totalFilesProcessed++;\r\n } catch (e: any) {\r\n console.error(`[Onto] Failed to process ${file}: ${e.message}`);\r\n }\r\n }\r\n console.log(`[Onto] Successfully generated ${totalFilesProcessed} semantic markdown endpoints.`);\r\n}\r\n","/**\r\n * Configuration schema for onto.config.ts\r\n * Used to dynamically generate llms.txt and other AI discovery files\r\n */\r\n\r\nexport type PageType = 'scoring' | 'about' | 'default';\r\n\r\nexport interface OntoRoute {\r\n /**\r\n * The URL path (e.g., '/docs', '/api/reference')\r\n */\r\n path: string;\r\n /**\r\n * Description of what this route contains\r\n */\r\n description: string;\r\n /**\r\n * Optional: Page type for automatic JSON-LD schema injection\r\n * - 'scoring': Injects Methodology schema with AIO scoring weights (40/35/25)\r\n * - 'about': Injects Organization/AboutPage schema\r\n * - 'default': No automatic schema injection\r\n */\r\n pageType?: PageType;\r\n}\r\n\r\nexport interface OntoConfig {\r\n /**\r\n * The name of your project or site (required)\r\n * Used as the H1 heading in llms.txt\r\n */\r\n name: string;\r\n\r\n /**\r\n * A short summary of your project (required)\r\n * Displayed as a blockquote in llms.txt\r\n * Should contain key information necessary for understanding the rest of the file\r\n */\r\n summary: string;\r\n\r\n /**\r\n * The base URL of your site (e.g., 'https://example.com')\r\n */\r\n baseUrl: string;\r\n\r\n /**\r\n * Optional: Additional sections to include in llms.txt\r\n * Each section can contain any markdown content\r\n */\r\n sections?: {\r\n heading: string;\r\n content: string;\r\n }[];\r\n\r\n /**\r\n * Key routes that AI agents should know about\r\n * These will be formatted as a markdown list in llms.txt\r\n */\r\n routes?: OntoRoute[];\r\n\r\n /**\r\n * Optional: Links to external resources (documentation, API references, etc.)\r\n */\r\n externalLinks?: {\r\n title: string;\r\n url: string;\r\n description?: string;\r\n }[];\r\n\r\n /**\r\n * Optional: Organization information for JSON-LD schemas\r\n */\r\n organization?: {\r\n name: string;\r\n description?: string;\r\n url?: string;\r\n logo?: string;\r\n foundingDate?: string;\r\n };\r\n}\r\n\r\n/**\r\n * Generate llms.txt content from OntoConfig\r\n * Follows the llms.txt specification:\r\n * - H1 with project name\r\n * - Blockquote with summary\r\n * - Additional markdown sections\r\n */\r\nexport function generateLlmsTxt(config: OntoConfig): string {\r\n const lines: string[] = [];\r\n\r\n // H1: Project name (required)\r\n lines.push(`# ${config.name}`);\r\n lines.push('');\r\n\r\n // Blockquote: Summary (required)\r\n lines.push(`> ${config.summary}`);\r\n lines.push('');\r\n\r\n // Key Routes section (if provided)\r\n if (config.routes && config.routes.length > 0) {\r\n lines.push('## Key Routes');\r\n lines.push('');\r\n for (const route of config.routes) {\r\n const fullUrl = `${config.baseUrl}${route.path}`;\r\n lines.push(`- [${route.path}](${fullUrl}): ${route.description}`);\r\n }\r\n lines.push('');\r\n }\r\n\r\n // External Links section (if provided)\r\n if (config.externalLinks && config.externalLinks.length > 0) {\r\n lines.push('## Resources');\r\n lines.push('');\r\n for (const link of config.externalLinks) {\r\n if (link.description) {\r\n lines.push(`- [${link.title}](${link.url}): ${link.description}`);\r\n } else {\r\n lines.push(`- [${link.title}](${link.url})`);\r\n }\r\n }\r\n lines.push('');\r\n }\r\n\r\n // Custom sections (if provided)\r\n if (config.sections && config.sections.length > 0) {\r\n for (const section of config.sections) {\r\n lines.push(`## ${section.heading}`);\r\n lines.push('');\r\n lines.push(section.content);\r\n lines.push('');\r\n }\r\n }\r\n\r\n return lines.join('\\n').trim() + '\\n';\r\n}\r\n"],"mappings":";wdACA,IAAAA,EAAqB,gBACrBC,EAAe,iBACfC,EAAiB,mBACjBC,EAAe,yBCJf,IAAAC,EAAyB,sBACzBC,EAA4B,uBAEtBC,EAAkB,IAAI,EAAAC,QAAgB,CACxC,aAAc,MACd,eAAgB,QACpB,CAAC,EAwBM,SAASC,EAAeC,EAAcC,EAAoB,mBAAsC,CACnG,IAAMC,EAAeF,EAAK,OAEpBG,EAAY,OAAKH,CAAI,EAGrBI,EAAQD,EAAE,OAAO,EAAE,KAAK,GAAKA,EAAE,IAAI,EAAE,MAAM,EAAE,KAAK,GAAK,gBACvDE,EAAcF,EAAE,0BAA0B,EAAE,KAAK,SAAS,GAAK,wBAE/DG,EAAuB,CAAC,EAC9BH,EAAE,oCAAoC,EAAE,KAAK,CAACI,EAAGC,IAAO,CACpD,GAAI,CACA,IAAMC,EAAMN,EAAEK,CAAE,EAAE,KAAK,GAAK,GACtBE,EAAS,KAAK,MAAMD,CAAG,EAC7BH,EAAc,KAAKI,CAAM,CAC7B,MAAQ,CAER,CACJ,CAAC,EAGDP,EAAE,uEAAuE,EAAE,OAAO,EAOlF,IAAIQ,EAAc,GACdR,EAAE,MAAM,EAAE,OAAS,EACnBQ,EAAcR,EAAE,MAAM,EAAE,KAAK,GAAK,GAC3BA,EAAE,SAAS,EAAE,OAAS,EAC7BQ,EAAcR,EAAE,SAAS,EAAE,KAAK,GAAK,GAErCQ,EAAcR,EAAE,MAAM,EAAE,KAAK,GAAK,GAItC,IAAIS,EAAWf,EAAgB,SAASc,CAAW,EAc/CE,EAXgB,CAChB,KAAKT,CAAK,GACV,KAAKC,CAAW,GAChB,GACA,eAAeJ,CAAS,GACxB,kBAAkB,IAAI,KAAK,EAAE,YAAY,CAAC,GAC1C,GACA,MACA,EACJ,EAEgC,KAAK;AAAA,CAAI,EAAIW,EAGzCN,EAAc,OAAS,IACvBO,GAAiB;AAAA;AAAA;AAAA;AAAA;AAAA,EACjBP,EAAc,QAAQQ,GAAK,CACvBD,GAAiB,KAAK,UAAUC,EAAG,KAAM,CAAC,EAAI;AAAA,CAClD,CAAC,EACDD,GAAiB,SAGrB,IAAME,EAAeF,EAAc,OAC7BG,EAAsBd,EAAe,GAAMA,EAAea,GAAgBb,EAAgB,IAAM,EAEtG,MAAO,CACH,SAAUW,EACV,SAAU,CACN,MAAAT,EACA,YAAAC,EACA,OAAQC,CACZ,EACA,MAAO,CACH,iBAAkBJ,EAClB,aAAAa,EACA,oBAAAC,CACJ,CACJ,CACJ,CCtBO,SAASC,EAAgBC,EAA4B,CAC1D,IAAMC,EAAkB,CAAC,EAWzB,GARAA,EAAM,KAAK,KAAKD,EAAO,IAAI,EAAE,EAC7BC,EAAM,KAAK,EAAE,EAGbA,EAAM,KAAK,KAAKD,EAAO,OAAO,EAAE,EAChCC,EAAM,KAAK,EAAE,EAGTD,EAAO,QAAUA,EAAO,OAAO,OAAS,EAAG,CAC7CC,EAAM,KAAK,eAAe,EAC1BA,EAAM,KAAK,EAAE,EACb,QAAWC,KAASF,EAAO,OAAQ,CACjC,IAAMG,EAAU,GAAGH,EAAO,OAAO,GAAGE,EAAM,IAAI,GAC9CD,EAAM,KAAK,MAAMC,EAAM,IAAI,KAAKC,CAAO,MAAMD,EAAM,WAAW,EAAE,CAClE,CACAD,EAAM,KAAK,EAAE,CACf,CAGA,GAAID,EAAO,eAAiBA,EAAO,cAAc,OAAS,EAAG,CAC3DC,EAAM,KAAK,cAAc,EACzBA,EAAM,KAAK,EAAE,EACb,QAAWG,KAAQJ,EAAO,cACpBI,EAAK,YACPH,EAAM,KAAK,MAAMG,EAAK,KAAK,KAAKA,EAAK,GAAG,MAAMA,EAAK,WAAW,EAAE,EAEhEH,EAAM,KAAK,MAAMG,EAAK,KAAK,KAAKA,EAAK,GAAG,GAAG,EAG/CH,EAAM,KAAK,EAAE,CACf,CAGA,GAAID,EAAO,UAAYA,EAAO,SAAS,OAAS,EAC9C,QAAWK,KAAWL,EAAO,SAC3BC,EAAM,KAAK,MAAMI,EAAQ,OAAO,EAAE,EAClCJ,EAAM,KAAK,EAAE,EACbA,EAAM,KAAKI,EAAQ,OAAO,EAC1BJ,EAAM,KAAK,EAAE,EAIjB,OAAOA,EAAM,KAAK;AAAA,CAAI,EAAE,KAAK,EAAI;AAAA,CACnC,CF9HA,eAAeK,GAA6C,CACxD,GAAI,CAGA,IAAMC,EAAS,MAAM,OAAO,UAFT,EAAAC,QAAK,KAAK,QAAQ,IAAI,EAAG,aAAa,EAEN,QAAQ,MAAO,GAAG,EAAI,OACzE,OAAOD,EAAO,SAAWA,CAC7B,MAAgB,CACZ,GAAI,CAEA,IAAMA,EAAS,MAAM,OAAO,UADT,EAAAC,QAAK,KAAK,QAAQ,IAAI,EAAG,aAAa,EACN,QAAQ,MAAO,GAAG,EAAI,OACzE,OAAOD,EAAO,SAAWA,CAC7B,MAAY,CACR,OAAO,IACX,CACJ,CACJ,CAGA,SAASE,GAAU,CACf,IAAMC,EAAU,EAAAF,QAAK,KAAK,QAAQ,IAAI,EAAG,YAAY,EACjD,EAAAG,QAAG,WAAWD,CAAO,GACF,EAAAC,QAAG,aAAaD,EAAS,MAAM,EACvC,MAAM,OAAO,EAAE,QAAQE,GAAQ,CACtC,IAAMC,EAAcD,EAAK,KAAK,EAC9B,GAAI,CAACC,GAAeA,EAAY,WAAW,GAAG,EAAG,OACjD,GAAM,CAACC,EAAK,GAAGC,CAAU,EAAIF,EAAY,MAAM,GAAG,EAC9CC,GAAOC,EAAW,OAAS,IAC3B,QAAQ,IAAID,EAAI,KAAK,CAAC,EAAIC,EAAW,KAAK,GAAG,EAAE,KAAK,EAAE,QAAQ,eAAgB,EAAE,EAExF,CAAC,CAET,CAEA,eAAeC,GAAO,CAClBP,EAAQ,EACR,QAAQ,IAAI,EAAAQ,QAAG,KAAK;AAAA,8CAAiD,CAAC,EAEtE,IAAMC,EAAM,QAAQ,IAAI,EAClBC,EAAgB,EAAAX,QAAK,KAAKU,EAAK,kBAAkB,EACjDE,EAAgB,EAAAZ,QAAK,KAAKU,EAAK,cAAc,EAEnD,GAAI,CAAC,EAAAP,QAAG,WAAWQ,CAAa,EAAG,CAC/B,QAAQ,IAAI,EAAAF,QAAG,OAAO,+CAA+CE,CAAa,EAAE,CAAC,EACrF,QAAQ,IAAI,EAAAF,QAAG,OAAO,gFAAgF,CAAC,EACvG,MACJ,CAGA,IAAMI,EAAQ,QAAM,QAAK,YAAa,CAAE,IAAKF,CAAc,CAAC,EAE5D,GAAIE,EAAM,SAAW,EAAG,CACpB,QAAQ,IAAI,EAAAJ,QAAG,OAAO,+CAA+C,CAAC,EACtE,MACJ,CAGK,EAAAN,QAAG,WAAWS,CAAa,GAC5B,EAAAT,QAAG,UAAUS,EAAe,CAAE,UAAW,EAAK,CAAC,EAGnD,IAAIE,EAAoB,EACpBC,EAAoB,EACpBC,EAAsB,EAE1B,QAAWC,KAAQJ,EAAO,CACtB,IAAMK,EAAY,EAAAlB,QAAK,KAAKW,EAAeM,CAAI,EAG3CE,EAAqBF,EAAK,QAAQ,UAAW,KAAK,EAEhDG,EAAa,EAAApB,QAAK,KAAKY,EAAeO,CAAkB,EAE9D,GAAI,CACA,IAAME,EAAc,EAAAlB,QAAG,aAAae,EAAW,MAAM,EAE/CI,EAASC,EAAeF,EAAa,IAAIF,EAAmB,QAAQ,QAAS,EAAE,CAAC,EAAE,EAGlFK,EAAY,EAAAxB,QAAK,QAAQoB,CAAU,EACpC,EAAAjB,QAAG,WAAWqB,CAAS,GACxB,EAAArB,QAAG,UAAUqB,EAAW,CAAE,UAAW,EAAK,CAAC,EAG/C,EAAArB,QAAG,cAAciB,EAAYE,EAAO,SAAU,MAAM,EAEpDR,GAAqBQ,EAAO,MAAM,iBAClCP,GAAqBO,EAAO,MAAM,aAClCN,IAEA,IAAMS,GAAUH,EAAO,MAAM,iBAAmB,MAAM,QAAQ,CAAC,EACzDI,GAAQJ,EAAO,MAAM,aAAe,MAAM,QAAQ,CAAC,EAGrDK,EAAYV,EAAK,QAAQ,UAAW,EAAE,EACtCU,IAAc,QAASA,EAAY,IAClCA,EAAY,IAAIA,CAAS,GAE9B,QAAQ,IACJ,EAAAlB,QAAG,MAAM,kBAAa,EACtB,EAAAA,QAAG,IAAI,IAAIkB,CAAS,GAAG,EACvB,EAAAlB,QAAG,KAAK,IAAIgB,CAAM,SAASC,CAAI,KAAK,CACxC,CACJ,OAASE,EAAQ,CACb,QAAQ,MAAM,EAAAnB,QAAG,IAAI,4BAAuBQ,CAAI,KAAKW,EAAE,OAAO,EAAE,CAAC,CACrE,CACJ,CAEA,QAAQ,IACJ,EAAAnB,QAAG,KACC,EAAAA,QAAG,QAAQ,aAAaO,CAAmB,wBAAwBF,EAAoB,MAAM,QAAQ,CAAC,CAAC,UAAUC,EAAoB,MAAM,QAAQ,CAAC,CAAC,IAAI,CAC7J,CACJ,EAGA,IAAMc,EAAe,QAAQ,IAAI,aAC3BC,EAAgB,QAAQ,IAAI,oBAAsB,4BAExD,GAAID,GAAgBb,EAAsB,EAAG,CACzC,QAAQ,IAAI,EAAAP,QAAG,KAAK,+CAA+CqB,CAAa,MAAM,CAAC,EACvF,GAAI,CACA,IAAMC,EAAWlB,EAAM,IAAII,GAAQ,CAC/B,IAAMU,EAAYV,EAAK,QAAQ,UAAW,EAAE,EACtCe,EAAQL,IAAc,QAAU,IAAM,IAAIA,CAAS,GACnDM,EAAS,EAAAjC,QAAK,KAAKY,EAAeK,EAAK,QAAQ,UAAW,KAAK,CAAC,EACtE,MAAO,CACH,MAAAe,EACA,SAAU,GAAGL,CAAS,MACtB,QAAS,EAAAxB,QAAG,aAAa8B,EAAQ,MAAM,CAC3C,CACJ,CAAC,EAEKC,EAAM,MAAM,MAAM,GAAGJ,CAAa,aAAc,CAClD,OAAQ,OACR,QAAS,CACL,aAAcD,EACd,eAAgB,kBACpB,EACA,KAAM,KAAK,UAAU,CAAE,MAAOE,CAAS,CAAC,CAC5C,CAAC,EAED,GAAIG,EAAI,GACJ,QAAQ,IAAI,EAAAzB,QAAG,MAAM,sCAAiC,CAAC,MACpD,CACH,IAAM0B,EAAU,MAAMD,EAAI,KAAK,EAAE,MAAM,KAAO,CAAC,EAAE,EACjD,QAAQ,IAAI,EAAAzB,QAAG,OAAO,sCAAiC0B,EAAQ,OAASD,EAAI,UAAU,EAAE,CAAC,CAC7F,CACJ,OAASN,EAAQ,CACb,QAAQ,IAAI,EAAAnB,QAAG,OAAO,qCAAgCmB,EAAE,OAAO,EAAE,CAAC,CACtE,CACJ,CAGA,IAAM7B,EAAS,MAAMD,EAAe,EACpC,GAAIC,EAAQ,CACR,IAAMqC,EAAiBC,EAAgBtC,CAAM,EACvCuC,EAAc,EAAAtC,QAAK,KAAKU,EAAK,iBAAiB,EAG9C6B,EAAY,EAAAvC,QAAK,KAAKU,EAAK,QAAQ,EACpC,EAAAP,QAAG,WAAWoC,CAAS,GACxB,EAAApC,QAAG,UAAUoC,EAAW,CAAE,UAAW,EAAK,CAAC,EAG/C,EAAApC,QAAG,cAAcmC,EAAaF,EAAgB,MAAM,EACpD,QAAQ,IAAI,EAAA3B,QAAG,MAAM,kBAAa,EAAI,EAAAA,QAAG,IAAI,YAAY,CAAC,CAC9D,CAEA,QAAQ,IAAI,EAAAA,QAAG,IAAI;AAAA,CAA8C,CAAC,CACtE,CAEAD,EAAK,EAAE,MAAM,GAAK,CACd,QAAQ,MAAM,EAAAC,QAAG,IAAI,uBAAuB,EAAE,OAAO,EAAE,CAAC,EACxD,QAAQ,KAAK,CAAC,CAClB,CAAC","names":["import_glob","import_fs","import_path","import_picocolors","cheerio","import_turndown","turndownService","TurndownService","extractContent","html","sourceUrl","originalSize","$","title","description","jsonLdScripts","_","el","raw","parsed","contentHtml","markdown","finalMarkdown","j","markdownSize","tokenReductionRatio","generateLlmsTxt","config","lines","route","fullUrl","link","section","loadOntoConfig","config","path","loadEnv","envPath","fs","line","trimmedLine","key","valueParts","main","pc","cwd","nextAppDirDir","ontoPublicDir","files","totalOriginalSize","totalMarkdownSize","totalFilesProcessed","file","inputPath","outputPathRelative","outputPath","htmlContent","result","extractContent","outputDir","origKb","mdKb","routeName","e","ONTO_API_KEY","DASHBOARD_URL","manifest","route","mdPath","res","errData","llmsTxtContent","generateLlmsTxt","llmsTxtPath","publicDir"]}
package/dist/cli.mjs CHANGED
@@ -1,12 +1,14 @@
1
1
  #!/usr/bin/env node
2
- import{glob as j}from"glob";import r from"fs";import h from"path";import e from"picocolors";import*as x from"cheerio";import v from"turndown";var P=new v({headingStyle:"atx",codeBlockStyle:"fenced"});function O(i,f="Generated Output"){let n=i.length,t=x.load(i),l=t("title").text()||t("h1").first().text()||"Untitled Page",d=t('meta[name="description"]').attr("content")||"No description found.",m=[];t('script[type="application/ld+json"]').each((p,c)=>{try{let w=t(c).html()||"",$=JSON.parse(w);m.push($)}catch{}}),t("script, style, noscript, iframe, svg, nav, footer, meta, link, header").remove();let g="";t("main").length>0?g=t("main").html()||"":t("article").length>0?g=t("article").html()||"":g=t("body").html()||"";let S=P.turndown(g),o=[`# ${l}`,`> ${d}`,"",`**Source:** ${f}`,`**Extracted:** ${new Date().toISOString()}`,"","---",""].join(`
3
- `)+S;m.length>0&&(o+=`
2
+ import{glob as C}from"glob";import r from"fs";import c from"path";import s from"picocolors";import*as w from"cheerio";import v from"turndown";var b=new v({headingStyle:"atx",codeBlockStyle:"fenced"});function O(n,t="Generated Output"){let e=n.length,o=w.load(n),d=o("title").text()||o("h1").first().text()||"Untitled Page",m=o('meta[name="description"]').attr("content")||"No description found.",f=[];o('script[type="application/ld+json"]').each((u,h)=>{try{let p=o(h).html()||"",$=JSON.parse(p);f.push($)}catch{}}),o("script, style, noscript, iframe, svg, nav, footer, meta, link, header").remove();let g="";o("main").length>0?g=o("main").html()||"":o("article").length>0?g=o("article").html()||"":g=o("body").html()||"";let S=b.turndown(g),i=[`# ${d}`,`> ${m}`,"",`**Source:** ${t}`,`**Extracted:** ${new Date().toISOString()}`,"","---",""].join(`
3
+ `)+S;f.length>0&&(i+=`
4
4
 
5
5
  ---
6
6
  ## Structured Data (JSON-LD)
7
7
  \`\`\`json
8
- `,m.forEach(p=>{o+=JSON.stringify(p,null,2)+`
9
- `}),o+="```\n");let s=o.length,u=n>0?(n-s)/n*100:0;return{markdown:o,metadata:{title:l,description:d,jsonLd:m},stats:{originalHtmlSize:n,markdownSize:s,tokenReductionRatio:u}}}function b(){let i=h.join(process.cwd(),".env.local");r.existsSync(i)&&r.readFileSync(i,"utf8").split(/\r?\n/).forEach(n=>{let t=n.trim();if(!t||t.startsWith("#"))return;let[l,...d]=t.split("=");l&&d.length>0&&(process.env[l.trim()]=d.join("=").trim().replace(/^["']|["']$/g,""))})}async function F(){b(),console.log(e.cyan(`
10
- [Onto] Starting Semantic Output Generation...`));let i=process.cwd(),f=h.join(i,".next/server/app"),n=h.join(i,"public/.onto");if(!r.existsSync(f)){console.log(e.yellow(`[Onto] Could not find Next.js app output at ${f}`)),console.log(e.yellow('[Onto] Ensure this is run after "next build" and you are using the App Router.'));return}let t=await j("**/*.html",{cwd:f});if(t.length===0){console.log(e.yellow("[Onto] No static HTML files found to process."));return}r.existsSync(n)||r.mkdirSync(n,{recursive:!0});let l=0,d=0,m=0;for(let a of t){let o=h.join(f,a),s=a.replace(/\.html$/,".md"),u=h.join(n,s);try{let p=r.readFileSync(o,"utf8"),c=O(p,`/${s.replace(/\.md$/,"")}`),w=h.dirname(u);r.existsSync(w)||r.mkdirSync(w,{recursive:!0}),r.writeFileSync(u,c.markdown,"utf8"),l+=c.stats.originalHtmlSize,d+=c.stats.markdownSize,m++;let $=(c.stats.originalHtmlSize/1024).toFixed(1),k=(c.stats.markdownSize/1024).toFixed(1),y=a.replace(/\.html$/,"");y==="index"?y="/":y=`/${y}`,console.log(e.green("\u2713 Optimized")+e.dim(` ${y} `)+e.blue(`[${$}KB -> ${k}KB]`))}catch(p){console.error(e.red(`\u2717 Failed to process ${a}: ${p.message}`))}}console.log(e.bold(e.magenta(`Processed ${m} pages. Total Size: ${(l/1024).toFixed(1)}KB -> ${(d/1024).toFixed(1)}KB`)));let g=process.env.ONTO_API_KEY,S=process.env.ONTO_DASHBOARD_URL||"https://app.buildonto.dev";if(g&&m>0){console.log(e.cyan(`[Onto] Syncing manifest with Control Plane [${S}]...`));try{let a=t.map(s=>{let u=s.replace(/\.html$/,""),p=u==="index"?"/":`/${u}`,c=h.join(n,s.replace(/\.html$/,".md"));return{route:p,filename:`${u}.md`,content:r.readFileSync(c,"utf8")}}),o=await fetch(`${S}/api/files`,{method:"POST",headers:{"x-onto-key":g,"Content-Type":"application/json"},body:JSON.stringify({files:a})});if(o.ok)console.log(e.green("\u2713 Control Plane sync successful"));else{let s=await o.json().catch(()=>({}));console.log(e.yellow(`\u26A0 Control Plane sync skipped: ${s.error||o.statusText}`))}}catch(a){console.log(e.yellow(`\u26A0 Control Plane sync failed: ${a.message}`))}}console.log(e.dim(`Edge payloads are ready at /public/.onto/*
11
- `))}F().catch(i=>{console.error(e.red(`[Onto] Fatal Error: ${i.message}`)),process.exit(1)});
8
+ `,f.forEach(u=>{i+=JSON.stringify(u,null,2)+`
9
+ `}),i+="```\n");let l=i.length,a=e>0?(e-l)/e*100:0;return{markdown:i,metadata:{title:d,description:m,jsonLd:f},stats:{originalHtmlSize:e,markdownSize:l,tokenReductionRatio:a}}}function k(n){let t=[];if(t.push(`# ${n.name}`),t.push(""),t.push(`> ${n.summary}`),t.push(""),n.routes&&n.routes.length>0){t.push("## Key Routes"),t.push("");for(let e of n.routes){let o=`${n.baseUrl}${e.path}`;t.push(`- [${e.path}](${o}): ${e.description}`)}t.push("")}if(n.externalLinks&&n.externalLinks.length>0){t.push("## Resources"),t.push("");for(let e of n.externalLinks)e.description?t.push(`- [${e.title}](${e.url}): ${e.description}`):t.push(`- [${e.title}](${e.url})`);t.push("")}if(n.sections&&n.sections.length>0)for(let e of n.sections)t.push(`## ${e.heading}`),t.push(""),t.push(e.content),t.push("");return t.join(`
10
+ `).trim()+`
11
+ `}async function R(){try{let t=await import("file://"+c.join(process.cwd(),"onto.config").replace(/\\/g,"/")+".ts");return t.default||t}catch{try{let e=await import("file://"+c.join(process.cwd(),"onto.config").replace(/\\/g,"/")+".js");return e.default||e}catch{return null}}}function F(){let n=c.join(process.cwd(),".env.local");r.existsSync(n)&&r.readFileSync(n,"utf8").split(/\r?\n/).forEach(e=>{let o=e.trim();if(!o||o.startsWith("#"))return;let[d,...m]=o.split("=");d&&m.length>0&&(process.env[d.trim()]=m.join("=").trim().replace(/^["']|["']$/g,""))})}async function L(){F(),console.log(s.cyan(`
12
+ [Onto] Starting Semantic Output Generation...`));let n=process.cwd(),t=c.join(n,".next/server/app"),e=c.join(n,"public/.onto");if(!r.existsSync(t)){console.log(s.yellow(`[Onto] Could not find Next.js app output at ${t}`)),console.log(s.yellow('[Onto] Ensure this is run after "next build" and you are using the App Router.'));return}let o=await C("**/*.html",{cwd:t});if(o.length===0){console.log(s.yellow("[Onto] No static HTML files found to process."));return}r.existsSync(e)||r.mkdirSync(e,{recursive:!0});let d=0,m=0,f=0;for(let i of o){let l=c.join(t,i),a=i.replace(/\.html$/,".md"),u=c.join(e,a);try{let h=r.readFileSync(l,"utf8"),p=O(h,`/${a.replace(/\.md$/,"")}`),$=c.dirname(u);r.existsSync($)||r.mkdirSync($,{recursive:!0}),r.writeFileSync(u,p.markdown,"utf8"),d+=p.stats.originalHtmlSize,m+=p.stats.markdownSize,f++;let P=(p.stats.originalHtmlSize/1024).toFixed(1),j=(p.stats.markdownSize/1024).toFixed(1),y=i.replace(/\.html$/,"");y==="index"?y="/":y=`/${y}`,console.log(s.green("\u2713 Optimized")+s.dim(` ${y} `)+s.blue(`[${P}KB -> ${j}KB]`))}catch(h){console.error(s.red(`\u2717 Failed to process ${i}: ${h.message}`))}}console.log(s.bold(s.magenta(`Processed ${f} pages. Total Size: ${(d/1024).toFixed(1)}KB -> ${(m/1024).toFixed(1)}KB`)));let g=process.env.ONTO_API_KEY,S=process.env.ONTO_DASHBOARD_URL||"https://app.buildonto.dev";if(g&&f>0){console.log(s.cyan(`[Onto] Syncing manifest with Control Plane [${S}]...`));try{let i=o.map(a=>{let u=a.replace(/\.html$/,""),h=u==="index"?"/":`/${u}`,p=c.join(e,a.replace(/\.html$/,".md"));return{route:h,filename:`${u}.md`,content:r.readFileSync(p,"utf8")}}),l=await fetch(`${S}/api/files`,{method:"POST",headers:{"x-onto-key":g,"Content-Type":"application/json"},body:JSON.stringify({files:i})});if(l.ok)console.log(s.green("\u2713 Control Plane sync successful"));else{let a=await l.json().catch(()=>({}));console.log(s.yellow(`\u26A0 Control Plane sync skipped: ${a.error||l.statusText}`))}}catch(i){console.log(s.yellow(`\u26A0 Control Plane sync failed: ${i.message}`))}}let x=await R();if(x){let i=k(x),l=c.join(n,"public/llms.txt"),a=c.join(n,"public");r.existsSync(a)||r.mkdirSync(a,{recursive:!0}),r.writeFileSync(l,i,"utf8"),console.log(s.green("\u2713 Generated")+s.dim(" /llms.txt"))}console.log(s.dim(`Edge payloads are ready at /public/.onto/*
13
+ `))}L().catch(n=>{console.error(s.red(`[Onto] Fatal Error: ${n.message}`)),process.exit(1)});
12
14
  //# sourceMappingURL=cli.mjs.map
package/dist/cli.mjs.map CHANGED
@@ -1 +1 @@
1
- {"version":3,"sources":["../src/cli.ts","../src/extractor.ts"],"sourcesContent":["#!/usr/bin/env node\r\nimport { glob } from 'glob';\r\nimport fs from 'fs';\r\nimport path from 'path';\r\nimport pc from 'picocolors';\r\nimport { extractContent } from './extractor';\r\n\r\n// Simple helper to load .env.local from the current working directory\r\nfunction loadEnv() {\r\n const envPath = path.join(process.cwd(), '.env.local');\r\n if (fs.existsSync(envPath)) {\r\n const envContent = fs.readFileSync(envPath, 'utf8');\r\n envContent.split(/\\r?\\n/).forEach(line => {\r\n const trimmedLine = line.trim();\r\n if (!trimmedLine || trimmedLine.startsWith('#')) return;\r\n const [key, ...valueParts] = trimmedLine.split('=');\r\n if (key && valueParts.length > 0) {\r\n process.env[key.trim()] = valueParts.join('=').trim().replace(/^[\"']|[\"']$/g, '');\r\n }\r\n });\r\n }\r\n}\r\n\r\nasync function main() {\r\n loadEnv();\r\n console.log(pc.cyan('\\n[Onto] Starting Semantic Output Generation...'));\r\n\r\n const cwd = process.cwd();\r\n const nextAppDirDir = path.join(cwd, '.next/server/app');\r\n const ontoPublicDir = path.join(cwd, 'public/.onto');\r\n\r\n if (!fs.existsSync(nextAppDirDir)) {\r\n console.log(pc.yellow(`[Onto] Could not find Next.js app output at ${nextAppDirDir}`));\r\n console.log(pc.yellow(`[Onto] Ensure this is run after \"next build\" and you are using the App Router.`));\r\n return;\r\n }\r\n\r\n // Find all HTML files rendered by Next.js in the app directory\r\n const files = await glob('**/*.html', { cwd: nextAppDirDir });\r\n\r\n if (files.length === 0) {\r\n console.log(pc.yellow(`[Onto] No static HTML files found to process.`));\r\n return;\r\n }\r\n\r\n // Ensure output directory exists\r\n if (!fs.existsSync(ontoPublicDir)) {\r\n fs.mkdirSync(ontoPublicDir, { recursive: true });\r\n }\r\n\r\n let totalOriginalSize = 0;\r\n let totalMarkdownSize = 0;\r\n let totalFilesProcessed = 0;\r\n\r\n for (const file of files) {\r\n const inputPath = path.join(nextAppDirDir, file);\r\n\r\n // We map file path e.g. \"pricing.html\" to \"pricing.md\", or \"blog/post.html\" to \"blog/post.md\"\r\n let outputPathRelative = file.replace(/\\.html$/, '.md');\r\n // If it's a dynamic route page, or purely root index.html\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 const result = extractContent(htmlContent, `/${outputPathRelative.replace(/\\.md$/, '')}`);\r\n\r\n // Ensure specific sub-directory exists (e.g., for blog/post.md)\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\r\n totalOriginalSize += result.stats.originalHtmlSize;\r\n totalMarkdownSize += result.stats.markdownSize;\r\n totalFilesProcessed++;\r\n\r\n const origKb = (result.stats.originalHtmlSize / 1024).toFixed(1);\r\n const mdKb = (result.stats.markdownSize / 1024).toFixed(1);\r\n\r\n // /index.html -> /\r\n let routeName = file.replace(/\\.html$/, '');\r\n if (routeName === 'index') routeName = '/';\r\n else routeName = `/${routeName}`;\r\n\r\n console.log(\r\n pc.green(`✓ Optimized`) +\r\n pc.dim(` ${routeName} `) +\r\n pc.blue(`[${origKb}KB -> ${mdKb}KB]`)\r\n );\r\n } catch (e: any) {\r\n console.error(pc.red(`✗ Failed to process ${file}: ${e.message}`));\r\n }\r\n }\r\n\r\n console.log(\r\n pc.bold(\r\n pc.magenta(`Processed ${totalFilesProcessed} pages. Total Size: ${(totalOriginalSize / 1024).toFixed(1)}KB -> ${(totalMarkdownSize / 1024).toFixed(1)}KB`)\r\n )\r\n );\r\n\r\n // Sync with Onto Control Plane (Premium)\r\n const ONTO_API_KEY = process.env.ONTO_API_KEY;\r\n const DASHBOARD_URL = process.env.ONTO_DASHBOARD_URL || 'https://app.buildonto.dev';\r\n\r\n if (ONTO_API_KEY && totalFilesProcessed > 0) {\r\n console.log(pc.cyan(`[Onto] Syncing manifest with Control Plane [${DASHBOARD_URL}]...`));\r\n try {\r\n const manifest = files.map(file => {\r\n const routeName = file.replace(/\\.html$/, '');\r\n const route = routeName === 'index' ? '/' : `/${routeName}`;\r\n const mdPath = path.join(ontoPublicDir, file.replace(/\\.html$/, '.md'));\r\n return {\r\n route,\r\n filename: `${routeName}.md`,\r\n content: fs.readFileSync(mdPath, 'utf8')\r\n };\r\n });\r\n\r\n const res = await fetch(`${DASHBOARD_URL}/api/files`, {\r\n method: 'POST',\r\n headers: {\r\n 'x-onto-key': ONTO_API_KEY,\r\n 'Content-Type': 'application/json'\r\n },\r\n body: JSON.stringify({ files: manifest })\r\n });\r\n\r\n if (res.ok) {\r\n console.log(pc.green('✓ Control Plane sync successful'));\r\n } else {\r\n const errData = await res.json().catch(() => ({}));\r\n console.log(pc.yellow(`⚠ Control Plane sync skipped: ${errData.error || res.statusText}`));\r\n }\r\n } catch (e: any) {\r\n console.log(pc.yellow(`⚠ Control Plane sync failed: ${e.message}`));\r\n }\r\n }\r\n\r\n console.log(pc.dim(`Edge payloads are ready at /public/.onto/*\\n`));\r\n}\r\n\r\nmain().catch(e => {\r\n console.error(pc.red(`[Onto] Fatal Error: ${e.message}`));\r\n process.exit(1);\r\n});\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":";AACA,OAAS,QAAAA,MAAY,OACrB,OAAOC,MAAQ,KACf,OAAOC,MAAU,OACjB,OAAOC,MAAQ,aCJf,UAAYC,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,CDrGA,SAASC,GAAU,CACf,IAAMC,EAAUC,EAAK,KAAK,QAAQ,IAAI,EAAG,YAAY,EACjDC,EAAG,WAAWF,CAAO,GACFE,EAAG,aAAaF,EAAS,MAAM,EACvC,MAAM,OAAO,EAAE,QAAQG,GAAQ,CACtC,IAAMC,EAAcD,EAAK,KAAK,EAC9B,GAAI,CAACC,GAAeA,EAAY,WAAW,GAAG,EAAG,OACjD,GAAM,CAACC,EAAK,GAAGC,CAAU,EAAIF,EAAY,MAAM,GAAG,EAC9CC,GAAOC,EAAW,OAAS,IAC3B,QAAQ,IAAID,EAAI,KAAK,CAAC,EAAIC,EAAW,KAAK,GAAG,EAAE,KAAK,EAAE,QAAQ,eAAgB,EAAE,EAExF,CAAC,CAET,CAEA,eAAeC,GAAO,CAClBR,EAAQ,EACR,QAAQ,IAAIS,EAAG,KAAK;AAAA,8CAAiD,CAAC,EAEtE,IAAMC,EAAM,QAAQ,IAAI,EAClBC,EAAgBT,EAAK,KAAKQ,EAAK,kBAAkB,EACjDE,EAAgBV,EAAK,KAAKQ,EAAK,cAAc,EAEnD,GAAI,CAACP,EAAG,WAAWQ,CAAa,EAAG,CAC/B,QAAQ,IAAIF,EAAG,OAAO,+CAA+CE,CAAa,EAAE,CAAC,EACrF,QAAQ,IAAIF,EAAG,OAAO,gFAAgF,CAAC,EACvG,MACJ,CAGA,IAAMI,EAAQ,MAAMC,EAAK,YAAa,CAAE,IAAKH,CAAc,CAAC,EAE5D,GAAIE,EAAM,SAAW,EAAG,CACpB,QAAQ,IAAIJ,EAAG,OAAO,+CAA+C,CAAC,EACtE,MACJ,CAGKN,EAAG,WAAWS,CAAa,GAC5BT,EAAG,UAAUS,EAAe,CAAE,UAAW,EAAK,CAAC,EAGnD,IAAIG,EAAoB,EACpBC,EAAoB,EACpBC,EAAsB,EAE1B,QAAWC,KAAQL,EAAO,CACtB,IAAMM,EAAYjB,EAAK,KAAKS,EAAeO,CAAI,EAG3CE,EAAqBF,EAAK,QAAQ,UAAW,KAAK,EAEhDG,EAAanB,EAAK,KAAKU,EAAeQ,CAAkB,EAE9D,GAAI,CACA,IAAME,EAAcnB,EAAG,aAAagB,EAAW,MAAM,EAE/CI,EAASC,EAAeF,EAAa,IAAIF,EAAmB,QAAQ,QAAS,EAAE,CAAC,EAAE,EAGlFK,EAAYvB,EAAK,QAAQmB,CAAU,EACpClB,EAAG,WAAWsB,CAAS,GACxBtB,EAAG,UAAUsB,EAAW,CAAE,UAAW,EAAK,CAAC,EAG/CtB,EAAG,cAAckB,EAAYE,EAAO,SAAU,MAAM,EAEpDR,GAAqBQ,EAAO,MAAM,iBAClCP,GAAqBO,EAAO,MAAM,aAClCN,IAEA,IAAMS,GAAUH,EAAO,MAAM,iBAAmB,MAAM,QAAQ,CAAC,EACzDI,GAAQJ,EAAO,MAAM,aAAe,MAAM,QAAQ,CAAC,EAGrDK,EAAYV,EAAK,QAAQ,UAAW,EAAE,EACtCU,IAAc,QAASA,EAAY,IAClCA,EAAY,IAAIA,CAAS,GAE9B,QAAQ,IACJnB,EAAG,MAAM,kBAAa,EACtBA,EAAG,IAAI,IAAImB,CAAS,GAAG,EACvBnB,EAAG,KAAK,IAAIiB,CAAM,SAASC,CAAI,KAAK,CACxC,CACJ,OAASE,EAAQ,CACb,QAAQ,MAAMpB,EAAG,IAAI,4BAAuBS,CAAI,KAAKW,EAAE,OAAO,EAAE,CAAC,CACrE,CACJ,CAEA,QAAQ,IACJpB,EAAG,KACCA,EAAG,QAAQ,aAAaQ,CAAmB,wBAAwBF,EAAoB,MAAM,QAAQ,CAAC,CAAC,UAAUC,EAAoB,MAAM,QAAQ,CAAC,CAAC,IAAI,CAC7J,CACJ,EAGA,IAAMc,EAAe,QAAQ,IAAI,aAC3BC,EAAgB,QAAQ,IAAI,oBAAsB,4BAExD,GAAID,GAAgBb,EAAsB,EAAG,CACzC,QAAQ,IAAIR,EAAG,KAAK,+CAA+CsB,CAAa,MAAM,CAAC,EACvF,GAAI,CACA,IAAMC,EAAWnB,EAAM,IAAIK,GAAQ,CAC/B,IAAMU,EAAYV,EAAK,QAAQ,UAAW,EAAE,EACtCe,EAAQL,IAAc,QAAU,IAAM,IAAIA,CAAS,GACnDM,EAAShC,EAAK,KAAKU,EAAeM,EAAK,QAAQ,UAAW,KAAK,CAAC,EACtE,MAAO,CACH,MAAAe,EACA,SAAU,GAAGL,CAAS,MACtB,QAASzB,EAAG,aAAa+B,EAAQ,MAAM,CAC3C,CACJ,CAAC,EAEKC,EAAM,MAAM,MAAM,GAAGJ,CAAa,aAAc,CAClD,OAAQ,OACR,QAAS,CACL,aAAcD,EACd,eAAgB,kBACpB,EACA,KAAM,KAAK,UAAU,CAAE,MAAOE,CAAS,CAAC,CAC5C,CAAC,EAED,GAAIG,EAAI,GACJ,QAAQ,IAAI1B,EAAG,MAAM,sCAAiC,CAAC,MACpD,CACH,IAAM2B,EAAU,MAAMD,EAAI,KAAK,EAAE,MAAM,KAAO,CAAC,EAAE,EACjD,QAAQ,IAAI1B,EAAG,OAAO,sCAAiC2B,EAAQ,OAASD,EAAI,UAAU,EAAE,CAAC,CAC7F,CACJ,OAASN,EAAQ,CACb,QAAQ,IAAIpB,EAAG,OAAO,qCAAgCoB,EAAE,OAAO,EAAE,CAAC,CACtE,CACJ,CAEA,QAAQ,IAAIpB,EAAG,IAAI;AAAA,CAA8C,CAAC,CACtE,CAEAD,EAAK,EAAE,MAAMqB,GAAK,CACd,QAAQ,MAAMpB,EAAG,IAAI,uBAAuBoB,EAAE,OAAO,EAAE,CAAC,EACxD,QAAQ,KAAK,CAAC,CAClB,CAAC","names":["glob","fs","path","pc","cheerio","TurndownService","turndownService","extractContent","html","sourceUrl","originalSize","$","title","description","jsonLdScripts","_","el","raw","parsed","contentHtml","markdown","finalMarkdown","j","markdownSize","tokenReductionRatio","loadEnv","envPath","path","fs","line","trimmedLine","key","valueParts","main","pc","cwd","nextAppDirDir","ontoPublicDir","files","glob","totalOriginalSize","totalMarkdownSize","totalFilesProcessed","file","inputPath","outputPathRelative","outputPath","htmlContent","result","extractContent","outputDir","origKb","mdKb","routeName","e","ONTO_API_KEY","DASHBOARD_URL","manifest","route","mdPath","res","errData"]}
1
+ {"version":3,"sources":["../src/cli.ts","../src/extractor.ts","../src/config.ts"],"sourcesContent":["#!/usr/bin/env node\r\nimport { glob } from 'glob';\r\nimport fs from 'fs';\r\nimport path from 'path';\r\nimport pc from 'picocolors';\r\nimport { extractContent } from './extractor';\r\nimport { generateLlmsTxt, OntoConfig } from './config';\r\n\r\nasync function loadOntoConfig(): Promise<OntoConfig | null> {\r\n try {\r\n const configPath = path.join(process.cwd(), 'onto.config');\r\n // Node.js dynamic import\r\n const config = await import('file://' + configPath.replace(/\\\\/g, '/') + '.ts');\r\n return config.default || config;\r\n } catch (error) {\r\n try {\r\n const configPath = path.join(process.cwd(), 'onto.config');\r\n const config = await import('file://' + configPath.replace(/\\\\/g, '/') + '.js');\r\n return config.default || config;\r\n } catch (e) {\r\n return null;\r\n }\r\n }\r\n}\r\n\r\n// Simple helper to load .env.local from the current working directory\r\nfunction loadEnv() {\r\n const envPath = path.join(process.cwd(), '.env.local');\r\n if (fs.existsSync(envPath)) {\r\n const envContent = fs.readFileSync(envPath, 'utf8');\r\n envContent.split(/\\r?\\n/).forEach(line => {\r\n const trimmedLine = line.trim();\r\n if (!trimmedLine || trimmedLine.startsWith('#')) return;\r\n const [key, ...valueParts] = trimmedLine.split('=');\r\n if (key && valueParts.length > 0) {\r\n process.env[key.trim()] = valueParts.join('=').trim().replace(/^[\"']|[\"']$/g, '');\r\n }\r\n });\r\n }\r\n}\r\n\r\nasync function main() {\r\n loadEnv();\r\n console.log(pc.cyan('\\n[Onto] Starting Semantic Output Generation...'));\r\n\r\n const cwd = process.cwd();\r\n const nextAppDirDir = path.join(cwd, '.next/server/app');\r\n const ontoPublicDir = path.join(cwd, 'public/.onto');\r\n\r\n if (!fs.existsSync(nextAppDirDir)) {\r\n console.log(pc.yellow(`[Onto] Could not find Next.js app output at ${nextAppDirDir}`));\r\n console.log(pc.yellow(`[Onto] Ensure this is run after \"next build\" and you are using the App Router.`));\r\n return;\r\n }\r\n\r\n // Find all HTML files rendered by Next.js in the app directory\r\n const files = await glob('**/*.html', { cwd: nextAppDirDir });\r\n\r\n if (files.length === 0) {\r\n console.log(pc.yellow(`[Onto] No static HTML files found to process.`));\r\n return;\r\n }\r\n\r\n // Ensure output directory exists\r\n if (!fs.existsSync(ontoPublicDir)) {\r\n fs.mkdirSync(ontoPublicDir, { recursive: true });\r\n }\r\n\r\n let totalOriginalSize = 0;\r\n let totalMarkdownSize = 0;\r\n let totalFilesProcessed = 0;\r\n\r\n for (const file of files) {\r\n const inputPath = path.join(nextAppDirDir, file);\r\n\r\n // We map file path e.g. \"pricing.html\" to \"pricing.md\", or \"blog/post.html\" to \"blog/post.md\"\r\n let outputPathRelative = file.replace(/\\.html$/, '.md');\r\n // If it's a dynamic route page, or purely root index.html\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 const result = extractContent(htmlContent, `/${outputPathRelative.replace(/\\.md$/, '')}`);\r\n\r\n // Ensure specific sub-directory exists (e.g., for blog/post.md)\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\r\n totalOriginalSize += result.stats.originalHtmlSize;\r\n totalMarkdownSize += result.stats.markdownSize;\r\n totalFilesProcessed++;\r\n\r\n const origKb = (result.stats.originalHtmlSize / 1024).toFixed(1);\r\n const mdKb = (result.stats.markdownSize / 1024).toFixed(1);\r\n\r\n // /index.html -> /\r\n let routeName = file.replace(/\\.html$/, '');\r\n if (routeName === 'index') routeName = '/';\r\n else routeName = `/${routeName}`;\r\n\r\n console.log(\r\n pc.green(`✓ Optimized`) +\r\n pc.dim(` ${routeName} `) +\r\n pc.blue(`[${origKb}KB -> ${mdKb}KB]`)\r\n );\r\n } catch (e: any) {\r\n console.error(pc.red(`✗ Failed to process ${file}: ${e.message}`));\r\n }\r\n }\r\n\r\n console.log(\r\n pc.bold(\r\n pc.magenta(`Processed ${totalFilesProcessed} pages. Total Size: ${(totalOriginalSize / 1024).toFixed(1)}KB -> ${(totalMarkdownSize / 1024).toFixed(1)}KB`)\r\n )\r\n );\r\n\r\n // Sync with Onto Control Plane (Premium)\r\n const ONTO_API_KEY = process.env.ONTO_API_KEY;\r\n const DASHBOARD_URL = process.env.ONTO_DASHBOARD_URL || 'https://app.buildonto.dev';\r\n\r\n if (ONTO_API_KEY && totalFilesProcessed > 0) {\r\n console.log(pc.cyan(`[Onto] Syncing manifest with Control Plane [${DASHBOARD_URL}]...`));\r\n try {\r\n const manifest = files.map(file => {\r\n const routeName = file.replace(/\\.html$/, '');\r\n const route = routeName === 'index' ? '/' : `/${routeName}`;\r\n const mdPath = path.join(ontoPublicDir, file.replace(/\\.html$/, '.md'));\r\n return {\r\n route,\r\n filename: `${routeName}.md`,\r\n content: fs.readFileSync(mdPath, 'utf8')\r\n };\r\n });\r\n\r\n const res = await fetch(`${DASHBOARD_URL}/api/files`, {\r\n method: 'POST',\r\n headers: {\r\n 'x-onto-key': ONTO_API_KEY,\r\n 'Content-Type': 'application/json'\r\n },\r\n body: JSON.stringify({ files: manifest })\r\n });\r\n\r\n if (res.ok) {\r\n console.log(pc.green('✓ Control Plane sync successful'));\r\n } else {\r\n const errData = await res.json().catch(() => ({}));\r\n console.log(pc.yellow(`⚠ Control Plane sync skipped: ${errData.error || res.statusText}`));\r\n }\r\n } catch (e: any) {\r\n console.log(pc.yellow(`⚠ Control Plane sync failed: ${e.message}`));\r\n }\r\n }\r\n\r\n // --- Generate llms.txt manifest ---\r\n const config = await loadOntoConfig();\r\n if (config) {\r\n const llmsTxtContent = generateLlmsTxt(config);\r\n const llmsTxtPath = path.join(cwd, 'public/llms.txt');\r\n \r\n // Ensure public dir exists\r\n const publicDir = path.join(cwd, 'public');\r\n if (!fs.existsSync(publicDir)) {\r\n fs.mkdirSync(publicDir, { recursive: true });\r\n }\r\n\r\n fs.writeFileSync(llmsTxtPath, llmsTxtContent, 'utf8');\r\n console.log(pc.green('✓ Generated') + pc.dim(' /llms.txt'));\r\n }\r\n\r\n console.log(pc.dim(`Edge payloads are ready at /public/.onto/*\\n`));\r\n}\r\n\r\nmain().catch(e => {\r\n console.error(pc.red(`[Onto] Fatal Error: ${e.message}`));\r\n process.exit(1);\r\n});\r\n","import * as cheerio from 'cheerio';\r\nimport TurndownService from 'turndown';\r\n\r\nconst turndownService = new TurndownService({\r\n headingStyle: 'atx',\r\n codeBlockStyle: 'fenced',\r\n});\r\n\r\n// Configure turndown to keep some layout or handle semantic tags differently if needed\r\n\r\nexport interface ExtractionResult {\r\n markdown: string;\r\n metadata: {\r\n title: string;\r\n description: string;\r\n jsonLd: any[];\r\n };\r\n stats: {\r\n originalHtmlSize: number;\r\n markdownSize: number;\r\n tokenReductionRatio: number;\r\n };\r\n}\r\n\r\n/**\r\n * Extracts pure semantic markdown and metadata from rendered Next.js HTML strings.\r\n * @param html The raw HTML string.\r\n * @param sourceUrl (Optional) the URL this was generated from, to attach as metadata.\r\n * @returns {ExtractionResult} The extracted payload.\r\n */\r\nexport function extractContent(html: string, sourceUrl: string = 'Generated Output'): ExtractionResult {\r\n const originalSize = html.length;\r\n\r\n const $ = cheerio.load(html);\r\n\r\n // 1. Extract Metadata BEFORE removing structure\r\n const title = $('title').text() || $('h1').first().text() || 'Untitled Page';\r\n const description = $('meta[name=\"description\"]').attr('content') || 'No description found.';\r\n\r\n const jsonLdScripts: any[] = [];\r\n $('script[type=\"application/ld+json\"]').each((_, el) => {\r\n try {\r\n const raw = $(el).html() || '';\r\n const parsed = JSON.parse(raw);\r\n jsonLdScripts.push(parsed);\r\n } catch {\r\n // ignore bad json\r\n }\r\n });\r\n\r\n // 2. Strip noise (React boilerplate, styles, unnecessary tags)\r\n $('script, style, noscript, iframe, svg, nav, footer, meta, link, header').remove();\r\n\r\n // Optionally remove typical Next.js hidden wrappers if they don't contain real content.\r\n // Next.js uses <div id=\"__next\"> but we mostly just want semantic content.\r\n\r\n // 3. Find the entry point for content\r\n // Prefer <main> or <article> over <body>\r\n let contentHtml = '';\r\n if ($('main').length > 0) {\r\n contentHtml = $('main').html() || '';\r\n } else if ($('article').length > 0) {\r\n contentHtml = $('article').html() || '';\r\n } else {\r\n contentHtml = $('body').html() || '';\r\n }\r\n\r\n // 4. Convert to Markdown\r\n let markdown = turndownService.turndown(contentHtml);\r\n\r\n // 5. Optionally inject Metadata header\r\n const headerLines = [\r\n `# ${title}`,\r\n `> ${description}`,\r\n ``,\r\n `**Source:** ${sourceUrl}`,\r\n `**Extracted:** ${new Date().toISOString()}`,\r\n ``,\r\n `---`,\r\n ``\r\n ];\r\n\r\n let finalMarkdown = headerLines.join('\\n') + markdown;\r\n\r\n // Add JSON-LD section if exists\r\n if (jsonLdScripts.length > 0) {\r\n finalMarkdown += '\\n\\n---\\n## Structured Data (JSON-LD)\\n```json\\n';\r\n jsonLdScripts.forEach(j => {\r\n finalMarkdown += JSON.stringify(j, null, 2) + '\\n';\r\n });\r\n finalMarkdown += '```\\n';\r\n }\r\n\r\n const markdownSize = finalMarkdown.length;\r\n const tokenReductionRatio = originalSize > 0 ? ((originalSize - markdownSize) / originalSize) * 100 : 0;\r\n\r\n return {\r\n markdown: finalMarkdown,\r\n metadata: {\r\n title,\r\n description,\r\n jsonLd: jsonLdScripts\r\n },\r\n stats: {\r\n originalHtmlSize: originalSize,\r\n markdownSize,\r\n tokenReductionRatio\r\n }\r\n };\r\n}\r\n\r\nexport async function generateStaticPayloads(nextAppDirDir: string, ontoPublicDir: string) {\r\n const fs = await import('fs');\r\n const path = await import('path');\r\n const { glob } = await import('glob');\r\n\r\n if (!fs.existsSync(nextAppDirDir)) {\r\n return;\r\n }\r\n\r\n const files = await glob('**/*.html', { cwd: nextAppDirDir });\r\n if (files.length === 0) return;\r\n\r\n if (!fs.existsSync(ontoPublicDir)) {\r\n fs.mkdirSync(ontoPublicDir, { recursive: true });\r\n }\r\n\r\n let totalFilesProcessed = 0;\r\n\r\n for (const file of files) {\r\n const inputPath = path.join(nextAppDirDir, file);\r\n const outputPathRelative = file.replace(/\\.html$/, '.md');\r\n const outputPath = path.join(ontoPublicDir, outputPathRelative);\r\n\r\n try {\r\n const htmlContent = fs.readFileSync(inputPath, 'utf8');\r\n\r\n let routeName = file.replace(/\\.html$/, '');\r\n if (routeName === 'index') routeName = '/';\r\n else routeName = `/${routeName}`;\r\n\r\n const result = extractContent(htmlContent, routeName);\r\n\r\n const outputDir = path.dirname(outputPath);\r\n if (!fs.existsSync(outputDir)) {\r\n fs.mkdirSync(outputDir, { recursive: true });\r\n }\r\n\r\n fs.writeFileSync(outputPath, result.markdown, 'utf8');\r\n totalFilesProcessed++;\r\n } catch (e: any) {\r\n console.error(`[Onto] Failed to process ${file}: ${e.message}`);\r\n }\r\n }\r\n console.log(`[Onto] Successfully generated ${totalFilesProcessed} semantic markdown endpoints.`);\r\n}\r\n","/**\r\n * Configuration schema for onto.config.ts\r\n * Used to dynamically generate llms.txt and other AI discovery files\r\n */\r\n\r\nexport type PageType = 'scoring' | 'about' | 'default';\r\n\r\nexport interface OntoRoute {\r\n /**\r\n * The URL path (e.g., '/docs', '/api/reference')\r\n */\r\n path: string;\r\n /**\r\n * Description of what this route contains\r\n */\r\n description: string;\r\n /**\r\n * Optional: Page type for automatic JSON-LD schema injection\r\n * - 'scoring': Injects Methodology schema with AIO scoring weights (40/35/25)\r\n * - 'about': Injects Organization/AboutPage schema\r\n * - 'default': No automatic schema injection\r\n */\r\n pageType?: PageType;\r\n}\r\n\r\nexport interface OntoConfig {\r\n /**\r\n * The name of your project or site (required)\r\n * Used as the H1 heading in llms.txt\r\n */\r\n name: string;\r\n\r\n /**\r\n * A short summary of your project (required)\r\n * Displayed as a blockquote in llms.txt\r\n * Should contain key information necessary for understanding the rest of the file\r\n */\r\n summary: string;\r\n\r\n /**\r\n * The base URL of your site (e.g., 'https://example.com')\r\n */\r\n baseUrl: string;\r\n\r\n /**\r\n * Optional: Additional sections to include in llms.txt\r\n * Each section can contain any markdown content\r\n */\r\n sections?: {\r\n heading: string;\r\n content: string;\r\n }[];\r\n\r\n /**\r\n * Key routes that AI agents should know about\r\n * These will be formatted as a markdown list in llms.txt\r\n */\r\n routes?: OntoRoute[];\r\n\r\n /**\r\n * Optional: Links to external resources (documentation, API references, etc.)\r\n */\r\n externalLinks?: {\r\n title: string;\r\n url: string;\r\n description?: string;\r\n }[];\r\n\r\n /**\r\n * Optional: Organization information for JSON-LD schemas\r\n */\r\n organization?: {\r\n name: string;\r\n description?: string;\r\n url?: string;\r\n logo?: string;\r\n foundingDate?: string;\r\n };\r\n}\r\n\r\n/**\r\n * Generate llms.txt content from OntoConfig\r\n * Follows the llms.txt specification:\r\n * - H1 with project name\r\n * - Blockquote with summary\r\n * - Additional markdown sections\r\n */\r\nexport function generateLlmsTxt(config: OntoConfig): string {\r\n const lines: string[] = [];\r\n\r\n // H1: Project name (required)\r\n lines.push(`# ${config.name}`);\r\n lines.push('');\r\n\r\n // Blockquote: Summary (required)\r\n lines.push(`> ${config.summary}`);\r\n lines.push('');\r\n\r\n // Key Routes section (if provided)\r\n if (config.routes && config.routes.length > 0) {\r\n lines.push('## Key Routes');\r\n lines.push('');\r\n for (const route of config.routes) {\r\n const fullUrl = `${config.baseUrl}${route.path}`;\r\n lines.push(`- [${route.path}](${fullUrl}): ${route.description}`);\r\n }\r\n lines.push('');\r\n }\r\n\r\n // External Links section (if provided)\r\n if (config.externalLinks && config.externalLinks.length > 0) {\r\n lines.push('## Resources');\r\n lines.push('');\r\n for (const link of config.externalLinks) {\r\n if (link.description) {\r\n lines.push(`- [${link.title}](${link.url}): ${link.description}`);\r\n } else {\r\n lines.push(`- [${link.title}](${link.url})`);\r\n }\r\n }\r\n lines.push('');\r\n }\r\n\r\n // Custom sections (if provided)\r\n if (config.sections && config.sections.length > 0) {\r\n for (const section of config.sections) {\r\n lines.push(`## ${section.heading}`);\r\n lines.push('');\r\n lines.push(section.content);\r\n lines.push('');\r\n }\r\n }\r\n\r\n return lines.join('\\n').trim() + '\\n';\r\n}\r\n"],"mappings":";AACA,OAAS,QAAAA,MAAY,OACrB,OAAOC,MAAQ,KACf,OAAOC,MAAU,OACjB,OAAOC,MAAQ,aCJf,UAAYC,MAAa,UACzB,OAAOC,MAAqB,WAE5B,IAAMC,EAAkB,IAAID,EAAgB,CACxC,aAAc,MACd,eAAgB,QACpB,CAAC,EAwBM,SAASE,EAAeC,EAAcC,EAAoB,mBAAsC,CACnG,IAAMC,EAAeF,EAAK,OAEpBG,EAAY,OAAKH,CAAI,EAGrBI,EAAQD,EAAE,OAAO,EAAE,KAAK,GAAKA,EAAE,IAAI,EAAE,MAAM,EAAE,KAAK,GAAK,gBACvDE,EAAcF,EAAE,0BAA0B,EAAE,KAAK,SAAS,GAAK,wBAE/DG,EAAuB,CAAC,EAC9BH,EAAE,oCAAoC,EAAE,KAAK,CAACI,EAAGC,IAAO,CACpD,GAAI,CACA,IAAMC,EAAMN,EAAEK,CAAE,EAAE,KAAK,GAAK,GACtBE,EAAS,KAAK,MAAMD,CAAG,EAC7BH,EAAc,KAAKI,CAAM,CAC7B,MAAQ,CAER,CACJ,CAAC,EAGDP,EAAE,uEAAuE,EAAE,OAAO,EAOlF,IAAIQ,EAAc,GACdR,EAAE,MAAM,EAAE,OAAS,EACnBQ,EAAcR,EAAE,MAAM,EAAE,KAAK,GAAK,GAC3BA,EAAE,SAAS,EAAE,OAAS,EAC7BQ,EAAcR,EAAE,SAAS,EAAE,KAAK,GAAK,GAErCQ,EAAcR,EAAE,MAAM,EAAE,KAAK,GAAK,GAItC,IAAIS,EAAWd,EAAgB,SAASa,CAAW,EAc/CE,EAXgB,CAChB,KAAKT,CAAK,GACV,KAAKC,CAAW,GAChB,GACA,eAAeJ,CAAS,GACxB,kBAAkB,IAAI,KAAK,EAAE,YAAY,CAAC,GAC1C,GACA,MACA,EACJ,EAEgC,KAAK;AAAA,CAAI,EAAIW,EAGzCN,EAAc,OAAS,IACvBO,GAAiB;AAAA;AAAA;AAAA;AAAA;AAAA,EACjBP,EAAc,QAAQQ,GAAK,CACvBD,GAAiB,KAAK,UAAUC,EAAG,KAAM,CAAC,EAAI;AAAA,CAClD,CAAC,EACDD,GAAiB,SAGrB,IAAME,EAAeF,EAAc,OAC7BG,EAAsBd,EAAe,GAAMA,EAAea,GAAgBb,EAAgB,IAAM,EAEtG,MAAO,CACH,SAAUW,EACV,SAAU,CACN,MAAAT,EACA,YAAAC,EACA,OAAQC,CACZ,EACA,MAAO,CACH,iBAAkBJ,EAClB,aAAAa,EACA,oBAAAC,CACJ,CACJ,CACJ,CCtBO,SAASC,EAAgBC,EAA4B,CAC1D,IAAMC,EAAkB,CAAC,EAWzB,GARAA,EAAM,KAAK,KAAKD,EAAO,IAAI,EAAE,EAC7BC,EAAM,KAAK,EAAE,EAGbA,EAAM,KAAK,KAAKD,EAAO,OAAO,EAAE,EAChCC,EAAM,KAAK,EAAE,EAGTD,EAAO,QAAUA,EAAO,OAAO,OAAS,EAAG,CAC7CC,EAAM,KAAK,eAAe,EAC1BA,EAAM,KAAK,EAAE,EACb,QAAWC,KAASF,EAAO,OAAQ,CACjC,IAAMG,EAAU,GAAGH,EAAO,OAAO,GAAGE,EAAM,IAAI,GAC9CD,EAAM,KAAK,MAAMC,EAAM,IAAI,KAAKC,CAAO,MAAMD,EAAM,WAAW,EAAE,CAClE,CACAD,EAAM,KAAK,EAAE,CACf,CAGA,GAAID,EAAO,eAAiBA,EAAO,cAAc,OAAS,EAAG,CAC3DC,EAAM,KAAK,cAAc,EACzBA,EAAM,KAAK,EAAE,EACb,QAAWG,KAAQJ,EAAO,cACpBI,EAAK,YACPH,EAAM,KAAK,MAAMG,EAAK,KAAK,KAAKA,EAAK,GAAG,MAAMA,EAAK,WAAW,EAAE,EAEhEH,EAAM,KAAK,MAAMG,EAAK,KAAK,KAAKA,EAAK,GAAG,GAAG,EAG/CH,EAAM,KAAK,EAAE,CACf,CAGA,GAAID,EAAO,UAAYA,EAAO,SAAS,OAAS,EAC9C,QAAWK,KAAWL,EAAO,SAC3BC,EAAM,KAAK,MAAMI,EAAQ,OAAO,EAAE,EAClCJ,EAAM,KAAK,EAAE,EACbA,EAAM,KAAKI,EAAQ,OAAO,EAC1BJ,EAAM,KAAK,EAAE,EAIjB,OAAOA,EAAM,KAAK;AAAA,CAAI,EAAE,KAAK,EAAI;AAAA,CACnC,CF9HA,eAAeK,GAA6C,CACxD,GAAI,CAGA,IAAMC,EAAS,MAAM,OAAO,UAFTC,EAAK,KAAK,QAAQ,IAAI,EAAG,aAAa,EAEN,QAAQ,MAAO,GAAG,EAAI,OACzE,OAAOD,EAAO,SAAWA,CAC7B,MAAgB,CACZ,GAAI,CAEA,IAAMA,EAAS,MAAM,OAAO,UADTC,EAAK,KAAK,QAAQ,IAAI,EAAG,aAAa,EACN,QAAQ,MAAO,GAAG,EAAI,OACzE,OAAOD,EAAO,SAAWA,CAC7B,MAAY,CACR,OAAO,IACX,CACJ,CACJ,CAGA,SAASE,GAAU,CACf,IAAMC,EAAUF,EAAK,KAAK,QAAQ,IAAI,EAAG,YAAY,EACjDG,EAAG,WAAWD,CAAO,GACFC,EAAG,aAAaD,EAAS,MAAM,EACvC,MAAM,OAAO,EAAE,QAAQE,GAAQ,CACtC,IAAMC,EAAcD,EAAK,KAAK,EAC9B,GAAI,CAACC,GAAeA,EAAY,WAAW,GAAG,EAAG,OACjD,GAAM,CAACC,EAAK,GAAGC,CAAU,EAAIF,EAAY,MAAM,GAAG,EAC9CC,GAAOC,EAAW,OAAS,IAC3B,QAAQ,IAAID,EAAI,KAAK,CAAC,EAAIC,EAAW,KAAK,GAAG,EAAE,KAAK,EAAE,QAAQ,eAAgB,EAAE,EAExF,CAAC,CAET,CAEA,eAAeC,GAAO,CAClBP,EAAQ,EACR,QAAQ,IAAIQ,EAAG,KAAK;AAAA,8CAAiD,CAAC,EAEtE,IAAMC,EAAM,QAAQ,IAAI,EAClBC,EAAgBX,EAAK,KAAKU,EAAK,kBAAkB,EACjDE,EAAgBZ,EAAK,KAAKU,EAAK,cAAc,EAEnD,GAAI,CAACP,EAAG,WAAWQ,CAAa,EAAG,CAC/B,QAAQ,IAAIF,EAAG,OAAO,+CAA+CE,CAAa,EAAE,CAAC,EACrF,QAAQ,IAAIF,EAAG,OAAO,gFAAgF,CAAC,EACvG,MACJ,CAGA,IAAMI,EAAQ,MAAMC,EAAK,YAAa,CAAE,IAAKH,CAAc,CAAC,EAE5D,GAAIE,EAAM,SAAW,EAAG,CACpB,QAAQ,IAAIJ,EAAG,OAAO,+CAA+C,CAAC,EACtE,MACJ,CAGKN,EAAG,WAAWS,CAAa,GAC5BT,EAAG,UAAUS,EAAe,CAAE,UAAW,EAAK,CAAC,EAGnD,IAAIG,EAAoB,EACpBC,EAAoB,EACpBC,EAAsB,EAE1B,QAAWC,KAAQL,EAAO,CACtB,IAAMM,EAAYnB,EAAK,KAAKW,EAAeO,CAAI,EAG3CE,EAAqBF,EAAK,QAAQ,UAAW,KAAK,EAEhDG,EAAarB,EAAK,KAAKY,EAAeQ,CAAkB,EAE9D,GAAI,CACA,IAAME,EAAcnB,EAAG,aAAagB,EAAW,MAAM,EAE/CI,EAASC,EAAeF,EAAa,IAAIF,EAAmB,QAAQ,QAAS,EAAE,CAAC,EAAE,EAGlFK,EAAYzB,EAAK,QAAQqB,CAAU,EACpClB,EAAG,WAAWsB,CAAS,GACxBtB,EAAG,UAAUsB,EAAW,CAAE,UAAW,EAAK,CAAC,EAG/CtB,EAAG,cAAckB,EAAYE,EAAO,SAAU,MAAM,EAEpDR,GAAqBQ,EAAO,MAAM,iBAClCP,GAAqBO,EAAO,MAAM,aAClCN,IAEA,IAAMS,GAAUH,EAAO,MAAM,iBAAmB,MAAM,QAAQ,CAAC,EACzDI,GAAQJ,EAAO,MAAM,aAAe,MAAM,QAAQ,CAAC,EAGrDK,EAAYV,EAAK,QAAQ,UAAW,EAAE,EACtCU,IAAc,QAASA,EAAY,IAClCA,EAAY,IAAIA,CAAS,GAE9B,QAAQ,IACJnB,EAAG,MAAM,kBAAa,EACtBA,EAAG,IAAI,IAAImB,CAAS,GAAG,EACvBnB,EAAG,KAAK,IAAIiB,CAAM,SAASC,CAAI,KAAK,CACxC,CACJ,OAASE,EAAQ,CACb,QAAQ,MAAMpB,EAAG,IAAI,4BAAuBS,CAAI,KAAKW,EAAE,OAAO,EAAE,CAAC,CACrE,CACJ,CAEA,QAAQ,IACJpB,EAAG,KACCA,EAAG,QAAQ,aAAaQ,CAAmB,wBAAwBF,EAAoB,MAAM,QAAQ,CAAC,CAAC,UAAUC,EAAoB,MAAM,QAAQ,CAAC,CAAC,IAAI,CAC7J,CACJ,EAGA,IAAMc,EAAe,QAAQ,IAAI,aAC3BC,EAAgB,QAAQ,IAAI,oBAAsB,4BAExD,GAAID,GAAgBb,EAAsB,EAAG,CACzC,QAAQ,IAAIR,EAAG,KAAK,+CAA+CsB,CAAa,MAAM,CAAC,EACvF,GAAI,CACA,IAAMC,EAAWnB,EAAM,IAAIK,GAAQ,CAC/B,IAAMU,EAAYV,EAAK,QAAQ,UAAW,EAAE,EACtCe,EAAQL,IAAc,QAAU,IAAM,IAAIA,CAAS,GACnDM,EAASlC,EAAK,KAAKY,EAAeM,EAAK,QAAQ,UAAW,KAAK,CAAC,EACtE,MAAO,CACH,MAAAe,EACA,SAAU,GAAGL,CAAS,MACtB,QAASzB,EAAG,aAAa+B,EAAQ,MAAM,CAC3C,CACJ,CAAC,EAEKC,EAAM,MAAM,MAAM,GAAGJ,CAAa,aAAc,CAClD,OAAQ,OACR,QAAS,CACL,aAAcD,EACd,eAAgB,kBACpB,EACA,KAAM,KAAK,UAAU,CAAE,MAAOE,CAAS,CAAC,CAC5C,CAAC,EAED,GAAIG,EAAI,GACJ,QAAQ,IAAI1B,EAAG,MAAM,sCAAiC,CAAC,MACpD,CACH,IAAM2B,EAAU,MAAMD,EAAI,KAAK,EAAE,MAAM,KAAO,CAAC,EAAE,EACjD,QAAQ,IAAI1B,EAAG,OAAO,sCAAiC2B,EAAQ,OAASD,EAAI,UAAU,EAAE,CAAC,CAC7F,CACJ,OAASN,EAAQ,CACb,QAAQ,IAAIpB,EAAG,OAAO,qCAAgCoB,EAAE,OAAO,EAAE,CAAC,CACtE,CACJ,CAGA,IAAM9B,EAAS,MAAMD,EAAe,EACpC,GAAIC,EAAQ,CACR,IAAMsC,EAAiBC,EAAgBvC,CAAM,EACvCwC,EAAcvC,EAAK,KAAKU,EAAK,iBAAiB,EAG9C8B,EAAYxC,EAAK,KAAKU,EAAK,QAAQ,EACpCP,EAAG,WAAWqC,CAAS,GACxBrC,EAAG,UAAUqC,EAAW,CAAE,UAAW,EAAK,CAAC,EAG/CrC,EAAG,cAAcoC,EAAaF,EAAgB,MAAM,EACpD,QAAQ,IAAI5B,EAAG,MAAM,kBAAa,EAAIA,EAAG,IAAI,YAAY,CAAC,CAC9D,CAEA,QAAQ,IAAIA,EAAG,IAAI;AAAA,CAA8C,CAAC,CACtE,CAEAD,EAAK,EAAE,MAAMqB,GAAK,CACd,QAAQ,MAAMpB,EAAG,IAAI,uBAAuBoB,EAAE,OAAO,EAAE,CAAC,EACxD,QAAQ,KAAK,CAAC,CAClB,CAAC","names":["glob","fs","path","pc","cheerio","TurndownService","turndownService","extractContent","html","sourceUrl","originalSize","$","title","description","jsonLdScripts","_","el","raw","parsed","contentHtml","markdown","finalMarkdown","j","markdownSize","tokenReductionRatio","generateLlmsTxt","config","lines","route","fullUrl","link","section","loadOntoConfig","config","path","loadEnv","envPath","fs","line","trimmedLine","key","valueParts","main","pc","cwd","nextAppDirDir","ontoPublicDir","files","glob","totalOriginalSize","totalMarkdownSize","totalFilesProcessed","file","inputPath","outputPathRelative","outputPath","htmlContent","result","extractContent","outputDir","origKb","mdKb","routeName","e","ONTO_API_KEY","DASHBOARD_URL","manifest","route","mdPath","res","errData","llmsTxtContent","generateLlmsTxt","llmsTxtPath","publicDir"]}
package/dist/config.js CHANGED
@@ -1,4 +1,4 @@
1
- "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(`
1
+ "use strict";var i=Object.defineProperty;var o=Object.getOwnPropertyDescriptor;var u=Object.getOwnPropertyNames;var p=Object.prototype.hasOwnProperty;var a=(s,t)=>{for(var e in t)i(s,e,{get:t[e],enumerable:!0})},h=(s,t,e,r)=>{if(t&&typeof t=="object"||typeof t=="function")for(let n of u(t))!p.call(s,n)&&n!==e&&i(s,n,{get:()=>t[n],enumerable:!(r=o(t,n))||r.enumerable});return s};var l=s=>h(i({},"__esModule",{value:!0}),s);var $={};a($,{generateLlmsTxt:()=>g});module.exports=l($);function g(s){let t=[];if(t.push(`# ${s.name}`),t.push(""),t.push(`> ${s.summary}`),t.push(""),s.routes&&s.routes.length>0){t.push("## Key Routes"),t.push("");for(let e of s.routes){let r=`${s.baseUrl}${e.path}`;t.push(`- [${e.path}](${r}): ${e.description}`)}t.push("")}if(s.externalLinks&&s.externalLinks.length>0){t.push("## Resources"),t.push("");for(let e of s.externalLinks)e.description?t.push(`- [${e.title}](${e.url}): ${e.description}`):t.push(`- [${e.title}](${e.url})`);t.push("")}if(s.sections&&s.sections.length>0)for(let e of s.sections)t.push(`## ${e.heading}`),t.push(""),t.push(e.content),t.push("");return t.join(`
2
2
  `).trim()+`
3
- `}0&&(module.exports={generateLlmsTxt,loadOntoConfig});
3
+ `}0&&(module.exports={generateLlmsTxt});
4
4
  //# sourceMappingURL=config.js.map
package/dist/config.js.map CHANGED
@@ -1 +1 @@
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"]}
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 * 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,IAAA,eAAAC,EAAAH,GAuFO,SAASE,EAAgBE,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","__toCommonJS","config","lines","route","fullUrl","link","section"]}
package/dist/config.mjs CHANGED
@@ -1,4 +1,4 @@
1
- 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(`
1
+ function r(e){let t=[];if(t.push(`# ${e.name}`),t.push(""),t.push(`> ${e.summary}`),t.push(""),e.routes&&e.routes.length>0){t.push("## Key Routes"),t.push("");for(let s of e.routes){let n=`${e.baseUrl}${s.path}`;t.push(`- [${s.path}](${n}): ${s.description}`)}t.push("")}if(e.externalLinks&&e.externalLinks.length>0){t.push("## Resources"),t.push("");for(let s of e.externalLinks)s.description?t.push(`- [${s.title}](${s.url}): ${s.description}`):t.push(`- [${s.title}](${s.url})`);t.push("")}if(e.sections&&e.sections.length>0)for(let s of e.sections)t.push(`## ${s.heading}`),t.push(""),t.push(s.content),t.push("");return t.join(`
2
2
  `).trim()+`
3
- `}export{i as generateLlmsTxt,r as loadOntoConfig};
3
+ `}export{r as generateLlmsTxt};
4
4
  //# sourceMappingURL=config.mjs.map
package/dist/config.mjs.map CHANGED
@@ -1 +1 @@
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"]}
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 * 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":"AAuFO,SAASA,EAAgBC,EAA4B,CAC1D,IAAMC,EAAkB,CAAC,EAWzB,GARAA,EAAM,KAAK,KAAKD,EAAO,IAAI,EAAE,EAC7BC,EAAM,KAAK,EAAE,EAGbA,EAAM,KAAK,KAAKD,EAAO,OAAO,EAAE,EAChCC,EAAM,KAAK,EAAE,EAGTD,EAAO,QAAUA,EAAO,OAAO,OAAS,EAAG,CAC7CC,EAAM,KAAK,eAAe,EAC1BA,EAAM,KAAK,EAAE,EACb,QAAWC,KAASF,EAAO,OAAQ,CACjC,IAAMG,EAAU,GAAGH,EAAO,OAAO,GAAGE,EAAM,IAAI,GAC9CD,EAAM,KAAK,MAAMC,EAAM,IAAI,KAAKC,CAAO,MAAMD,EAAM,WAAW,EAAE,CAClE,CACAD,EAAM,KAAK,EAAE,CACf,CAGA,GAAID,EAAO,eAAiBA,EAAO,cAAc,OAAS,EAAG,CAC3DC,EAAM,KAAK,cAAc,EACzBA,EAAM,KAAK,EAAE,EACb,QAAWG,KAAQJ,EAAO,cACpBI,EAAK,YACPH,EAAM,KAAK,MAAMG,EAAK,KAAK,KAAKA,EAAK,GAAG,MAAMA,EAAK,WAAW,EAAE,EAEhEH,EAAM,KAAK,MAAMG,EAAK,KAAK,KAAKA,EAAK,GAAG,GAAG,EAG/CH,EAAM,KAAK,EAAE,CACf,CAGA,GAAID,EAAO,UAAYA,EAAO,SAAS,OAAS,EAC9C,QAAWK,KAAWL,EAAO,SAC3BC,EAAM,KAAK,MAAMI,EAAQ,OAAO,EAAE,EAClCJ,EAAM,KAAK,EAAE,EACbA,EAAM,KAAKI,EAAQ,OAAO,EAC1BJ,EAAM,KAAK,EAAE,EAIjB,OAAOA,EAAM,KAAK;AAAA,CAAI,EAAE,KAAK,EAAI;AAAA,CACnC","names":["generateLlmsTxt","config","lines","route","fullUrl","link","section"]}
package/dist/index.js CHANGED
@@ -1,11 +1,11 @@
1
- "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:** ${new Date().toISOString()}`,"","---",""].join(`
2
- `)+T;a.length>0&&(r+=`
1
+ "use strict";var P=Object.create;var c=Object.defineProperty;var L=Object.getOwnPropertyDescriptor;var C=Object.getOwnPropertyNames;var H=Object.getPrototypeOf,R=Object.prototype.hasOwnProperty;var v=(t,e)=>{for(var n in e)c(t,n,{get:e[n],enumerable:!0})},d=(t,e,n,o)=>{if(e&&typeof e=="object"||typeof e=="function")for(let i of C(e))!R.call(t,i)&&i!==n&&c(t,i,{get:()=>e[i],enumerable:!(o=L(e,i))||o.enumerable});return t};var f=(t,e,n)=>(n=t!=null?P(H(t)):{},d(e||!t||!t.__esModule?c(n,"default",{value:t,enumerable:!0}):n,t)),I=t=>d(c({},"__esModule",{value:!0}),t);var j={};v(j,{extractContent:()=>x,generateAIOMethodologySchema:()=>l,generateAboutPageSchema:()=>p,generateLlmsTxt:()=>O,generateOrganizationSchema:()=>u,generateSchemaForPageType:()=>w,serializeSchema:()=>z});module.exports=I(j);var y=f(require("cheerio")),S=f(require("turndown")),M=new S.default({headingStyle:"atx",codeBlockStyle:"fenced"});function x(t,e="Generated Output"){let n=t.length,o=y.load(t),i=o("title").text()||o("h1").first().text()||"Untitled Page",g=o('meta[name="description"]').attr("content")||"No description found.",a=[];o('script[type="application/ld+json"]').each((h,$)=>{try{let b=o($).html()||"",A=JSON.parse(b);a.push(A)}catch{}}),o("script, style, noscript, iframe, svg, nav, footer, meta, link, header").remove();let s="";o("main").length>0?s=o("main").html()||"":o("article").length>0?s=o("article").html()||"":s=o("body").html()||"";let k=M.turndown(s),r=[`# ${i}`,`> ${g}`,"",`**Source:** ${e}`,`**Extracted:** ${new Date().toISOString()}`,"","---",""].join(`
2
+ `)+k;a.length>0&&(r+=`
3
3
 
4
4
  ---
5
5
  ## Structured Data (JSON-LD)
6
6
  \`\`\`json
7
7
  `,a.forEach(h=>{r+=JSON.stringify(h,null,2)+`
8
- `}),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(`
8
+ `}),r+="```\n");let m=r.length,T=n>0?(n-m)/n*100:0;return{markdown:r,metadata:{title:i,description:g,jsonLd:a},stats:{originalHtmlSize:n,markdownSize:m,tokenReductionRatio:T}}}function O(t){let e=[];if(e.push(`# ${t.name}`),e.push(""),e.push(`> ${t.summary}`),e.push(""),t.routes&&t.routes.length>0){e.push("## Key Routes"),e.push("");for(let n of t.routes){let o=`${t.baseUrl}${n.path}`;e.push(`- [${n.path}](${o}): ${n.description}`)}e.push("")}if(t.externalLinks&&t.externalLinks.length>0){e.push("## Resources"),e.push("");for(let n of t.externalLinks)n.description?e.push(`- [${n.title}](${n.url}): ${n.description}`):e.push(`- [${n.title}](${n.url})`);e.push("")}if(t.sections&&t.sections.length>0)for(let n of t.sections)e.push(`## ${n.heading}`),e.push(""),e.push(n.content),e.push("");return e.join(`
9
9
  `).trim()+`
10
- `}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});
10
+ `}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 w(t,e,n){switch(t){case"scoring":return l(e,n);case"about":return p(e,n);default:return null}}function z(t){return t?JSON.stringify(t,null,2):null}0&&(module.exports={extractContent,generateAIOMethodologySchema,generateAboutPageSchema,generateLlmsTxt,generateOrganizationSchema,generateSchemaForPageType,serializeSchema});
11
11
  //# sourceMappingURL=index.js.map
package/dist/index.js.map CHANGED
@@ -1 +1 @@
1
- {"version":3,"sources":["../src/index.ts","../src/extractor.ts","../src/config.ts","../src/schemas.ts"],"sourcesContent":["// We cannot use Webpack plugins reliably in Next.js Turbopack due to WorkerError restrictions.\r\n// Users must instead run `npx onto-next` as a postbuild script.\r\nexport { extractContent } from './extractor';\r\nexport { OntoConfig, OntoRoute, 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"]}
1
+ {"version":3,"sources":["../src/index.ts","../src/extractor.ts","../src/config.ts","../src/schemas.ts"],"sourcesContent":["// We cannot use Webpack plugins reliably in Next.js Turbopack due to WorkerError restrictions.\r\n// Users must instead run `npx onto-next` as a postbuild script.\r\nexport { extractContent } from './extractor';\r\nexport { OntoConfig, OntoRoute, 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 * Generate llms.txt content from OntoConfig\r\n * Follows the llms.txt specification:\r\n * - H1 with project name\r\n * - Blockquote with summary\r\n * - Additional markdown sections\r\n */\r\nexport function generateLlmsTxt(config: OntoConfig): string {\r\n const lines: string[] = [];\r\n\r\n // H1: Project name (required)\r\n lines.push(`# ${config.name}`);\r\n lines.push('');\r\n\r\n // Blockquote: Summary (required)\r\n lines.push(`> ${config.summary}`);\r\n lines.push('');\r\n\r\n // Key Routes section (if provided)\r\n if (config.routes && config.routes.length > 0) {\r\n lines.push('## Key Routes');\r\n lines.push('');\r\n for (const route of config.routes) {\r\n const fullUrl = `${config.baseUrl}${route.path}`;\r\n lines.push(`- [${route.path}](${fullUrl}): ${route.description}`);\r\n }\r\n lines.push('');\r\n }\r\n\r\n // External Links section (if provided)\r\n if (config.externalLinks && config.externalLinks.length > 0) {\r\n lines.push('## Resources');\r\n lines.push('');\r\n for (const link of config.externalLinks) {\r\n if (link.description) {\r\n lines.push(`- [${link.title}](${link.url}): ${link.description}`);\r\n } else {\r\n lines.push(`- [${link.title}](${link.url})`);\r\n }\r\n }\r\n lines.push('');\r\n }\r\n\r\n // Custom sections (if provided)\r\n if (config.sections && config.sections.length > 0) {\r\n for (const section of config.sections) {\r\n lines.push(`## ${section.heading}`);\r\n lines.push('');\r\n lines.push(section.content);\r\n lines.push('');\r\n }\r\n }\r\n\r\n return lines.join('\\n').trim() + '\\n';\r\n}\r\n","/**\r\n * JSON-LD Schema generators for automatic structured data injection\r\n * Follows Schema.org standards for AI-friendly metadata\r\n */\r\n\r\nimport { OntoConfig } from './config';\r\n\r\n/**\r\n * Standard AIO (AI Optimization) scoring methodology\r\n * Based on the Onto scoring algorithm:\r\n * - Content Negotiation: 40% (30 points)\r\n * - React Tax / Token Efficiency: 35% (30 points)\r\n * - Structured Data: 25% (25 points)\r\n * - Semantic HTML: Bonus (15 points)\r\n */\r\nexport interface AIOMethodologySchema {\r\n '@context': 'https://schema.org';\r\n '@type': 'HowTo';\r\n name: string;\r\n description: string;\r\n step: Array<{\r\n '@type': 'HowToStep';\r\n name: string;\r\n text: string;\r\n position: number;\r\n }>;\r\n}\r\n\r\n/**\r\n * Generate AIO Scoring Methodology JSON-LD schema\r\n * This explains to AI agents how the scoring system works\r\n */\r\nexport function generateAIOMethodologySchema(\r\n config: OntoConfig,\r\n pageUrl: string\r\n): AIOMethodologySchema {\r\n return {\r\n '@context': 'https://schema.org',\r\n '@type': 'HowTo',\r\n name: 'AIO Score Calculation Methodology',\r\n description: 'AI Optimization (AIO) Score measures how well a website is optimized for AI agents and LLM crawlers. Scored out of 100 points based on four key metrics.',\r\n step: [\r\n {\r\n '@type': 'HowToStep',\r\n name: 'Content Negotiation',\r\n text: 'Check if the site responds to Accept: text/markdown header. Weight: 40%. Penalty: -30 points if missing. This ensures AI bots receive optimized content instead of heavy HTML.',\r\n position: 1\r\n },\r\n {\r\n '@type': 'HowToStep',\r\n name: 'Token Efficiency (React Tax)',\r\n text: 'Measure the ratio of visible text to total HTML size. Weight: 35%. Penalty: -30 points if HTML > 100KB but text < 1KB. Detects JavaScript-heavy sites that are difficult for AI to parse.',\r\n position: 2\r\n },\r\n {\r\n '@type': 'HowToStep',\r\n name: 'Structured Data',\r\n text: 'Verify presence of JSON-LD structured data (Schema.org). Weight: 25%. Penalty: -25 points if missing. Enables AI to confidently extract pricing, products, and entities.',\r\n position: 3\r\n },\r\n {\r\n '@type': 'HowToStep',\r\n name: 'Semantic HTML',\r\n text: 'Check for semantic tags like <main> and <article>. Bonus: +15 points if present. Helps AI agents separate navigation from core content.',\r\n position: 4\r\n }\r\n ]\r\n };\r\n}\r\n\r\n/**\r\n * Organization schema for About pages\r\n */\r\nexport interface OrganizationSchema {\r\n '@context': 'https://schema.org';\r\n '@type': 'Organization';\r\n name: string;\r\n url?: string;\r\n description?: string;\r\n logo?: string;\r\n foundingDate?: string;\r\n}\r\n\r\n/**\r\n * Generate Organization JSON-LD schema for About pages\r\n */\r\nexport function generateOrganizationSchema(\r\n config: OntoConfig,\r\n pageUrl: string\r\n): OrganizationSchema | null {\r\n if (!config.organization) {\r\n return null;\r\n }\r\n\r\n const schema: OrganizationSchema = {\r\n '@context': 'https://schema.org',\r\n '@type': 'Organization',\r\n name: config.organization.name\r\n };\r\n\r\n if (config.organization.url) {\r\n schema.url = config.organization.url;\r\n }\r\n\r\n if (config.organization.description) {\r\n schema.description = config.organization.description;\r\n }\r\n\r\n if (config.organization.logo) {\r\n schema.logo = config.organization.logo;\r\n }\r\n\r\n if (config.organization.foundingDate) {\r\n schema.foundingDate = config.organization.foundingDate;\r\n }\r\n\r\n return schema;\r\n}\r\n\r\n/**\r\n * AboutPage schema combining Organization and WebPage\r\n */\r\nexport interface AboutPageSchema {\r\n '@context': 'https://schema.org';\r\n '@type': 'AboutPage';\r\n name: string;\r\n url: string;\r\n description?: string;\r\n mainEntity?: OrganizationSchema;\r\n}\r\n\r\n/**\r\n * Generate AboutPage JSON-LD schema\r\n */\r\nexport function generateAboutPageSchema(\r\n config: OntoConfig,\r\n pageUrl: string\r\n): AboutPageSchema {\r\n const orgSchema = generateOrganizationSchema(config, pageUrl);\r\n\r\n const schema: AboutPageSchema = {\r\n '@context': 'https://schema.org',\r\n '@type': 'AboutPage',\r\n name: `About ${config.name}`,\r\n url: pageUrl\r\n };\r\n\r\n if (config.summary) {\r\n schema.description = config.summary;\r\n }\r\n\r\n if (orgSchema) {\r\n schema.mainEntity = orgSchema;\r\n }\r\n\r\n return schema;\r\n}\r\n\r\n/**\r\n * Determine which schema to generate based on page type\r\n */\r\nexport function generateSchemaForPageType(\r\n pageType: 'scoring' | 'about' | 'default',\r\n config: OntoConfig,\r\n pageUrl: string\r\n): any | null {\r\n switch (pageType) {\r\n case 'scoring':\r\n return generateAIOMethodologySchema(config, pageUrl);\r\n case 'about':\r\n return generateAboutPageSchema(config, pageUrl);\r\n case 'default':\r\n default:\r\n return null;\r\n }\r\n}\r\n\r\n/**\r\n * Serialize schema to JSON-LD script tag content\r\n */\r\nexport function serializeSchema(schema: any | null): string | null {\r\n if (!schema) {\r\n return null;\r\n }\r\n return JSON.stringify(schema, null, 2);\r\n}\r\n"],"mappings":"0jBAAA,IAAAA,EAAA,GAAAC,EAAAD,EAAA,oBAAAE,EAAA,iCAAAC,EAAA,4BAAAC,EAAA,oBAAAC,EAAA,+BAAAC,EAAA,8BAAAC,EAAA,oBAAAC,IAAA,eAAAC,EAAAT,GCAA,IAAAU,EAAyB,sBACzBC,EAA4B,uBAEtBC,EAAkB,IAAI,EAAAC,QAAgB,CACxC,aAAc,MACd,eAAgB,QACpB,CAAC,EAwBM,SAASC,EAAeC,EAAcC,EAAoB,mBAAsC,CACnG,IAAMC,EAAeF,EAAK,OAEpBG,EAAY,OAAKH,CAAI,EAGrBI,EAAQD,EAAE,OAAO,EAAE,KAAK,GAAKA,EAAE,IAAI,EAAE,MAAM,EAAE,KAAK,GAAK,gBACvDE,EAAcF,EAAE,0BAA0B,EAAE,KAAK,SAAS,GAAK,wBAE/DG,EAAuB,CAAC,EAC9BH,EAAE,oCAAoC,EAAE,KAAK,CAACI,EAAGC,IAAO,CACpD,GAAI,CACA,IAAMC,EAAMN,EAAEK,CAAE,EAAE,KAAK,GAAK,GACtBE,EAAS,KAAK,MAAMD,CAAG,EAC7BH,EAAc,KAAKI,CAAM,CAC7B,MAAQ,CAER,CACJ,CAAC,EAGDP,EAAE,uEAAuE,EAAE,OAAO,EAOlF,IAAIQ,EAAc,GACdR,EAAE,MAAM,EAAE,OAAS,EACnBQ,EAAcR,EAAE,MAAM,EAAE,KAAK,GAAK,GAC3BA,EAAE,SAAS,EAAE,OAAS,EAC7BQ,EAAcR,EAAE,SAAS,EAAE,KAAK,GAAK,GAErCQ,EAAcR,EAAE,MAAM,EAAE,KAAK,GAAK,GAItC,IAAIS,EAAWf,EAAgB,SAASc,CAAW,EAc/CE,EAXgB,CAChB,KAAKT,CAAK,GACV,KAAKC,CAAW,GAChB,GACA,eAAeJ,CAAS,GACxB,kBAAkB,IAAI,KAAK,EAAE,YAAY,CAAC,GAC1C,GACA,MACA,EACJ,EAEgC,KAAK;AAAA,CAAI,EAAIW,EAGzCN,EAAc,OAAS,IACvBO,GAAiB;AAAA;AAAA;AAAA;AAAA;AAAA,EACjBP,EAAc,QAAQQ,GAAK,CACvBD,GAAiB,KAAK,UAAUC,EAAG,KAAM,CAAC,EAAI;AAAA,CAClD,CAAC,EACDD,GAAiB,SAGrB,IAAME,EAAeF,EAAc,OAC7BG,EAAsBd,EAAe,GAAMA,EAAea,GAAgBb,EAAgB,IAAM,EAEtG,MAAO,CACH,SAAUW,EACV,SAAU,CACN,MAAAT,EACA,YAAAC,EACA,OAAQC,CACZ,EACA,MAAO,CACH,iBAAkBJ,EAClB,aAAAa,EACA,oBAAAC,CACJ,CACJ,CACJ,CCtBO,SAASC,EAAgBC,EAA4B,CAC1D,IAAMC,EAAkB,CAAC,EAWzB,GARAA,EAAM,KAAK,KAAKD,EAAO,IAAI,EAAE,EAC7BC,EAAM,KAAK,EAAE,EAGbA,EAAM,KAAK,KAAKD,EAAO,OAAO,EAAE,EAChCC,EAAM,KAAK,EAAE,EAGTD,EAAO,QAAUA,EAAO,OAAO,OAAS,EAAG,CAC7CC,EAAM,KAAK,eAAe,EAC1BA,EAAM,KAAK,EAAE,EACb,QAAWC,KAASF,EAAO,OAAQ,CACjC,IAAMG,EAAU,GAAGH,EAAO,OAAO,GAAGE,EAAM,IAAI,GAC9CD,EAAM,KAAK,MAAMC,EAAM,IAAI,KAAKC,CAAO,MAAMD,EAAM,WAAW,EAAE,CAClE,CACAD,EAAM,KAAK,EAAE,CACf,CAGA,GAAID,EAAO,eAAiBA,EAAO,cAAc,OAAS,EAAG,CAC3DC,EAAM,KAAK,cAAc,EACzBA,EAAM,KAAK,EAAE,EACb,QAAWG,KAAQJ,EAAO,cACpBI,EAAK,YACPH,EAAM,KAAK,MAAMG,EAAK,KAAK,KAAKA,EAAK,GAAG,MAAMA,EAAK,WAAW,EAAE,EAEhEH,EAAM,KAAK,MAAMG,EAAK,KAAK,KAAKA,EAAK,GAAG,GAAG,EAG/CH,EAAM,KAAK,EAAE,CACf,CAGA,GAAID,EAAO,UAAYA,EAAO,SAAS,OAAS,EAC9C,QAAWK,KAAWL,EAAO,SAC3BC,EAAM,KAAK,MAAMI,EAAQ,OAAO,EAAE,EAClCJ,EAAM,KAAK,EAAE,EACbA,EAAM,KAAKI,EAAQ,OAAO,EAC1BJ,EAAM,KAAK,EAAE,EAIjB,OAAOA,EAAM,KAAK;AAAA,CAAI,EAAE,KAAK,EAAI;AAAA,CACnC,CCtGO,SAASK,EACdC,EACAC,EACsB,CACtB,MAAO,CACL,WAAY,qBACZ,QAAS,QACT,KAAM,oCACN,YAAa,2JACb,KAAM,CACJ,CACE,QAAS,YACT,KAAM,sBACN,KAAM,iLACN,SAAU,CACZ,EACA,CACE,QAAS,YACT,KAAM,+BACN,KAAM,4LACN,SAAU,CACZ,EACA,CACE,QAAS,YACT,KAAM,kBACN,KAAM,2KACN,SAAU,CACZ,EACA,CACE,QAAS,YACT,KAAM,gBACN,KAAM,0IACN,SAAU,CACZ,CACF,CACF,CACF,CAkBO,SAASC,EACdF,EACAC,EAC2B,CAC3B,GAAI,CAACD,EAAO,aACV,OAAO,KAGT,IAAMG,EAA6B,CACjC,WAAY,qBACZ,QAAS,eACT,KAAMH,EAAO,aAAa,IAC5B,EAEA,OAAIA,EAAO,aAAa,MACtBG,EAAO,IAAMH,EAAO,aAAa,KAG/BA,EAAO,aAAa,cACtBG,EAAO,YAAcH,EAAO,aAAa,aAGvCA,EAAO,aAAa,OACtBG,EAAO,KAAOH,EAAO,aAAa,MAGhCA,EAAO,aAAa,eACtBG,EAAO,aAAeH,EAAO,aAAa,cAGrCG,CACT,CAiBO,SAASC,EACdJ,EACAC,EACiB,CACjB,IAAMI,EAAYH,EAA2BF,EAAQC,CAAO,EAEtDE,EAA0B,CAC9B,WAAY,qBACZ,QAAS,YACT,KAAM,SAASH,EAAO,IAAI,GAC1B,IAAKC,CACP,EAEA,OAAID,EAAO,UACTG,EAAO,YAAcH,EAAO,SAG1BK,IACFF,EAAO,WAAaE,GAGfF,CACT,CAKO,SAASG,EACdC,EACAP,EACAC,EACY,CACZ,OAAQM,EAAU,CAChB,IAAK,UACH,OAAOR,EAA6BC,EAAQC,CAAO,EACrD,IAAK,QACH,OAAOG,EAAwBJ,EAAQC,CAAO,EAEhD,QACE,OAAO,IACX,CACF,CAKO,SAASO,EAAgBL,EAAmC,CACjE,OAAKA,EAGE,KAAK,UAAUA,EAAQ,KAAM,CAAC,EAF5B,IAGX","names":["index_exports","__export","extractContent","generateAIOMethodologySchema","generateAboutPageSchema","generateLlmsTxt","generateOrganizationSchema","generateSchemaForPageType","serializeSchema","__toCommonJS","cheerio","import_turndown","turndownService","TurndownService","extractContent","html","sourceUrl","originalSize","$","title","description","jsonLdScripts","_","el","raw","parsed","contentHtml","markdown","finalMarkdown","j","markdownSize","tokenReductionRatio","generateLlmsTxt","config","lines","route","fullUrl","link","section","generateAIOMethodologySchema","config","pageUrl","generateOrganizationSchema","schema","generateAboutPageSchema","orgSchema","generateSchemaForPageType","pageType","serializeSchema"]}
package/dist/index.mjs CHANGED
@@ -5,7 +5,7 @@ import*as p from"cheerio";import O from"turndown";var w=new O({headingStyle:"atx
5
5
  ## Structured Data (JSON-LD)
6
6
  \`\`\`json
7
7
  `,r.forEach(u=>{i+=JSON.stringify(u,null,2)+`
8
- `}),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(`
8
+ `}),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}}}function k(t){let n=[];if(n.push(`# ${t.name}`),n.push(""),n.push(`> ${t.summary}`),n.push(""),t.routes&&t.routes.length>0){n.push("## Key Routes"),n.push("");for(let e of t.routes){let o=`${t.baseUrl}${e.path}`;n.push(`- [${e.path}](${o}): ${e.description}`)}n.push("")}if(t.externalLinks&&t.externalLinks.length>0){n.push("## Resources"),n.push("");for(let e of t.externalLinks)e.description?n.push(`- [${e.title}](${e.url}): ${e.description}`):n.push(`- [${e.title}](${e.url})`);n.push("")}if(t.sections&&t.sections.length>0)for(let e of t.sections)n.push(`## ${e.heading}`),n.push(""),n.push(e.content),n.push("");return n.join(`
9
9
  `).trim()+`
10
- `}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};
10
+ `}function g(t,n){return{"@context":"https://schema.org","@type":"HowTo",name:"AIO Score Calculation Methodology",description:"AI Optimization (AIO) Score measures how well a website is optimized for AI agents and LLM crawlers. Scored out of 100 points based on four key metrics.",step:[{"@type":"HowToStep",name:"Content Negotiation",text:"Check if the site responds to Accept: text/markdown header. Weight: 40%. Penalty: -30 points if missing. This ensures AI bots receive optimized content instead of heavy HTML.",position:1},{"@type":"HowToStep",name:"Token Efficiency (React Tax)",text:"Measure the ratio of visible text to total HTML size. Weight: 35%. Penalty: -30 points if HTML > 100KB but text < 1KB. Detects JavaScript-heavy sites that are difficult for AI to parse.",position:2},{"@type":"HowToStep",name:"Structured Data",text:"Verify presence of JSON-LD structured data (Schema.org). Weight: 25%. Penalty: -25 points if missing. Enables AI to confidently extract pricing, products, and entities.",position:3},{"@type":"HowToStep",name:"Semantic HTML",text:"Check for semantic tags like <main> and <article>. Bonus: +15 points if present. Helps AI agents separate navigation from core content.",position:4}]}}function m(t,n){if(!t.organization)return null;let e={"@context":"https://schema.org","@type":"Organization",name:t.organization.name};return t.organization.url&&(e.url=t.organization.url),t.organization.description&&(e.description=t.organization.description),t.organization.logo&&(e.logo=t.organization.logo),t.organization.foundingDate&&(e.foundingDate=t.organization.foundingDate),e}function h(t,n){let e=m(t,n),o={"@context":"https://schema.org","@type":"AboutPage",name:`About ${t.name}`,url:n};return t.summary&&(o.description=t.summary),e&&(o.mainEntity=e),o}function T(t,n,e){switch(t){case"scoring":return g(n,e);case"about":return h(n,e);default:return null}}function $(t){return t?JSON.stringify(t,null,2):null}export{z as extractContent,g as generateAIOMethodologySchema,h as generateAboutPageSchema,k as generateLlmsTxt,m as generateOrganizationSchema,T as generateSchemaForPageType,$ as serializeSchema};
11
11
  //# sourceMappingURL=index.mjs.map