npm - @ontosdk/next - Versions diffs - 1.3.1 → 1.3.2 - Mend

@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

package/dist/index.mjs.map CHANGED Viewed

	@@ -1 +1 @@
1	- {"version":3,"sources":["../src/extractor.ts","../src/config.ts","../src/schemas.ts"],"sourcesContent":["import * as cheerio from 'cheerio';\r\nimport TurndownService from 'turndown';\r\n\r\nconst turndownService = new TurndownService({\r\n headingStyle: 'atx',\r\n codeBlockStyle: 'fenced',\r\n});\r\n\r\n// Configure turndown to keep some layout or handle semantic tags differently if needed\r\n\r\nexport interface ExtractionResult {\r\n markdown: string;\r\n metadata: {\r\n title: string;\r\n description: string;\r\n jsonLd: any[];\r\n };\r\n stats: {\r\n originalHtmlSize: number;\r\n markdownSize: number;\r\n tokenReductionRatio: number;\r\n };\r\n}\r\n\r\n/*\r\n Extracts pure semantic markdown and metadata from rendered Next.js HTML strings.\r\n * @param html The raw HTML string.\r\n * @param sourceUrl (Optional) the URL this was generated from, to attach as metadata.\r\n * @returns {ExtractionResult} The extracted payload.\r\n /\r\nexport function extractContent(html: string, sourceUrl: string = 'Generated Output'): ExtractionResult {\r\n const originalSize = html.length;\r\n\r\n const $ = cheerio.load(html);\r\n\r\n // 1. Extract Metadata BEFORE removing structure\r\n const title = $('title').text() \|\| $('h1').first().text() \|\| 'Untitled Page';\r\n const description = $('meta[name=\"description\"]').attr('content') \|\| 'No description found.';\r\n\r\n const jsonLdScripts: any[] = [];\r\n $('script[type=\"application/ld+json\"]').each((_, el) => {\r\n try {\r\n const raw = $(el).html() \|\| '';\r\n const parsed = JSON.parse(raw);\r\n jsonLdScripts.push(parsed);\r\n } catch {\r\n // ignore bad json\r\n }\r\n });\r\n\r\n // 2. Strip noise (React boilerplate, styles, unnecessary tags)\r\n $('script, style, noscript, iframe, svg, nav, footer, meta, link, header').remove();\r\n\r\n // Optionally remove typical Next.js hidden wrappers if they don't contain real content.\r\n // Next.js uses <div id=\"__next\"> but we mostly just want semantic content.\r\n\r\n // 3. Find the entry point for content\r\n // Prefer <main> or <article> over <body>\r\n let contentHtml = '';\r\n if ($('main').length > 0) {\r\n contentHtml = $('main').html() \|\| '';\r\n } else if ($('article').length > 0) {\r\n contentHtml = $('article').html() \|\| '';\r\n } else {\r\n contentHtml = $('body').html() \|\| '';\r\n }\r\n\r\n // 4. Convert to Markdown\r\n let markdown = turndownService.turndown(contentHtml);\r\n\r\n // 5. Optionally inject Metadata header\r\n const headerLines = [\r\n `# ${title}`,\r\n `> ${description}`,\r\n ``,\r\n `Source:* ${sourceUrl}`,\r\n `Extracted: ${new Date().toISOString()}`,\r\n ``,\r\n `---`,\r\n ``\r\n ];\r\n\r\n let finalMarkdown = headerLines.join('\\n') + markdown;\r\n\r\n // Add JSON-LD section if exists\r\n if (jsonLdScripts.length > 0) {\r\n finalMarkdown += '\\n\\n---\\n## Structured Data (JSON-LD)\\n```json\\n';\r\n jsonLdScripts.forEach(j => {\r\n finalMarkdown += JSON.stringify(j, null, 2) + '\\n';\r\n });\r\n finalMarkdown += '```\\n';\r\n }\r\n\r\n const markdownSize = finalMarkdown.length;\r\n const tokenReductionRatio = originalSize > 0 ? ((originalSize - markdownSize) / originalSize) * 100 : 0;\r\n\r\n return {\r\n markdown: finalMarkdown,\r\n metadata: {\r\n title,\r\n description,\r\n jsonLd: jsonLdScripts\r\n },\r\n stats: {\r\n originalHtmlSize: originalSize,\r\n markdownSize,\r\n tokenReductionRatio\r\n }\r\n };\r\n}\r\n\r\nexport async function generateStaticPayloads(nextAppDirDir: string, ontoPublicDir: string) {\r\n const fs = await import('fs');\r\n const path = await import('path');\r\n const { glob } = await import('glob');\r\n\r\n if (!fs.existsSync(nextAppDirDir)) {\r\n return;\r\n }\r\n\r\n const files = await glob('*/.html', { cwd: nextAppDirDir });\r\n if (files.length === 0) return;\r\n\r\n if (!fs.existsSync(ontoPublicDir)) {\r\n fs.mkdirSync(ontoPublicDir, { recursive: true });\r\n }\r\n\r\n let totalFilesProcessed = 0;\r\n\r\n for (const file of files) {\r\n const inputPath = path.join(nextAppDirDir, file);\r\n const outputPathRelative = file.replace(/\\.html$/, '.md');\r\n const outputPath = path.join(ontoPublicDir, outputPathRelative);\r\n\r\n try {\r\n const htmlContent = fs.readFileSync(inputPath, 'utf8');\r\n\r\n let routeName = file.replace(/\\.html$/, '');\r\n if (routeName === 'index') routeName = '/';\r\n else routeName = `/${routeName}`;\r\n\r\n const result = extractContent(htmlContent, routeName);\r\n\r\n const outputDir = path.dirname(outputPath);\r\n if (!fs.existsSync(outputDir)) {\r\n fs.mkdirSync(outputDir, { recursive: true });\r\n }\r\n\r\n fs.writeFileSync(outputPath, result.markdown, 'utf8');\r\n totalFilesProcessed++;\r\n } catch (e: any) {\r\n console.error(`[Onto] Failed to process ${file}: ${e.message}`);\r\n }\r\n }\r\n console.log(`[Onto] Successfully generated ${totalFilesProcessed} semantic markdown endpoints.`);\r\n}\r\n","/*\r\n Configuration schema for onto.config.ts\r\n * Used to dynamically generate llms.txt and other AI discovery files\r\n /\r\n\r\nexport type PageType = 'scoring' \| 'about' \| 'default';\r\n\r\nexport interface OntoRoute {\r\n /\r\n The URL path (e.g., '/docs', '/api/reference')\r\n /\r\n path: string;\r\n /\r\n Description of what this route contains\r\n /\r\n description: string;\r\n /\r\n Optional: Page type for automatic JSON-LD schema injection\r\n * - 'scoring': Injects Methodology schema with AIO scoring weights (40/35/25)\r\n * - 'about': Injects Organization/AboutPage schema\r\n * - 'default': No automatic schema injection\r\n /\r\n pageType?: PageType;\r\n}\r\n\r\nexport interface OntoConfig {\r\n /\r\n The name of your project or site (required)\r\n * Used as the H1 heading in llms.txt\r\n /\r\n name: string;\r\n\r\n /\r\n A short summary of your project (required)\r\n * Displayed as a blockquote in llms.txt\r\n * Should contain key information necessary for understanding the rest of the file\r\n /\r\n summary: string;\r\n\r\n /\r\n The base URL of your site (e.g., 'https://example.com')\r\n /\r\n baseUrl: string;\r\n\r\n /\r\n Optional: Additional sections to include in llms.txt\r\n * Each section can contain any markdown content\r\n /\r\n sections?: {\r\n heading: string;\r\n content: string;\r\n }[];\r\n\r\n /\r\n Key routes that AI agents should know about\r\n * These will be formatted as a markdown list in llms.txt\r\n /\r\n routes?: OntoRoute[];\r\n\r\n /\r\n Optional: Links to external resources (documentation, API references, etc.)\r\n /\r\n externalLinks?: {\r\n title: string;\r\n url: string;\r\n description?: string;\r\n }[];\r\n\r\n /\r\n Optional: Organization information for JSON-LD schemas\r\n /\r\n organization?: {\r\n name: string;\r\n description?: string;\r\n url?: string;\r\n logo?: string;\r\n foundingDate?: string;\r\n };\r\n}\r\n\r\n/\r\n Load the onto.config.ts file from the user's project\r\n * This is used by the middleware to dynamically generate llms.txt\r\n /\r\nexport async function loadOntoConfig(): Promise<OntoConfig \| null> {\r\n try {\r\n // Try to dynamically import the config file from the user's project root\r\n // This runs in the middleware context, so we look in the project root\r\n const config = await import(process.cwd() + '/onto.config');\r\n return config.default \|\| config;\r\n } catch (error) {\r\n // Config file doesn't exist or failed to load\r\n return null;\r\n }\r\n}\r\n\r\n/\r\n Generate llms.txt content from OntoConfig\r\n * Follows the llms.txt specification:\r\n * - H1 with project name\r\n * - Blockquote with summary\r\n * - Additional markdown sections\r\n /\r\nexport function generateLlmsTxt(config: OntoConfig): string {\r\n const lines: string[] = [];\r\n\r\n // H1: Project name (required)\r\n lines.push(`# ${config.name}`);\r\n lines.push('');\r\n\r\n // Blockquote: Summary (required)\r\n lines.push(`> ${config.summary}`);\r\n lines.push('');\r\n\r\n // Key Routes section (if provided)\r\n if (config.routes && config.routes.length > 0) {\r\n lines.push('## Key Routes');\r\n lines.push('');\r\n for (const route of config.routes) {\r\n const fullUrl = `${config.baseUrl}${route.path}`;\r\n lines.push(`- [${route.path}](${fullUrl}): ${route.description}`);\r\n }\r\n lines.push('');\r\n }\r\n\r\n // External Links section (if provided)\r\n if (config.externalLinks && config.externalLinks.length > 0) {\r\n lines.push('## Resources');\r\n lines.push('');\r\n for (const link of config.externalLinks) {\r\n if (link.description) {\r\n lines.push(`- [${link.title}](${link.url}): ${link.description}`);\r\n } else {\r\n lines.push(`- [${link.title}](${link.url})`);\r\n }\r\n }\r\n lines.push('');\r\n }\r\n\r\n // Custom sections (if provided)\r\n if (config.sections && config.sections.length > 0) {\r\n for (const section of config.sections) {\r\n lines.push(`## ${section.heading}`);\r\n lines.push('');\r\n lines.push(section.content);\r\n lines.push('');\r\n }\r\n }\r\n\r\n return lines.join('\\n').trim() + '\\n';\r\n}\r\n","/\r\n JSON-LD Schema generators for automatic structured data injection\r\n * Follows Schema.org standards for AI-friendly metadata\r\n /\r\n\r\nimport { OntoConfig } from './config';\r\n\r\n/\r\n Standard AIO (AI Optimization) scoring methodology\r\n * Based on the Onto scoring algorithm:\r\n * - Content Negotiation: 40% (30 points)\r\n * - React Tax / Token Efficiency: 35% (30 points)\r\n * - Structured Data: 25% (25 points)\r\n * - Semantic HTML: Bonus (15 points)\r\n /\r\nexport interface AIOMethodologySchema {\r\n '@context': 'https://schema.org';\r\n '@type': 'HowTo';\r\n name: string;\r\n description: string;\r\n step: Array<{\r\n '@type': 'HowToStep';\r\n name: string;\r\n text: string;\r\n position: number;\r\n }>;\r\n}\r\n\r\n/\r\n Generate AIO Scoring Methodology JSON-LD schema\r\n * This explains to AI agents how the scoring system works\r\n /\r\nexport function generateAIOMethodologySchema(\r\n config: OntoConfig,\r\n pageUrl: string\r\n): AIOMethodologySchema {\r\n return {\r\n '@context': 'https://schema.org',\r\n '@type': 'HowTo',\r\n name: 'AIO Score Calculation Methodology',\r\n description: 'AI Optimization (AIO) Score measures how well a website is optimized for AI agents and LLM crawlers. Scored out of 100 points based on four key metrics.',\r\n step: [\r\n {\r\n '@type': 'HowToStep',\r\n name: 'Content Negotiation',\r\n text: 'Check if the site responds to Accept: text/markdown header. Weight: 40%. Penalty: -30 points if missing. This ensures AI bots receive optimized content instead of heavy HTML.',\r\n position: 1\r\n },\r\n {\r\n '@type': 'HowToStep',\r\n name: 'Token Efficiency (React Tax)',\r\n text: 'Measure the ratio of visible text to total HTML size. Weight: 35%. Penalty: -30 points if HTML > 100KB but text < 1KB. Detects JavaScript-heavy sites that are difficult for AI to parse.',\r\n position: 2\r\n },\r\n {\r\n '@type': 'HowToStep',\r\n name: 'Structured Data',\r\n text: 'Verify presence of JSON-LD structured data (Schema.org). Weight: 25%. Penalty: -25 points if missing. Enables AI to confidently extract pricing, products, and entities.',\r\n position: 3\r\n },\r\n {\r\n '@type': 'HowToStep',\r\n name: 'Semantic HTML',\r\n text: 'Check for semantic tags like <main> and <article>. Bonus: +15 points if present. Helps AI agents separate navigation from core content.',\r\n position: 4\r\n }\r\n ]\r\n };\r\n}\r\n\r\n/\r\n Organization schema for About pages\r\n /\r\nexport interface OrganizationSchema {\r\n '@context': 'https://schema.org';\r\n '@type': 'Organization';\r\n name: string;\r\n url?: string;\r\n description?: string;\r\n logo?: string;\r\n foundingDate?: string;\r\n}\r\n\r\n/\r\n Generate Organization JSON-LD schema for About pages\r\n /\r\nexport function generateOrganizationSchema(\r\n config: OntoConfig,\r\n pageUrl: string\r\n): OrganizationSchema \| null {\r\n if (!config.organization) {\r\n return null;\r\n }\r\n\r\n const schema: OrganizationSchema = {\r\n '@context': 'https://schema.org',\r\n '@type': 'Organization',\r\n name: config.organization.name\r\n };\r\n\r\n if (config.organization.url) {\r\n schema.url = config.organization.url;\r\n }\r\n\r\n if (config.organization.description) {\r\n schema.description = config.organization.description;\r\n }\r\n\r\n if (config.organization.logo) {\r\n schema.logo = config.organization.logo;\r\n }\r\n\r\n if (config.organization.foundingDate) {\r\n schema.foundingDate = config.organization.foundingDate;\r\n }\r\n\r\n return schema;\r\n}\r\n\r\n/\r\n AboutPage schema combining Organization and WebPage\r\n /\r\nexport interface AboutPageSchema {\r\n '@context': 'https://schema.org';\r\n '@type': 'AboutPage';\r\n name: string;\r\n url: string;\r\n description?: string;\r\n mainEntity?: OrganizationSchema;\r\n}\r\n\r\n/\r\n Generate AboutPage JSON-LD schema\r\n /\r\nexport function generateAboutPageSchema(\r\n config: OntoConfig,\r\n pageUrl: string\r\n): AboutPageSchema {\r\n const orgSchema = generateOrganizationSchema(config, pageUrl);\r\n\r\n const schema: AboutPageSchema = {\r\n '@context': 'https://schema.org',\r\n '@type': 'AboutPage',\r\n name: `About ${config.name}`,\r\n url: pageUrl\r\n };\r\n\r\n if (config.summary) {\r\n schema.description = config.summary;\r\n }\r\n\r\n if (orgSchema) {\r\n schema.mainEntity = orgSchema;\r\n }\r\n\r\n return schema;\r\n}\r\n\r\n/\r\n Determine which schema to generate based on page type\r\n /\r\nexport function generateSchemaForPageType(\r\n pageType: 'scoring' \| 'about' \| 'default',\r\n config: OntoConfig,\r\n pageUrl: string\r\n): any \| null {\r\n switch (pageType) {\r\n case 'scoring':\r\n return generateAIOMethodologySchema(config, pageUrl);\r\n case 'about':\r\n return generateAboutPageSchema(config, pageUrl);\r\n case 'default':\r\n default:\r\n return null;\r\n }\r\n}\r\n\r\n/\r\n Serialize schema to JSON-LD script tag content\r\n */\r\nexport function serializeSchema(schema: any \| null): string \| null {\r\n if (!schema) {\r\n return null;\r\n }\r\n return JSON.stringify(schema, null, 2);\r\n}\r\n"],"mappings":"AAAA,UAAYA,MAAa,UACzB,OAAOC,MAAqB,WAE5B,IAAMC,EAAkB,IAAID,EAAgB,CACxC,aAAc,MACd,eAAgB,QACpB,CAAC,EAwBM,SAASE,EAAeC,EAAcC,EAAoB,mBAAsC,CACnG,IAAMC,EAAeF,EAAK,OAEpBG,EAAY,OAAKH,CAAI,EAGrBI,EAAQD,EAAE,OAAO,EAAE,KAAK,GAAKA,EAAE,IAAI,EAAE,MAAM,EAAE,KAAK,GAAK,gBACvDE,EAAcF,EAAE,0BAA0B,EAAE,KAAK,SAAS,GAAK,wBAE/DG,EAAuB,CAAC,EAC9BH,EAAE,oCAAoC,EAAE,KAAK,CAACI,EAAGC,IAAO,CACpD,GAAI,CACA,IAAMC,EAAMN,EAAEK,CAAE,EAAE,KAAK,GAAK,GACtBE,EAAS,KAAK,MAAMD,CAAG,EAC7BH,EAAc,KAAKI,CAAM,CAC7B,MAAQ,CAER,CACJ,CAAC,EAGDP,EAAE,uEAAuE,EAAE,OAAO,EAOlF,IAAIQ,EAAc,GACdR,EAAE,MAAM,EAAE,OAAS,EACnBQ,EAAcR,EAAE,MAAM,EAAE,KAAK,GAAK,GAC3BA,EAAE,SAAS,EAAE,OAAS,EAC7BQ,EAAcR,EAAE,SAAS,EAAE,KAAK,GAAK,GAErCQ,EAAcR,EAAE,MAAM,EAAE,KAAK,GAAK,GAItC,IAAIS,EAAWd,EAAgB,SAASa,CAAW,EAc/CE,EAXgB,CAChB,KAAKT,CAAK,GACV,KAAKC,CAAW,GAChB,GACA,eAAeJ,CAAS,GACxB,kBAAkB,IAAI,KAAK,EAAE,YAAY,CAAC,GAC1C,GACA,MACA,EACJ,EAEgC,KAAK;AAAA,CAAI,EAAIW,EAGzCN,EAAc,OAAS,IACvBO,GAAiB;AAAA;AAAA;AAAA;AAAA;AAAA,EACjBP,EAAc,QAAQQ,GAAK,CACvBD,GAAiB,KAAK,UAAUC,EAAG,KAAM,CAAC,EAAI;AAAA,CAClD,CAAC,EACDD,GAAiB,SAGrB,IAAME,EAAeF,EAAc,OAC7BG,EAAsBd,EAAe,GAAMA,EAAea,GAAgBb,EAAgB,IAAM,EAEtG,MAAO,CACH,SAAUW,EACV,SAAU,CACN,MAAAT,EACA,YAAAC,EACA,OAAQC,CACZ,EACA,MAAO,CACH,iBAAkBJ,EAClB,aAAAa,EACA,oBAAAC,CACJ,CACJ,CACJ,CCzBA,eAAsBC,GAA6C,CACjE,GAAI,CAGF,IAAMC,EAAS,MAAM,OAAO,QAAQ,IAAI,EAAI,gBAC5C,OAAOA,EAAO,SAAWA,CAC3B,MAAgB,CAEd,OAAO,IACT,CACF,CASO,SAASC,EAAgBD,EAA4B,CAC1D,IAAME,EAAkB,CAAC,EAWzB,GARAA,EAAM,KAAK,KAAKF,EAAO,IAAI,EAAE,EAC7BE,EAAM,KAAK,EAAE,EAGbA,EAAM,KAAK,KAAKF,EAAO,OAAO,EAAE,EAChCE,EAAM,KAAK,EAAE,EAGTF,EAAO,QAAUA,EAAO,OAAO,OAAS,EAAG,CAC7CE,EAAM,KAAK,eAAe,EAC1BA,EAAM,KAAK,EAAE,EACb,QAAWC,KAASH,EAAO,OAAQ,CACjC,IAAMI,EAAU,GAAGJ,EAAO,OAAO,GAAGG,EAAM,IAAI,GAC9CD,EAAM,KAAK,MAAMC,EAAM,IAAI,KAAKC,CAAO,MAAMD,EAAM,WAAW,EAAE,CAClE,CACAD,EAAM,KAAK,EAAE,CACf,CAGA,GAAIF,EAAO,eAAiBA,EAAO,cAAc,OAAS,EAAG,CAC3DE,EAAM,KAAK,cAAc,EACzBA,EAAM,KAAK,EAAE,EACb,QAAWG,KAAQL,EAAO,cACpBK,EAAK,YACPH,EAAM,KAAK,MAAMG,EAAK,KAAK,KAAKA,EAAK,GAAG,MAAMA,EAAK,WAAW,EAAE,EAEhEH,EAAM,KAAK,MAAMG,EAAK,KAAK,KAAKA,EAAK,GAAG,GAAG,EAG/CH,EAAM,KAAK,EAAE,CACf,CAGA,GAAIF,EAAO,UAAYA,EAAO,SAAS,OAAS,EAC9C,QAAWM,KAAWN,EAAO,SAC3BE,EAAM,KAAK,MAAMI,EAAQ,OAAO,EAAE,EAClCJ,EAAM,KAAK,EAAE,EACbA,EAAM,KAAKI,EAAQ,OAAO,EAC1BJ,EAAM,KAAK,EAAE,EAIjB,OAAOA,EAAM,KAAK;AAAA,CAAI,EAAE,KAAK,EAAI;AAAA,CACnC,CCtHO,SAASK,EACdC,EACAC,EACsB,CACtB,MAAO,CACL,WAAY,qBACZ,QAAS,QACT,KAAM,oCACN,YAAa,2JACb,KAAM,CACJ,CACE,QAAS,YACT,KAAM,sBACN,KAAM,iLACN,SAAU,CACZ,EACA,CACE,QAAS,YACT,KAAM,+BACN,KAAM,4LACN,SAAU,CACZ,EACA,CACE,QAAS,YACT,KAAM,kBACN,KAAM,2KACN,SAAU,CACZ,EACA,CACE,QAAS,YACT,KAAM,gBACN,KAAM,0IACN,SAAU,CACZ,CACF,CACF,CACF,CAkBO,SAASC,EACdF,EACAC,EAC2B,CAC3B,GAAI,CAACD,EAAO,aACV,OAAO,KAGT,IAAMG,EAA6B,CACjC,WAAY,qBACZ,QAAS,eACT,KAAMH,EAAO,aAAa,IAC5B,EAEA,OAAIA,EAAO,aAAa,MACtBG,EAAO,IAAMH,EAAO,aAAa,KAG/BA,EAAO,aAAa,cACtBG,EAAO,YAAcH,EAAO,aAAa,aAGvCA,EAAO,aAAa,OACtBG,EAAO,KAAOH,EAAO,aAAa,MAGhCA,EAAO,aAAa,eACtBG,EAAO,aAAeH,EAAO,aAAa,cAGrCG,CACT,CAiBO,SAASC,EACdJ,EACAC,EACiB,CACjB,IAAMI,EAAYH,EAA2BF,EAAQC,CAAO,EAEtDE,EAA0B,CAC9B,WAAY,qBACZ,QAAS,YACT,KAAM,SAASH,EAAO,IAAI,GAC1B,IAAKC,CACP,EAEA,OAAID,EAAO,UACTG,EAAO,YAAcH,EAAO,SAG1BK,IACFF,EAAO,WAAaE,GAGfF,CACT,CAKO,SAASG,EACdC,EACAP,EACAC,EACY,CACZ,OAAQM,EAAU,CAChB,IAAK,UACH,OAAOR,EAA6BC,EAAQC,CAAO,EACrD,IAAK,QACH,OAAOG,EAAwBJ,EAAQC,CAAO,EAEhD,QACE,OAAO,IACX,CACF,CAKO,SAASO,EAAgBL,EAAmC,CACjE,OAAKA,EAGE,KAAK,UAAUA,EAAQ,KAAM,CAAC,EAF5B,IAGX","names":["cheerio","TurndownService","turndownService","extractContent","html","sourceUrl","originalSize","$","title","description","jsonLdScripts","_","el","raw","parsed","contentHtml","markdown","finalMarkdown","j","markdownSize","tokenReductionRatio","loadOntoConfig","config","generateLlmsTxt","lines","route","fullUrl","link","section","generateAIOMethodologySchema","config","pageUrl","generateOrganizationSchema","schema","generateAboutPageSchema","orgSchema","generateSchemaForPageType","pageType","serializeSchema"]}
1	+ {"version":3,"sources":["../src/extractor.ts","../src/config.ts","../src/schemas.ts"],"sourcesContent":["import * as cheerio from 'cheerio';\r\nimport TurndownService from 'turndown';\r\n\r\nconst turndownService = new TurndownService({\r\n headingStyle: 'atx',\r\n codeBlockStyle: 'fenced',\r\n});\r\n\r\n// Configure turndown to keep some layout or handle semantic tags differently if needed\r\n\r\nexport interface ExtractionResult {\r\n markdown: string;\r\n metadata: {\r\n title: string;\r\n description: string;\r\n jsonLd: any[];\r\n };\r\n stats: {\r\n originalHtmlSize: number;\r\n markdownSize: number;\r\n tokenReductionRatio: number;\r\n };\r\n}\r\n\r\n/*\r\n Extracts pure semantic markdown and metadata from rendered Next.js HTML strings.\r\n * @param html The raw HTML string.\r\n * @param sourceUrl (Optional) the URL this was generated from, to attach as metadata.\r\n * @returns {ExtractionResult} The extracted payload.\r\n /\r\nexport function extractContent(html: string, sourceUrl: string = 'Generated Output'): ExtractionResult {\r\n const originalSize = html.length;\r\n\r\n const $ = cheerio.load(html);\r\n\r\n // 1. Extract Metadata BEFORE removing structure\r\n const title = $('title').text() \|\| $('h1').first().text() \|\| 'Untitled Page';\r\n const description = $('meta[name=\"description\"]').attr('content') \|\| 'No description found.';\r\n\r\n const jsonLdScripts: any[] = [];\r\n $('script[type=\"application/ld+json\"]').each((_, el) => {\r\n try {\r\n const raw = $(el).html() \|\| '';\r\n const parsed = JSON.parse(raw);\r\n jsonLdScripts.push(parsed);\r\n } catch {\r\n // ignore bad json\r\n }\r\n });\r\n\r\n // 2. Strip noise (React boilerplate, styles, unnecessary tags)\r\n $('script, style, noscript, iframe, svg, nav, footer, meta, link, header').remove();\r\n\r\n // Optionally remove typical Next.js hidden wrappers if they don't contain real content.\r\n // Next.js uses <div id=\"__next\"> but we mostly just want semantic content.\r\n\r\n // 3. Find the entry point for content\r\n // Prefer <main> or <article> over <body>\r\n let contentHtml = '';\r\n if ($('main').length > 0) {\r\n contentHtml = $('main').html() \|\| '';\r\n } else if ($('article').length > 0) {\r\n contentHtml = $('article').html() \|\| '';\r\n } else {\r\n contentHtml = $('body').html() \|\| '';\r\n }\r\n\r\n // 4. Convert to Markdown\r\n let markdown = turndownService.turndown(contentHtml);\r\n\r\n // 5. Optionally inject Metadata header\r\n const headerLines = [\r\n `# ${title}`,\r\n `> ${description}`,\r\n ``,\r\n `Source:* ${sourceUrl}`,\r\n `Extracted: ${new Date().toISOString()}`,\r\n ``,\r\n `---`,\r\n ``\r\n ];\r\n\r\n let finalMarkdown = headerLines.join('\\n') + markdown;\r\n\r\n // Add JSON-LD section if exists\r\n if (jsonLdScripts.length > 0) {\r\n finalMarkdown += '\\n\\n---\\n## Structured Data (JSON-LD)\\n```json\\n';\r\n jsonLdScripts.forEach(j => {\r\n finalMarkdown += JSON.stringify(j, null, 2) + '\\n';\r\n });\r\n finalMarkdown += '```\\n';\r\n }\r\n\r\n const markdownSize = finalMarkdown.length;\r\n const tokenReductionRatio = originalSize > 0 ? ((originalSize - markdownSize) / originalSize) * 100 : 0;\r\n\r\n return {\r\n markdown: finalMarkdown,\r\n metadata: {\r\n title,\r\n description,\r\n jsonLd: jsonLdScripts\r\n },\r\n stats: {\r\n originalHtmlSize: originalSize,\r\n markdownSize,\r\n tokenReductionRatio\r\n }\r\n };\r\n}\r\n\r\nexport async function generateStaticPayloads(nextAppDirDir: string, ontoPublicDir: string) {\r\n const fs = await import('fs');\r\n const path = await import('path');\r\n const { glob } = await import('glob');\r\n\r\n if (!fs.existsSync(nextAppDirDir)) {\r\n return;\r\n }\r\n\r\n const files = await glob('*/.html', { cwd: nextAppDirDir });\r\n if (files.length === 0) return;\r\n\r\n if (!fs.existsSync(ontoPublicDir)) {\r\n fs.mkdirSync(ontoPublicDir, { recursive: true });\r\n }\r\n\r\n let totalFilesProcessed = 0;\r\n\r\n for (const file of files) {\r\n const inputPath = path.join(nextAppDirDir, file);\r\n const outputPathRelative = file.replace(/\\.html$/, '.md');\r\n const outputPath = path.join(ontoPublicDir, outputPathRelative);\r\n\r\n try {\r\n const htmlContent = fs.readFileSync(inputPath, 'utf8');\r\n\r\n let routeName = file.replace(/\\.html$/, '');\r\n if (routeName === 'index') routeName = '/';\r\n else routeName = `/${routeName}`;\r\n\r\n const result = extractContent(htmlContent, routeName);\r\n\r\n const outputDir = path.dirname(outputPath);\r\n if (!fs.existsSync(outputDir)) {\r\n fs.mkdirSync(outputDir, { recursive: true });\r\n }\r\n\r\n fs.writeFileSync(outputPath, result.markdown, 'utf8');\r\n totalFilesProcessed++;\r\n } catch (e: any) {\r\n console.error(`[Onto] Failed to process ${file}: ${e.message}`);\r\n }\r\n }\r\n console.log(`[Onto] Successfully generated ${totalFilesProcessed} semantic markdown endpoints.`);\r\n}\r\n","/*\r\n Configuration schema for onto.config.ts\r\n * Used to dynamically generate llms.txt and other AI discovery files\r\n /\r\n\r\nexport type PageType = 'scoring' \| 'about' \| 'default';\r\n\r\nexport interface OntoRoute {\r\n /\r\n The URL path (e.g., '/docs', '/api/reference')\r\n /\r\n path: string;\r\n /\r\n Description of what this route contains\r\n /\r\n description: string;\r\n /\r\n Optional: Page type for automatic JSON-LD schema injection\r\n * - 'scoring': Injects Methodology schema with AIO scoring weights (40/35/25)\r\n * - 'about': Injects Organization/AboutPage schema\r\n * - 'default': No automatic schema injection\r\n /\r\n pageType?: PageType;\r\n}\r\n\r\nexport interface OntoConfig {\r\n /\r\n The name of your project or site (required)\r\n * Used as the H1 heading in llms.txt\r\n /\r\n name: string;\r\n\r\n /\r\n A short summary of your project (required)\r\n * Displayed as a blockquote in llms.txt\r\n * Should contain key information necessary for understanding the rest of the file\r\n /\r\n summary: string;\r\n\r\n /\r\n The base URL of your site (e.g., 'https://example.com')\r\n /\r\n baseUrl: string;\r\n\r\n /\r\n Optional: Additional sections to include in llms.txt\r\n * Each section can contain any markdown content\r\n /\r\n sections?: {\r\n heading: string;\r\n content: string;\r\n }[];\r\n\r\n /\r\n Key routes that AI agents should know about\r\n * These will be formatted as a markdown list in llms.txt\r\n /\r\n routes?: OntoRoute[];\r\n\r\n /\r\n Optional: Links to external resources (documentation, API references, etc.)\r\n /\r\n externalLinks?: {\r\n title: string;\r\n url: string;\r\n description?: string;\r\n }[];\r\n\r\n /\r\n Optional: Organization information for JSON-LD schemas\r\n /\r\n organization?: {\r\n name: string;\r\n description?: string;\r\n url?: string;\r\n logo?: string;\r\n foundingDate?: string;\r\n };\r\n}\r\n\r\n/\r\n Generate llms.txt content from OntoConfig\r\n * Follows the llms.txt specification:\r\n * - H1 with project name\r\n * - Blockquote with summary\r\n * - Additional markdown sections\r\n /\r\nexport function generateLlmsTxt(config: OntoConfig): string {\r\n const lines: string[] = [];\r\n\r\n // H1: Project name (required)\r\n lines.push(`# ${config.name}`);\r\n lines.push('');\r\n\r\n // Blockquote: Summary (required)\r\n lines.push(`> ${config.summary}`);\r\n lines.push('');\r\n\r\n // Key Routes section (if provided)\r\n if (config.routes && config.routes.length > 0) {\r\n lines.push('## Key Routes');\r\n lines.push('');\r\n for (const route of config.routes) {\r\n const fullUrl = `${config.baseUrl}${route.path}`;\r\n lines.push(`- [${route.path}](${fullUrl}): ${route.description}`);\r\n }\r\n lines.push('');\r\n }\r\n\r\n // External Links section (if provided)\r\n if (config.externalLinks && config.externalLinks.length > 0) {\r\n lines.push('## Resources');\r\n lines.push('');\r\n for (const link of config.externalLinks) {\r\n if (link.description) {\r\n lines.push(`- [${link.title}](${link.url}): ${link.description}`);\r\n } else {\r\n lines.push(`- [${link.title}](${link.url})`);\r\n }\r\n }\r\n lines.push('');\r\n }\r\n\r\n // Custom sections (if provided)\r\n if (config.sections && config.sections.length > 0) {\r\n for (const section of config.sections) {\r\n lines.push(`## ${section.heading}`);\r\n lines.push('');\r\n lines.push(section.content);\r\n lines.push('');\r\n }\r\n }\r\n\r\n return lines.join('\\n').trim() + '\\n';\r\n}\r\n","/\r\n JSON-LD Schema generators for automatic structured data injection\r\n * Follows Schema.org standards for AI-friendly metadata\r\n /\r\n\r\nimport { OntoConfig } from './config';\r\n\r\n/\r\n Standard AIO (AI Optimization) scoring methodology\r\n * Based on the Onto scoring algorithm:\r\n * - Content Negotiation: 40% (30 points)\r\n * - React Tax / Token Efficiency: 35% (30 points)\r\n * - Structured Data: 25% (25 points)\r\n * - Semantic HTML: Bonus (15 points)\r\n /\r\nexport interface AIOMethodologySchema {\r\n '@context': 'https://schema.org';\r\n '@type': 'HowTo';\r\n name: string;\r\n description: string;\r\n step: Array<{\r\n '@type': 'HowToStep';\r\n name: string;\r\n text: string;\r\n position: number;\r\n }>;\r\n}\r\n\r\n/\r\n Generate AIO Scoring Methodology JSON-LD schema\r\n * This explains to AI agents how the scoring system works\r\n /\r\nexport function generateAIOMethodologySchema(\r\n config: OntoConfig,\r\n pageUrl: string\r\n): AIOMethodologySchema {\r\n return {\r\n '@context': 'https://schema.org',\r\n '@type': 'HowTo',\r\n name: 'AIO Score Calculation Methodology',\r\n description: 'AI Optimization (AIO) Score measures how well a website is optimized for AI agents and LLM crawlers. Scored out of 100 points based on four key metrics.',\r\n step: [\r\n {\r\n '@type': 'HowToStep',\r\n name: 'Content Negotiation',\r\n text: 'Check if the site responds to Accept: text/markdown header. Weight: 40%. Penalty: -30 points if missing. This ensures AI bots receive optimized content instead of heavy HTML.',\r\n position: 1\r\n },\r\n {\r\n '@type': 'HowToStep',\r\n name: 'Token Efficiency (React Tax)',\r\n text: 'Measure the ratio of visible text to total HTML size. Weight: 35%. Penalty: -30 points if HTML > 100KB but text < 1KB. Detects JavaScript-heavy sites that are difficult for AI to parse.',\r\n position: 2\r\n },\r\n {\r\n '@type': 'HowToStep',\r\n name: 'Structured Data',\r\n text: 'Verify presence of JSON-LD structured data (Schema.org). Weight: 25%. Penalty: -25 points if missing. Enables AI to confidently extract pricing, products, and entities.',\r\n position: 3\r\n },\r\n {\r\n '@type': 'HowToStep',\r\n name: 'Semantic HTML',\r\n text: 'Check for semantic tags like <main> and <article>. Bonus: +15 points if present. Helps AI agents separate navigation from core content.',\r\n position: 4\r\n }\r\n ]\r\n };\r\n}\r\n\r\n/\r\n Organization schema for About pages\r\n /\r\nexport interface OrganizationSchema {\r\n '@context': 'https://schema.org';\r\n '@type': 'Organization';\r\n name: string;\r\n url?: string;\r\n description?: string;\r\n logo?: string;\r\n foundingDate?: string;\r\n}\r\n\r\n/\r\n Generate Organization JSON-LD schema for About pages\r\n /\r\nexport function generateOrganizationSchema(\r\n config: OntoConfig,\r\n pageUrl: string\r\n): OrganizationSchema \| null {\r\n if (!config.organization) {\r\n return null;\r\n }\r\n\r\n const schema: OrganizationSchema = {\r\n '@context': 'https://schema.org',\r\n '@type': 'Organization',\r\n name: config.organization.name\r\n };\r\n\r\n if (config.organization.url) {\r\n schema.url = config.organization.url;\r\n }\r\n\r\n if (config.organization.description) {\r\n schema.description = config.organization.description;\r\n }\r\n\r\n if (config.organization.logo) {\r\n schema.logo = config.organization.logo;\r\n }\r\n\r\n if (config.organization.foundingDate) {\r\n schema.foundingDate = config.organization.foundingDate;\r\n }\r\n\r\n return schema;\r\n}\r\n\r\n/\r\n AboutPage schema combining Organization and WebPage\r\n /\r\nexport interface AboutPageSchema {\r\n '@context': 'https://schema.org';\r\n '@type': 'AboutPage';\r\n name: string;\r\n url: string;\r\n description?: string;\r\n mainEntity?: OrganizationSchema;\r\n}\r\n\r\n/\r\n Generate AboutPage JSON-LD schema\r\n /\r\nexport function generateAboutPageSchema(\r\n config: OntoConfig,\r\n pageUrl: string\r\n): AboutPageSchema {\r\n const orgSchema = generateOrganizationSchema(config, pageUrl);\r\n\r\n const schema: AboutPageSchema = {\r\n '@context': 'https://schema.org',\r\n '@type': 'AboutPage',\r\n name: `About ${config.name}`,\r\n url: pageUrl\r\n };\r\n\r\n if (config.summary) {\r\n schema.description = config.summary;\r\n }\r\n\r\n if (orgSchema) {\r\n schema.mainEntity = orgSchema;\r\n }\r\n\r\n return schema;\r\n}\r\n\r\n/\r\n Determine which schema to generate based on page type\r\n /\r\nexport function generateSchemaForPageType(\r\n pageType: 'scoring' \| 'about' \| 'default',\r\n config: OntoConfig,\r\n pageUrl: string\r\n): any \| null {\r\n switch (pageType) {\r\n case 'scoring':\r\n return generateAIOMethodologySchema(config, pageUrl);\r\n case 'about':\r\n return generateAboutPageSchema(config, pageUrl);\r\n case 'default':\r\n default:\r\n return null;\r\n }\r\n}\r\n\r\n/\r\n Serialize schema to JSON-LD script tag content\r\n */\r\nexport function serializeSchema(schema: any \| null): string \| null {\r\n if (!schema) {\r\n return null;\r\n }\r\n return JSON.stringify(schema, null, 2);\r\n}\r\n"],"mappings":"AAAA,UAAYA,MAAa,UACzB,OAAOC,MAAqB,WAE5B,IAAMC,EAAkB,IAAID,EAAgB,CACxC,aAAc,MACd,eAAgB,QACpB,CAAC,EAwBM,SAASE,EAAeC,EAAcC,EAAoB,mBAAsC,CACnG,IAAMC,EAAeF,EAAK,OAEpBG,EAAY,OAAKH,CAAI,EAGrBI,EAAQD,EAAE,OAAO,EAAE,KAAK,GAAKA,EAAE,IAAI,EAAE,MAAM,EAAE,KAAK,GAAK,gBACvDE,EAAcF,EAAE,0BAA0B,EAAE,KAAK,SAAS,GAAK,wBAE/DG,EAAuB,CAAC,EAC9BH,EAAE,oCAAoC,EAAE,KAAK,CAACI,EAAGC,IAAO,CACpD,GAAI,CACA,IAAMC,EAAMN,EAAEK,CAAE,EAAE,KAAK,GAAK,GACtBE,EAAS,KAAK,MAAMD,CAAG,EAC7BH,EAAc,KAAKI,CAAM,CAC7B,MAAQ,CAER,CACJ,CAAC,EAGDP,EAAE,uEAAuE,EAAE,OAAO,EAOlF,IAAIQ,EAAc,GACdR,EAAE,MAAM,EAAE,OAAS,EACnBQ,EAAcR,EAAE,MAAM,EAAE,KAAK,GAAK,GAC3BA,EAAE,SAAS,EAAE,OAAS,EAC7BQ,EAAcR,EAAE,SAAS,EAAE,KAAK,GAAK,GAErCQ,EAAcR,EAAE,MAAM,EAAE,KAAK,GAAK,GAItC,IAAIS,EAAWd,EAAgB,SAASa,CAAW,EAc/CE,EAXgB,CAChB,KAAKT,CAAK,GACV,KAAKC,CAAW,GAChB,GACA,eAAeJ,CAAS,GACxB,kBAAkB,IAAI,KAAK,EAAE,YAAY,CAAC,GAC1C,GACA,MACA,EACJ,EAEgC,KAAK;AAAA,CAAI,EAAIW,EAGzCN,EAAc,OAAS,IACvBO,GAAiB;AAAA;AAAA;AAAA;AAAA;AAAA,EACjBP,EAAc,QAAQQ,GAAK,CACvBD,GAAiB,KAAK,UAAUC,EAAG,KAAM,CAAC,EAAI;AAAA,CAClD,CAAC,EACDD,GAAiB,SAGrB,IAAME,EAAeF,EAAc,OAC7BG,EAAsBd,EAAe,GAAMA,EAAea,GAAgBb,EAAgB,IAAM,EAEtG,MAAO,CACH,SAAUW,EACV,SAAU,CACN,MAAAT,EACA,YAAAC,EACA,OAAQC,CACZ,EACA,MAAO,CACH,iBAAkBJ,EAClB,aAAAa,EACA,oBAAAC,CACJ,CACJ,CACJ,CCtBO,SAASC,EAAgBC,EAA4B,CAC1D,IAAMC,EAAkB,CAAC,EAWzB,GARAA,EAAM,KAAK,KAAKD,EAAO,IAAI,EAAE,EAC7BC,EAAM,KAAK,EAAE,EAGbA,EAAM,KAAK,KAAKD,EAAO,OAAO,EAAE,EAChCC,EAAM,KAAK,EAAE,EAGTD,EAAO,QAAUA,EAAO,OAAO,OAAS,EAAG,CAC7CC,EAAM,KAAK,eAAe,EAC1BA,EAAM,KAAK,EAAE,EACb,QAAWC,KAASF,EAAO,OAAQ,CACjC,IAAMG,EAAU,GAAGH,EAAO,OAAO,GAAGE,EAAM,IAAI,GAC9CD,EAAM,KAAK,MAAMC,EAAM,IAAI,KAAKC,CAAO,MAAMD,EAAM,WAAW,EAAE,CAClE,CACAD,EAAM,KAAK,EAAE,CACf,CAGA,GAAID,EAAO,eAAiBA,EAAO,cAAc,OAAS,EAAG,CAC3DC,EAAM,KAAK,cAAc,EACzBA,EAAM,KAAK,EAAE,EACb,QAAWG,KAAQJ,EAAO,cACpBI,EAAK,YACPH,EAAM,KAAK,MAAMG,EAAK,KAAK,KAAKA,EAAK,GAAG,MAAMA,EAAK,WAAW,EAAE,EAEhEH,EAAM,KAAK,MAAMG,EAAK,KAAK,KAAKA,EAAK,GAAG,GAAG,EAG/CH,EAAM,KAAK,EAAE,CACf,CAGA,GAAID,EAAO,UAAYA,EAAO,SAAS,OAAS,EAC9C,QAAWK,KAAWL,EAAO,SAC3BC,EAAM,KAAK,MAAMI,EAAQ,OAAO,EAAE,EAClCJ,EAAM,KAAK,EAAE,EACbA,EAAM,KAAKI,EAAQ,OAAO,EAC1BJ,EAAM,KAAK,EAAE,EAIjB,OAAOA,EAAM,KAAK;AAAA,CAAI,EAAE,KAAK,EAAI;AAAA,CACnC,CCtGO,SAASK,EACdC,EACAC,EACsB,CACtB,MAAO,CACL,WAAY,qBACZ,QAAS,QACT,KAAM,oCACN,YAAa,2JACb,KAAM,CACJ,CACE,QAAS,YACT,KAAM,sBACN,KAAM,iLACN,SAAU,CACZ,EACA,CACE,QAAS,YACT,KAAM,+BACN,KAAM,4LACN,SAAU,CACZ,EACA,CACE,QAAS,YACT,KAAM,kBACN,KAAM,2KACN,SAAU,CACZ,EACA,CACE,QAAS,YACT,KAAM,gBACN,KAAM,0IACN,SAAU,CACZ,CACF,CACF,CACF,CAkBO,SAASC,EACdF,EACAC,EAC2B,CAC3B,GAAI,CAACD,EAAO,aACV,OAAO,KAGT,IAAMG,EAA6B,CACjC,WAAY,qBACZ,QAAS,eACT,KAAMH,EAAO,aAAa,IAC5B,EAEA,OAAIA,EAAO,aAAa,MACtBG,EAAO,IAAMH,EAAO,aAAa,KAG/BA,EAAO,aAAa,cACtBG,EAAO,YAAcH,EAAO,aAAa,aAGvCA,EAAO,aAAa,OACtBG,EAAO,KAAOH,EAAO,aAAa,MAGhCA,EAAO,aAAa,eACtBG,EAAO,aAAeH,EAAO,aAAa,cAGrCG,CACT,CAiBO,SAASC,EACdJ,EACAC,EACiB,CACjB,IAAMI,EAAYH,EAA2BF,EAAQC,CAAO,EAEtDE,EAA0B,CAC9B,WAAY,qBACZ,QAAS,YACT,KAAM,SAASH,EAAO,IAAI,GAC1B,IAAKC,CACP,EAEA,OAAID,EAAO,UACTG,EAAO,YAAcH,EAAO,SAG1BK,IACFF,EAAO,WAAaE,GAGfF,CACT,CAKO,SAASG,EACdC,EACAP,EACAC,EACY,CACZ,OAAQM,EAAU,CAChB,IAAK,UACH,OAAOR,EAA6BC,EAAQC,CAAO,EACrD,IAAK,QACH,OAAOG,EAAwBJ,EAAQC,CAAO,EAEhD,QACE,OAAO,IACX,CACF,CAKO,SAASO,EAAgBL,EAAmC,CACjE,OAAKA,EAGE,KAAK,UAAUA,EAAQ,KAAM,CAAC,EAF5B,IAGX","names":["cheerio","TurndownService","turndownService","extractContent","html","sourceUrl","originalSize","$","title","description","jsonLdScripts","_","el","raw","parsed","contentHtml","markdown","finalMarkdown","j","markdownSize","tokenReductionRatio","generateLlmsTxt","config","lines","route","fullUrl","link","section","generateAIOMethodologySchema","config","pageUrl","generateOrganizationSchema","schema","generateAboutPageSchema","orgSchema","generateSchemaForPageType","pageType","serializeSchema"]}

package/dist/middleware.js CHANGED Viewed

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

package/dist/middleware.js.map CHANGED Viewed

	@@ -1 +1 @@
1	- {"version":3,"sources":["../src/middleware.ts","../src/bots.ts","../src/config.ts"],"sourcesContent":["import { NextRequest, NextResponse } from 'next/server';\r\nimport { matchBot } from './bots';\r\nimport { loadOntoConfig, generateLlmsTxt, OntoConfig } from './config';\r\n\r\nexport async function ontoMiddleware(request: NextRequest, config?: OntoConfig) {\r\n const userAgent = request.headers.get('user-agent');\r\n const url = request.nextUrl.clone();\r\n const matched = matchBot(userAgent);\r\n\r\n const accept = request.headers.get('accept') \|\| '';\r\n\r\n const isAiBot = !!matched;\r\n const isMarkdownRequested = accept.includes('text/markdown');\r\n\r\n // If traffic is identified as an AI Bot or markdown is requested\r\n if (isAiBot \|\| isMarkdownRequested) {\r\n\r\n // Ignore internal next.js requests & static assets (but not llms.txt)\r\n if (url.pathname.startsWith('/_next')) {\r\n return NextResponse.next();\r\n }\r\n\r\n // --- llms.txt Auto-Discovery ---\r\n // Serve the llms.txt manifest to AI agents\r\n if (url.pathname === '/llms.txt') {\r\n try {\r\n // Use provided config or try to load dynamic one\r\n const activeConfig = config \|\| await loadOntoConfig();\r\n\r\n if (activeConfig) {\r\n // Generate llms.txt dynamically from config\r\n const llmsTxtContent = generateLlmsTxt(activeConfig);\r\n const response = new NextResponse(llmsTxtContent, {\r\n headers: {\r\n 'Content-Type': 'text/plain; charset=utf-8',\r\n 'Cache-Control': 'public, max-age=3600, s-maxage=3600, stale-while-revalidate=86400',\r\n }\r\n });\r\n\r\n if (matched) {\r\n response.headers.set('X-Onto-Bot', `${matched.name} (${matched.company})`);\r\n }\r\n\r\n return response;\r\n } else {\r\n // Fallback: try to serve static llms.txt from public folder\r\n url.pathname = '/llms.txt';\r\n const response = NextResponse.rewrite(url);\r\n response.headers.set('Content-Type', 'text/plain; charset=utf-8');\r\n response.headers.set('Cache-Control', 'public, max-age=3600, s-maxage=3600, stale-while-revalidate=86400');\r\n if (matched) {\r\n response.headers.set('X-Onto-Bot', `${matched.name} (${matched.company})`);\r\n }\r\n return response;\r\n }\r\n } catch (error) {\r\n console.error('[Onto] Failed to generate llms.txt:', error);\r\n // Fallback to static file on error\r\n url.pathname = '/llms.txt';\r\n const response = NextResponse.rewrite(url);\r\n response.headers.set('Content-Type', 'text/plain; charset=utf-8');\r\n if (matched) {\r\n response.headers.set('X-Onto-Bot', `${matched.name} (${matched.company})`);\r\n }\r\n return response;\r\n }\r\n }\r\n\r\n // Skip other static assets\r\n if (url.pathname.includes('.')) {\r\n return NextResponse.next();\r\n }\r\n\r\n // Determine the corresponding payload path\r\n let payloadPath = url.pathname;\r\n if (payloadPath === '/' \|\| payloadPath === '') {\r\n payloadPath = '/index';\r\n }\r\n\r\n // Strip trailing slash if present\r\n if (payloadPath.endsWith('/') && payloadPath !== '/') {\r\n payloadPath = payloadPath.slice(0, -1);\r\n }\r\n\r\n // Common response headers for all bot responses\r\n const botHeaders: Record<string, string> = {\r\n 'Content-Type': 'text/markdown; charset=utf-8',\r\n 'Cache-Control': 'public, max-age=3600, s-maxage=3600, stale-while-revalidate=86400',\r\n };\r\n if (matched) {\r\n botHeaders['X-Onto-Bot'] = `${matched.name} (${matched.company})`;\r\n }\r\n\r\n // --- Onto Control Plane Integration (Premium) ---\r\n const ONTO_API_KEY = process.env.ONTO_API_KEY;\r\n const DASHBOARD_URL = process.env.ONTO_DASHBOARD_URL \|\| 'https://app.buildonto.dev';\r\n\r\n if (ONTO_API_KEY) {\r\n // 1. Fire-and-forget tracking — includes structured bot info\r\n fetch(`${DASHBOARD_URL}/api/track`, {\r\n method: 'POST',\r\n headers: {\r\n 'x-onto-key': ONTO_API_KEY,\r\n 'Content-Type': 'application/json'\r\n },\r\n body: JSON.stringify({\r\n route: url.pathname,\r\n userAgent: userAgent,\r\n bot: matched ? matched.name : null,\r\n company: matched ? matched.company : null,\r\n })\r\n }).catch(() => {});\r\n\r\n // 2. Dynamic Context Injection\r\n try {\r\n const injectRes = await fetch(`${DASHBOARD_URL}/api/sdk/inject?route=${url.pathname}`, {\r\n headers: { 'x-onto-key': ONTO_API_KEY },\r\n signal: AbortSignal.timeout(1500)\r\n });\r\n\r\n if (injectRes.ok) {\r\n const { injection } = await injectRes.json();\r\n \r\n if (injection) {\r\n const localMdUrl = `${url.origin}/.onto${payloadPath}.md`;\r\n const mdRes = await fetch(localMdUrl);\r\n \r\n if (mdRes.ok) {\r\n const baseMarkdown = await mdRes.text();\r\n const finalMarkdown = `${baseMarkdown}\\n\\n---\\n\\n${injection}`;\r\n \r\n return new NextResponse(finalMarkdown, {\r\n headers: {\r\n ...botHeaders,\r\n 'X-Onto-Injected': 'true'\r\n }\r\n });\r\n }\r\n }\r\n }\r\n } catch (err) {\r\n console.error('[Onto] Injection failed', err);\r\n }\r\n }\r\n // ------------------------------------------------\r\n\r\n url.pathname = `/.onto${payloadPath}.md`;\r\n\r\n // Rewrite implicitly serves the target URL transparently to the client.\r\n const response = NextResponse.rewrite(url);\r\n if (matched) {\r\n response.headers.set('X-Onto-Bot', `${matched.name} (${matched.company})`);\r\n }\r\n return response;\r\n }\r\n\r\n return NextResponse.next();\r\n}\r\n\r\n// Re-export the bot registry for consumers who want to extend or inspect it\r\nexport { AI_BOT_USER_AGENTS, matchBot } from './bots';\r\nexport type { AiBot } from './bots';\r\n\r\n\r\n","/*\n Comprehensive registry of AI bot user-agent strings.\n * The middleware uses this list to detect AI crawlers and serve optimized markdown.\n /\n\nexport interface AiBot {\n /* The user-agent substring to match against /\n name: string;\n /* The company operating this bot /\n company: string;\n}\n\n/\n Structured registry of all known AI bots, grouped by company.\n * Useful for analytics and the Control Plane dashboard.\n /\nexport const AI_BOTS: AiBot[] = [\n // OpenAI\n { name: 'GPTBot', company: 'OpenAI' },\n { name: 'ChatGPT-User', company: 'OpenAI' },\n { name: 'OAI-SearchBot', company: 'OpenAI' },\n\n // Google\n { name: 'Googlebot', company: 'Google' },\n { name: 'Google-CloudVertexBot', company: 'Google' },\n { name: 'Google-Extended', company: 'Google' },\n { name: 'GoogleOther', company: 'Google' },\n\n // Anthropic\n { name: 'ClaudeBot', company: 'Anthropic' },\n { name: 'Claude-User', company: 'Anthropic' },\n { name: 'anthropic-ai', company: 'Anthropic' },\n\n // Perplexity\n { name: 'PerplexityBot', company: 'Perplexity' },\n { name: 'Perplexity-User', company: 'Perplexity' },\n\n // Meta\n { name: 'Meta-ExternalAgent', company: 'Meta' },\n { name: 'Meta-ExternalFetcher', company: 'Meta' },\n { name: 'FacebookBot', company: 'Meta' },\n\n // Common Crawl (used by most smaller AI companies)\n { name: 'CCBot', company: 'Common Crawl' },\n\n // Other notable AI crawlers\n { name: 'Bytespider', company: 'ByteDance' },\n { name: 'Applebot-Extended', company: 'Apple' },\n { name: 'cohere-ai', company: 'Cohere' },\n { name: 'YouBot', company: 'You.com' },\n];\n\n/\n Flat list of user-agent substrings for fast matching in the middleware.\n /\nexport const AI_BOT_USER_AGENTS: string[] = AI_BOTS.map(bot => bot.name);\n\n/\n Given a raw user-agent string, returns the matched AiBot entry or undefined.\n * Comparison is case-insensitive to handle inconsistent agent casing.\n /\nexport function matchBot(userAgent: string \| null): AiBot \| undefined {\n if (!userAgent) return undefined;\n const lowerUA = userAgent.toLowerCase();\n return AI_BOTS.find(bot => \n lowerUA.includes(bot.name.toLowerCase())\n );\n}\n","/\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,wBAAAE,EAAA,aAAAC,EAAA,mBAAAC,IAAA,eAAAC,EAAAL,GAAA,IAAAM,EAA0C,uBCgBnC,IAAMC,EAAmB,CAE5B,CAAE,KAAM,SAAqB,QAAS,QAAS,EAC/C,CAAE,KAAM,eAAoB,QAAS,QAAS,EAC9C,CAAE,KAAM,gBAAoB,QAAS,QAAS,EAG9C,CAAE,KAAM,YAA0B,QAAS,QAAS,EACpD,CAAE,KAAM,wBAA2B,QAAS,QAAS,EACrD,CAAE,KAAM,kBAA2B,QAAS,QAAS,EACrD,CAAE,KAAM,cAA2B,QAAS,QAAS,EAGrD,CAAE,KAAM,YAAmB,QAAS,WAAY,EAChD,CAAE,KAAM,cAAkB,QAAS,WAAY,EAC/C,CAAE,KAAM,eAAkB,QAAS,WAAY,EAG/C,CAAE,KAAM,gBAAmB,QAAS,YAAa,EACjD,CAAE,KAAM,kBAAmB,QAAS,YAAa,EAGjD,CAAE,KAAM,qBAAwB,QAAS,MAAO,EAChD,CAAE,KAAM,uBAAwB,QAAS,MAAO,EAChD,CAAE,KAAM,cAAuB,QAAS,MAAO,EAG/C,CAAE,KAAM,QAAS,QAAS,cAAe,EAGzC,CAAE,KAAM,aAAqB,QAAS,WAAY,EAClD,CAAE,KAAM,oBAAqB,QAAS,OAAQ,EAC9C,CAAE,KAAM,YAAoB,QAAS,QAAS,EAC9C,CAAE,KAAM,SAAoB,QAAS,SAAU,CACnD,EAKaC,EAA+BD,EAAQ,IAAIE,GAAOA,EAAI,IAAI,EAMhE,SAASC,EAASC,EAA6C,CAClE,GAAI,CAACA,EAAW,OAChB,IAAMC,EAAUD,EAAU,YAAY,EACtC,OAAOJ,EAAQ,KAAKE,GAChBG,EAAQ,SAASH,EAAI,KAAK,YAAY,CAAC,CAC3C,CACJ,CCiBA,eAAsBI,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,CFlJA,eAAsBK,EAAeC,EAAsBC,EAAqB,CAC5E,IAAMC,EAAYF,EAAQ,QAAQ,IAAI,YAAY,EAC5CG,EAAMH,EAAQ,QAAQ,MAAM,EAC5BI,EAAUC,EAASH,CAAS,EAE5BI,EAASN,EAAQ,QAAQ,IAAI,QAAQ,GAAK,GAE1CO,EAAU,CAAC,CAACH,EACZI,EAAsBF,EAAO,SAAS,eAAe,EAG3D,GAAIC,GAAWC,EAAqB,CAGhC,GAAIL,EAAI,SAAS,WAAW,QAAQ,EAChC,OAAO,eAAa,KAAK,EAK7B,GAAIA,EAAI,WAAa,YACjB,GAAI,CAEA,IAAMM,EAAeR,GAAU,MAAMS,EAAe,EAEpD,GAAID,EAAc,CAEd,IAAME,EAAiBC,EAAgBH,CAAY,EAC7CI,EAAW,IAAI,eAAaF,EAAgB,CAC9C,QAAS,CACL,eAAgB,4BAChB,gBAAiB,mEACrB,CACJ,CAAC,EAED,OAAIP,GACAS,EAAS,QAAQ,IAAI,aAAc,GAAGT,EAAQ,IAAI,KAAKA,EAAQ,OAAO,GAAG,EAGtES,CACX,KAAO,CAEHV,EAAI,SAAW,YACf,IAAMU,EAAW,eAAa,QAAQV,CAAG,EACzC,OAAAU,EAAS,QAAQ,IAAI,eAAgB,2BAA2B,EAChEA,EAAS,QAAQ,IAAI,gBAAiB,mEAAmE,EACrGT,GACAS,EAAS,QAAQ,IAAI,aAAc,GAAGT,EAAQ,IAAI,KAAKA,EAAQ,OAAO,GAAG,EAEtES,CACX,CACJ,OAASC,EAAO,CACZ,QAAQ,MAAM,sCAAuCA,CAAK,EAE1DX,EAAI,SAAW,YACf,IAAMU,EAAW,eAAa,QAAQV,CAAG,EACzC,OAAAU,EAAS,QAAQ,IAAI,eAAgB,2BAA2B,EAC5DT,GACAS,EAAS,QAAQ,IAAI,aAAc,GAAGT,EAAQ,IAAI,KAAKA,EAAQ,OAAO,GAAG,EAEtES,CACX,CAIJ,GAAIV,EAAI,SAAS,SAAS,GAAG,EACzB,OAAO,eAAa,KAAK,EAI7B,IAAIY,EAAcZ,EAAI,UAClBY,IAAgB,KAAOA,IAAgB,MACvCA,EAAc,UAIdA,EAAY,SAAS,GAAG,GAAKA,IAAgB,MAC7CA,EAAcA,EAAY,MAAM,EAAG,EAAE,GAIzC,IAAMC,EAAqC,CACvC,eAAgB,+BAChB,gBAAiB,mEACrB,EACIZ,IACAY,EAAW,YAAY,EAAI,GAAGZ,EAAQ,IAAI,KAAKA,EAAQ,OAAO,KAIlE,IAAMa,EAAe,QAAQ,IAAI,aAC3BC,EAAgB,QAAQ,IAAI,oBAAsB,4BAExD,GAAID,EAAc,CAEd,MAAM,GAAGC,CAAa,aAAc,CAChC,OAAQ,OACR,QAAS,CACL,aAAcD,EACd,eAAgB,kBACpB,EACA,KAAM,KAAK,UAAU,CACjB,MAAOd,EAAI,SACX,UAAWD,EACX,IAAKE,EAAUA,EAAQ,KAAO,KAC9B,QAASA,EAAUA,EAAQ,QAAU,IACzC,CAAC,CACL,CAAC,EAAE,MAAM,IAAM,CAAC,CAAC,EAGjB,GAAI,CACA,IAAMe,EAAY,MAAM,MAAM,GAAGD,CAAa,yBAAyBf,EAAI,QAAQ,GAAI,CACnF,QAAS,CAAE,aAAcc,CAAa,EACtC,OAAQ,YAAY,QAAQ,IAAI,CACpC,CAAC,EAED,GAAIE,EAAU,GAAI,CACd,GAAM,CAAE,UAAAC,CAAU,EAAI,MAAMD,EAAU,KAAK,EAE3C,GAAIC,EAAW,CACX,IAAMC,EAAa,GAAGlB,EAAI,MAAM,SAASY,CAAW,MAC9CO,EAAQ,MAAM,MAAMD,CAAU,EAEpC,GAAIC,EAAM,GAAI,CAEV,IAAMC,EAAgB,GADD,MAAMD,EAAM,KAAK,CACD;AAAA;AAAA;AAAA;AAAA,EAAcF,CAAS,GAE5D,OAAO,IAAI,eAAaG,EAAe,CACnC,QAAS,CACL,GAAGP,EACH,kBAAmB,MACvB,CACJ,CAAC,CACL,CACJ,CACJ,CACJ,OAASQ,EAAK,CACV,QAAQ,MAAM,0BAA2BA,CAAG,CAChD,CACJ,CAGArB,EAAI,SAAW,SAASY,CAAW,MAGnC,IAAMF,EAAW,eAAa,QAAQV,CAAG,EACzC,OAAIC,GACAS,EAAS,QAAQ,IAAI,aAAc,GAAGT,EAAQ,IAAI,KAAKA,EAAQ,OAAO,GAAG,EAEtES,CACX,CAEA,OAAO,eAAa,KAAK,CAC7B","names":["middleware_exports","__export","AI_BOT_USER_AGENTS","matchBot","ontoMiddleware","__toCommonJS","import_server","AI_BOTS","AI_BOT_USER_AGENTS","bot","matchBot","userAgent","lowerUA","loadOntoConfig","config","generateLlmsTxt","lines","route","fullUrl","link","section","ontoMiddleware","request","config","userAgent","url","matched","matchBot","accept","isAiBot","isMarkdownRequested","activeConfig","loadOntoConfig","llmsTxtContent","generateLlmsTxt","response","error","payloadPath","botHeaders","ONTO_API_KEY","DASHBOARD_URL","injectRes","injection","localMdUrl","mdRes","finalMarkdown","err"]}
1	+ {"version":3,"sources":["../src/middleware.ts","../src/bots.ts","../src/config.ts"],"sourcesContent":["import { NextRequest, NextResponse } from 'next/server';\r\nimport { matchBot } from './bots';\r\nimport { generateLlmsTxt, OntoConfig } from './config';\r\n\r\nexport async function ontoMiddleware(request: NextRequest, config?: OntoConfig) {\r\n const userAgent = request.headers.get('user-agent');\r\n const url = request.nextUrl.clone();\r\n const matched = matchBot(userAgent);\r\n\r\n const accept = request.headers.get('accept') \|\| '';\r\n\r\n const isAiBot = !!matched;\r\n const isMarkdownRequested = accept.includes('text/markdown');\r\n\r\n // If traffic is identified as an AI Bot or markdown is requested\r\n if (isAiBot \|\| isMarkdownRequested) {\r\n\r\n // Ignore internal next.js requests & static assets (but not llms.txt)\r\n if (url.pathname.startsWith('/_next')) {\r\n return NextResponse.next();\r\n }\r\n\r\n // --- llms.txt Auto-Discovery ---\r\n // Serve the llms.txt manifest to AI agents\r\n if (url.pathname === '/llms.txt') {\r\n try {\r\n if (config) {\r\n // Generate llms.txt dynamically from config\r\n const llmsTxtContent = generateLlmsTxt(config);\r\n const response = new NextResponse(llmsTxtContent, {\r\n headers: {\r\n 'Content-Type': 'text/plain; charset=utf-8',\r\n 'Cache-Control': 'public, max-age=3600, s-maxage=3600, stale-while-revalidate=86400',\r\n }\r\n });\r\n\r\n if (matched) {\r\n response.headers.set('X-Onto-Bot', `${matched.name} (${matched.company})`);\r\n }\r\n\r\n return response;\r\n } else {\r\n // Fallback: try to serve static llms.txt from public folder\r\n url.pathname = '/llms.txt';\r\n const response = NextResponse.rewrite(url);\r\n response.headers.set('Content-Type', 'text/plain; charset=utf-8');\r\n response.headers.set('Cache-Control', 'public, max-age=3600, s-maxage=3600, stale-while-revalidate=86400');\r\n if (matched) {\r\n response.headers.set('X-Onto-Bot', `${matched.name} (${matched.company})`);\r\n }\r\n return response;\r\n }\r\n } catch (error) {\r\n console.error('[Onto] Failed to generate llms.txt:', error);\r\n // Fallback to static file on error\r\n url.pathname = '/llms.txt';\r\n const response = NextResponse.rewrite(url);\r\n response.headers.set('Content-Type', 'text/plain; charset=utf-8');\r\n if (matched) {\r\n response.headers.set('X-Onto-Bot', `${matched.name} (${matched.company})`);\r\n }\r\n return response;\r\n }\r\n }\r\n\r\n // Skip other static assets\r\n if (url.pathname.includes('.')) {\r\n return NextResponse.next();\r\n }\r\n\r\n // Determine the corresponding payload path\r\n let payloadPath = url.pathname;\r\n if (payloadPath === '/' \|\| payloadPath === '') {\r\n payloadPath = '/index';\r\n }\r\n\r\n // Strip trailing slash if present\r\n if (payloadPath.endsWith('/') && payloadPath !== '/') {\r\n payloadPath = payloadPath.slice(0, -1);\r\n }\r\n\r\n // Common response headers for all bot responses\r\n const botHeaders: Record<string, string> = {\r\n 'Content-Type': 'text/markdown; charset=utf-8',\r\n 'Cache-Control': 'public, max-age=3600, s-maxage=3600, stale-while-revalidate=86400',\r\n };\r\n if (matched) {\r\n botHeaders['X-Onto-Bot'] = `${matched.name} (${matched.company})`;\r\n }\r\n\r\n // --- Onto Control Plane Integration (Premium) ---\r\n const ONTO_API_KEY = process.env.ONTO_API_KEY;\r\n const DASHBOARD_URL = process.env.ONTO_DASHBOARD_URL \|\| 'https://app.buildonto.dev';\r\n\r\n if (ONTO_API_KEY) {\r\n // 1. Fire-and-forget tracking — includes structured bot info\r\n fetch(`${DASHBOARD_URL}/api/track`, {\r\n method: 'POST',\r\n headers: {\r\n 'x-onto-key': ONTO_API_KEY,\r\n 'Content-Type': 'application/json'\r\n },\r\n body: JSON.stringify({\r\n route: url.pathname,\r\n userAgent: userAgent,\r\n bot: matched ? matched.name : null,\r\n company: matched ? matched.company : null,\r\n })\r\n }).catch(() => {});\r\n\r\n // 2. Dynamic Context Injection\r\n try {\r\n const injectRes = await fetch(`${DASHBOARD_URL}/api/sdk/inject?route=${url.pathname}`, {\r\n headers: { 'x-onto-key': ONTO_API_KEY },\r\n signal: AbortSignal.timeout(1500)\r\n });\r\n\r\n if (injectRes.ok) {\r\n const { injection } = await injectRes.json();\r\n \r\n if (injection) {\r\n const localMdUrl = `${url.origin}/.onto${payloadPath}.md`;\r\n const mdRes = await fetch(localMdUrl);\r\n \r\n if (mdRes.ok) {\r\n const baseMarkdown = await mdRes.text();\r\n const finalMarkdown = `${baseMarkdown}\\n\\n---\\n\\n${injection}`;\r\n \r\n return new NextResponse(finalMarkdown, {\r\n headers: {\r\n ...botHeaders,\r\n 'X-Onto-Injected': 'true'\r\n }\r\n });\r\n }\r\n }\r\n }\r\n } catch (err) {\r\n console.error('[Onto] Injection failed', err);\r\n }\r\n }\r\n // ------------------------------------------------\r\n\r\n url.pathname = `/.onto${payloadPath}.md`;\r\n\r\n // Rewrite implicitly serves the target URL transparently to the client.\r\n const response = NextResponse.rewrite(url);\r\n if (matched) {\r\n response.headers.set('X-Onto-Bot', `${matched.name} (${matched.company})`);\r\n }\r\n return response;\r\n }\r\n\r\n return NextResponse.next();\r\n}\r\n\r\n// Re-export the bot registry for consumers who want to extend or inspect it\r\nexport { AI_BOT_USER_AGENTS, matchBot } from './bots';\r\nexport type { AiBot } from './bots';\r\n\r\n\r\n","/*\n Comprehensive registry of AI bot user-agent strings.\n * The middleware uses this list to detect AI crawlers and serve optimized markdown.\n /\n\nexport interface AiBot {\n /* The user-agent substring to match against /\n name: string;\n /* The company operating this bot /\n company: string;\n}\n\n/\n Structured registry of all known AI bots, grouped by company.\n * Useful for analytics and the Control Plane dashboard.\n /\nexport const AI_BOTS: AiBot[] = [\n // OpenAI\n { name: 'GPTBot', company: 'OpenAI' },\n { name: 'ChatGPT-User', company: 'OpenAI' },\n { name: 'OAI-SearchBot', company: 'OpenAI' },\n\n // Google\n { name: 'Googlebot', company: 'Google' },\n { name: 'Google-CloudVertexBot', company: 'Google' },\n { name: 'Google-Extended', company: 'Google' },\n { name: 'GoogleOther', company: 'Google' },\n\n // Anthropic\n { name: 'ClaudeBot', company: 'Anthropic' },\n { name: 'Claude-User', company: 'Anthropic' },\n { name: 'anthropic-ai', company: 'Anthropic' },\n\n // Perplexity\n { name: 'PerplexityBot', company: 'Perplexity' },\n { name: 'Perplexity-User', company: 'Perplexity' },\n\n // Meta\n { name: 'Meta-ExternalAgent', company: 'Meta' },\n { name: 'Meta-ExternalFetcher', company: 'Meta' },\n { name: 'FacebookBot', company: 'Meta' },\n\n // Common Crawl (used by most smaller AI companies)\n { name: 'CCBot', company: 'Common Crawl' },\n\n // Other notable AI crawlers\n { name: 'Bytespider', company: 'ByteDance' },\n { name: 'Applebot-Extended', company: 'Apple' },\n { name: 'cohere-ai', company: 'Cohere' },\n { name: 'YouBot', company: 'You.com' },\n];\n\n/\n Flat list of user-agent substrings for fast matching in the middleware.\n /\nexport const AI_BOT_USER_AGENTS: string[] = AI_BOTS.map(bot => bot.name);\n\n/\n Given a raw user-agent string, returns the matched AiBot entry or undefined.\n * Comparison is case-insensitive to handle inconsistent agent casing.\n /\nexport function matchBot(userAgent: string \| null): AiBot \| undefined {\n if (!userAgent) return undefined;\n const lowerUA = userAgent.toLowerCase();\n return AI_BOTS.find(bot => \n lowerUA.includes(bot.name.toLowerCase())\n );\n}\n","/\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,wBAAAE,EAAA,aAAAC,EAAA,mBAAAC,IAAA,eAAAC,EAAAL,GAAA,IAAAM,EAA0C,uBCgBnC,IAAMC,EAAmB,CAE5B,CAAE,KAAM,SAAqB,QAAS,QAAS,EAC/C,CAAE,KAAM,eAAoB,QAAS,QAAS,EAC9C,CAAE,KAAM,gBAAoB,QAAS,QAAS,EAG9C,CAAE,KAAM,YAA0B,QAAS,QAAS,EACpD,CAAE,KAAM,wBAA2B,QAAS,QAAS,EACrD,CAAE,KAAM,kBAA2B,QAAS,QAAS,EACrD,CAAE,KAAM,cAA2B,QAAS,QAAS,EAGrD,CAAE,KAAM,YAAmB,QAAS,WAAY,EAChD,CAAE,KAAM,cAAkB,QAAS,WAAY,EAC/C,CAAE,KAAM,eAAkB,QAAS,WAAY,EAG/C,CAAE,KAAM,gBAAmB,QAAS,YAAa,EACjD,CAAE,KAAM,kBAAmB,QAAS,YAAa,EAGjD,CAAE,KAAM,qBAAwB,QAAS,MAAO,EAChD,CAAE,KAAM,uBAAwB,QAAS,MAAO,EAChD,CAAE,KAAM,cAAuB,QAAS,MAAO,EAG/C,CAAE,KAAM,QAAS,QAAS,cAAe,EAGzC,CAAE,KAAM,aAAqB,QAAS,WAAY,EAClD,CAAE,KAAM,oBAAqB,QAAS,OAAQ,EAC9C,CAAE,KAAM,YAAoB,QAAS,QAAS,EAC9C,CAAE,KAAM,SAAoB,QAAS,SAAU,CACnD,EAKaC,EAA+BD,EAAQ,IAAIE,GAAOA,EAAI,IAAI,EAMhE,SAASC,EAASC,EAA6C,CAClE,GAAI,CAACA,EAAW,OAChB,IAAMC,EAAUD,EAAU,YAAY,EACtC,OAAOJ,EAAQ,KAAKE,GAChBG,EAAQ,SAASH,EAAI,KAAK,YAAY,CAAC,CAC3C,CACJ,CCoBO,SAASI,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,CFlIA,eAAsBK,EAAeC,EAAsBC,EAAqB,CAC5E,IAAMC,EAAYF,EAAQ,QAAQ,IAAI,YAAY,EAC5CG,EAAMH,EAAQ,QAAQ,MAAM,EAC5BI,EAAUC,EAASH,CAAS,EAE5BI,EAASN,EAAQ,QAAQ,IAAI,QAAQ,GAAK,GAE1CO,EAAU,CAAC,CAACH,EACZI,EAAsBF,EAAO,SAAS,eAAe,EAG3D,GAAIC,GAAWC,EAAqB,CAGhC,GAAIL,EAAI,SAAS,WAAW,QAAQ,EAChC,OAAO,eAAa,KAAK,EAK7B,GAAIA,EAAI,WAAa,YACjB,GAAI,CACA,GAAIF,EAAQ,CAER,IAAMQ,EAAiBC,EAAgBT,CAAM,EACvCU,EAAW,IAAI,eAAaF,EAAgB,CAC9C,QAAS,CACL,eAAgB,4BAChB,gBAAiB,mEACrB,CACJ,CAAC,EAED,OAAIL,GACAO,EAAS,QAAQ,IAAI,aAAc,GAAGP,EAAQ,IAAI,KAAKA,EAAQ,OAAO,GAAG,EAGtEO,CACX,KAAO,CAEHR,EAAI,SAAW,YACf,IAAMQ,EAAW,eAAa,QAAQR,CAAG,EACzC,OAAAQ,EAAS,QAAQ,IAAI,eAAgB,2BAA2B,EAChEA,EAAS,QAAQ,IAAI,gBAAiB,mEAAmE,EACrGP,GACAO,EAAS,QAAQ,IAAI,aAAc,GAAGP,EAAQ,IAAI,KAAKA,EAAQ,OAAO,GAAG,EAEtEO,CACX,CACJ,OAASC,EAAO,CACZ,QAAQ,MAAM,sCAAuCA,CAAK,EAE1DT,EAAI,SAAW,YACf,IAAMQ,EAAW,eAAa,QAAQR,CAAG,EACzC,OAAAQ,EAAS,QAAQ,IAAI,eAAgB,2BAA2B,EAC5DP,GACAO,EAAS,QAAQ,IAAI,aAAc,GAAGP,EAAQ,IAAI,KAAKA,EAAQ,OAAO,GAAG,EAEtEO,CACX,CAIJ,GAAIR,EAAI,SAAS,SAAS,GAAG,EACzB,OAAO,eAAa,KAAK,EAI7B,IAAIU,EAAcV,EAAI,UAClBU,IAAgB,KAAOA,IAAgB,MACvCA,EAAc,UAIdA,EAAY,SAAS,GAAG,GAAKA,IAAgB,MAC7CA,EAAcA,EAAY,MAAM,EAAG,EAAE,GAIzC,IAAMC,EAAqC,CACvC,eAAgB,+BAChB,gBAAiB,mEACrB,EACIV,IACAU,EAAW,YAAY,EAAI,GAAGV,EAAQ,IAAI,KAAKA,EAAQ,OAAO,KAIlE,IAAMW,EAAe,QAAQ,IAAI,aAC3BC,EAAgB,QAAQ,IAAI,oBAAsB,4BAExD,GAAID,EAAc,CAEd,MAAM,GAAGC,CAAa,aAAc,CAChC,OAAQ,OACR,QAAS,CACL,aAAcD,EACd,eAAgB,kBACpB,EACA,KAAM,KAAK,UAAU,CACjB,MAAOZ,EAAI,SACX,UAAWD,EACX,IAAKE,EAAUA,EAAQ,KAAO,KAC9B,QAASA,EAAUA,EAAQ,QAAU,IACzC,CAAC,CACL,CAAC,EAAE,MAAM,IAAM,CAAC,CAAC,EAGjB,GAAI,CACA,IAAMa,EAAY,MAAM,MAAM,GAAGD,CAAa,yBAAyBb,EAAI,QAAQ,GAAI,CACnF,QAAS,CAAE,aAAcY,CAAa,EACtC,OAAQ,YAAY,QAAQ,IAAI,CACpC,CAAC,EAED,GAAIE,EAAU,GAAI,CACd,GAAM,CAAE,UAAAC,CAAU,EAAI,MAAMD,EAAU,KAAK,EAE3C,GAAIC,EAAW,CACX,IAAMC,EAAa,GAAGhB,EAAI,MAAM,SAASU,CAAW,MAC9CO,EAAQ,MAAM,MAAMD,CAAU,EAEpC,GAAIC,EAAM,GAAI,CAEV,IAAMC,EAAgB,GADD,MAAMD,EAAM,KAAK,CACD;AAAA;AAAA;AAAA;AAAA,EAAcF,CAAS,GAE5D,OAAO,IAAI,eAAaG,EAAe,CACnC,QAAS,CACL,GAAGP,EACH,kBAAmB,MACvB,CACJ,CAAC,CACL,CACJ,CACJ,CACJ,OAASQ,EAAK,CACV,QAAQ,MAAM,0BAA2BA,CAAG,CAChD,CACJ,CAGAnB,EAAI,SAAW,SAASU,CAAW,MAGnC,IAAMF,EAAW,eAAa,QAAQR,CAAG,EACzC,OAAIC,GACAO,EAAS,QAAQ,IAAI,aAAc,GAAGP,EAAQ,IAAI,KAAKA,EAAQ,OAAO,GAAG,EAEtEO,CACX,CAEA,OAAO,eAAa,KAAK,CAC7B","names":["middleware_exports","__export","AI_BOT_USER_AGENTS","matchBot","ontoMiddleware","__toCommonJS","import_server","AI_BOTS","AI_BOT_USER_AGENTS","bot","matchBot","userAgent","lowerUA","generateLlmsTxt","config","lines","route","fullUrl","link","section","ontoMiddleware","request","config","userAgent","url","matched","matchBot","accept","isAiBot","isMarkdownRequested","llmsTxtContent","generateLlmsTxt","response","error","payloadPath","botHeaders","ONTO_API_KEY","DASHBOARD_URL","injectRes","injection","localMdUrl","mdRes","finalMarkdown","err"]}

package/dist/middleware.mjs CHANGED Viewed

@@ -1,8 +1,8 @@
-import{NextResponse as c}from"next/server";var y=[{name:"GPTBot",company:"OpenAI"},{name:"ChatGPT-User",company:"OpenAI"},{name:"OAI-SearchBot",company:"OpenAI"},{name:"Googlebot",company:"Google"},{name:"Google-CloudVertexBot",company:"Google"},{name:"Google-Extended",company:"Google"},{name:"GoogleOther",company:"Google"},{name:"ClaudeBot",company:"Anthropic"},{name:"Claude-User",company:"Anthropic"},{name:"anthropic-ai",company:"Anthropic"},{name:"PerplexityBot",company:"Perplexity"},{name:"Perplexity-User",company:"Perplexity"},{name:"Meta-ExternalAgent",company:"Meta"},{name:"Meta-ExternalFetcher",company:"Meta"},{name:"FacebookBot",company:"Meta"},{name:"CCBot",company:"Common Crawl"},{name:"Bytespider",company:"ByteDance"},{name:"Applebot-Extended",company:"Apple"},{name:"cohere-ai",company:"Cohere"},{name:"YouBot",company:"You.com"}],B=y.map(e=>e.name);function l(e){if(!e)return;let t=e.toLowerCase();return y.find(o=>t.includes(o.name.toLowerCase()))}async function f(){try{let e=await import(process.cwd()+"/onto.config");return e.default||e}catch{return null}}function x(e){let t=[];if(t.push(`# ${e.name}`),t.push(""),t.push(`> ${e.summary}`),t.push(""),e.routes&&e.routes.length>0){t.push("## Key Routes"),t.push("");for(let o of e.routes){let a=`${e.baseUrl}${o.path}`;t.push(`- [${o.path}](${a}): ${o.description}`)}t.push("")}if(e.externalLinks&&e.externalLinks.length>0){t.push("## Resources"),t.push("");for(let o of e.externalLinks)o.description?t.push(`- [${o.title}](${o.url}): ${o.description}`):t.push(`- [${o.title}](${o.url})`);t.push("")}if(e.sections&&e.sections.length>0)for(let o of e.sections)t.push(`## ${o.heading}`),t.push(""),t.push(o.content),t.push("");return t.join(`
+import{NextResponse as p}from"next/server";var y=[{name:"GPTBot",company:"OpenAI"},{name:"ChatGPT-User",company:"OpenAI"},{name:"OAI-SearchBot",company:"OpenAI"},{name:"Googlebot",company:"Google"},{name:"Google-CloudVertexBot",company:"Google"},{name:"Google-Extended",company:"Google"},{name:"GoogleOther",company:"Google"},{name:"ClaudeBot",company:"Anthropic"},{name:"Claude-User",company:"Anthropic"},{name:"anthropic-ai",company:"Anthropic"},{name:"PerplexityBot",company:"Perplexity"},{name:"Perplexity-User",company:"Perplexity"},{name:"Meta-ExternalAgent",company:"Meta"},{name:"Meta-ExternalFetcher",company:"Meta"},{name:"FacebookBot",company:"Meta"},{name:"CCBot",company:"Common Crawl"},{name:"Bytespider",company:"ByteDance"},{name:"Applebot-Extended",company:"Apple"},{name:"cohere-ai",company:"Cohere"},{name:"YouBot",company:"You.com"}],B=y.map(t=>t.name);function m(t){if(!t)return;let e=t.toLowerCase();return y.find(o=>e.includes(o.name.toLowerCase()))}function x(t){let e=[];if(e.push(`# ${t.name}`),e.push(""),e.push(`> ${t.summary}`),e.push(""),t.routes&&t.routes.length>0){e.push("## Key Routes"),e.push("");for(let o of t.routes){let a=`${t.baseUrl}${o.path}`;e.push(`- [${o.path}](${a}): ${o.description}`)}e.push("")}if(t.externalLinks&&t.externalLinks.length>0){e.push("## Resources"),e.push("");for(let o of t.externalLinks)o.description?e.push(`- [${o.title}](${o.url}): ${o.description}`):e.push(`- [${o.title}](${o.url})`);e.push("")}if(t.sections&&t.sections.length>0)for(let o of t.sections)e.push(`## ${o.heading}`),e.push(""),e.push(o.content),e.push("");return e.join(`
 `).trim()+`
-`}async function G(e,t){let o=e.headers.get("user-agent"),a=e.nextUrl.clone(),n=l(o),O=e.headers.get("accept")||"",C=!!n,$=O.includes("text/markdown");if(C||$){if(a.pathname.startsWith("/_next"))return c.next();if(a.pathname==="/llms.txt")try{let i=t||await f();if(i){let r=x(i),p=new c(r,{headers:{"Content-Type":"text/plain; charset=utf-8","Cache-Control":"public, max-age=3600, s-maxage=3600, stale-while-revalidate=86400"}});return n&&p.headers.set("X-Onto-Bot",`${n.name} (${n.company})`),p}else{a.pathname="/llms.txt";let r=c.rewrite(a);return r.headers.set("Content-Type","text/plain; charset=utf-8"),r.headers.set("Cache-Control","public, max-age=3600, s-maxage=3600, stale-while-revalidate=86400"),n&&r.headers.set("X-Onto-Bot",`${n.name} (${n.company})`),r}}catch(i){console.error("[Onto] Failed to generate llms.txt:",i),a.pathname="/llms.txt";let r=c.rewrite(a);return r.headers.set("Content-Type","text/plain; charset=utf-8"),n&&r.headers.set("X-Onto-Bot",`${n.name} (${n.company})`),r}if(a.pathname.includes("."))return c.next();let s=a.pathname;(s==="/"||s==="")&&(s="/index"),s.endsWith("/")&&s!=="/"&&(s=s.slice(0,-1));let u={"Content-Type":"text/markdown; charset=utf-8","Cache-Control":"public, max-age=3600, s-maxage=3600, stale-while-revalidate=86400"};n&&(u["X-Onto-Bot"]=`${n.name} (${n.company})`);let m=process.env.ONTO_API_KEY,h=process.env.ONTO_DASHBOARD_URL||"https://app.buildonto.dev";if(m){fetch(`${h}/api/track`,{method:"POST",headers:{"x-onto-key":m,"Content-Type":"application/json"},body:JSON.stringify({route:a.pathname,userAgent:o,bot:n?n.name:null,company:n?n.company:null})}).catch(()=>{});try{let i=await fetch(`${h}/api/sdk/inject?route=${a.pathname}`,{headers:{"x-onto-key":m},signal:AbortSignal.timeout(1500)});if(i.ok){let{injection:r}=await i.json();if(r){let p=`${a.origin}/.onto${s}.md`,g=await fetch(p);if(g.ok){let A=`${await g.text()}
+`}async function G(t,e){let o=t.headers.get("user-agent"),a=t.nextUrl.clone(),n=m(o),g=t.headers.get("accept")||"",f=!!n,O=g.includes("text/markdown");if(f||O){if(a.pathname.startsWith("/_next"))return p.next();if(a.pathname==="/llms.txt")try{if(e){let r=x(e),i=new p(r,{headers:{"Content-Type":"text/plain; charset=utf-8","Cache-Control":"public, max-age=3600, s-maxage=3600, stale-while-revalidate=86400"}});return n&&i.headers.set("X-Onto-Bot",`${n.name} (${n.company})`),i}else{a.pathname="/llms.txt";let r=p.rewrite(a);return r.headers.set("Content-Type","text/plain; charset=utf-8"),r.headers.set("Cache-Control","public, max-age=3600, s-maxage=3600, stale-while-revalidate=86400"),n&&r.headers.set("X-Onto-Bot",`${n.name} (${n.company})`),r}}catch(r){console.error("[Onto] Failed to generate llms.txt:",r),a.pathname="/llms.txt";let i=p.rewrite(a);return i.headers.set("Content-Type","text/plain; charset=utf-8"),n&&i.headers.set("X-Onto-Bot",`${n.name} (${n.company})`),i}if(a.pathname.includes("."))return p.next();let s=a.pathname;(s==="/"||s==="")&&(s="/index"),s.endsWith("/")&&s!=="/"&&(s=s.slice(0,-1));let l={"Content-Type":"text/markdown; charset=utf-8","Cache-Control":"public, max-age=3600, s-maxage=3600, stale-while-revalidate=86400"};n&&(l["X-Onto-Bot"]=`${n.name} (${n.company})`);let c=process.env.ONTO_API_KEY,h=process.env.ONTO_DASHBOARD_URL||"https://app.buildonto.dev";if(c){fetch(`${h}/api/track`,{method:"POST",headers:{"x-onto-key":c,"Content-Type":"application/json"},body:JSON.stringify({route:a.pathname,userAgent:o,bot:n?n.name:null,company:n?n.company:null})}).catch(()=>{});try{let r=await fetch(`${h}/api/sdk/inject?route=${a.pathname}`,{headers:{"x-onto-key":c},signal:AbortSignal.timeout(1500)});if(r.ok){let{injection:i}=await r.json();if(i){let $=`${a.origin}/.onto${s}.md`,d=await fetch($);if(d.ok){let A=`${await d.text()}
 ---
-${r}`;return new c(A,{headers:{...u,"X-Onto-Injected":"true"}})}}}}catch(i){console.error("[Onto] Injection failed",i)}}a.pathname=`/.onto${s}.md`;let d=c.rewrite(a);return n&&d.headers.set("X-Onto-Bot",`${n.name} (${n.company})`),d}return c.next()}export{B as AI_BOT_USER_AGENTS,l as matchBot,G as ontoMiddleware};
+${i}`;return new p(A,{headers:{...l,"X-Onto-Injected":"true"}})}}}}catch(r){console.error("[Onto] Injection failed",r)}}a.pathname=`/.onto${s}.md`;let u=p.rewrite(a);return n&&u.headers.set("X-Onto-Bot",`${n.name} (${n.company})`),u}return p.next()}export{B as AI_BOT_USER_AGENTS,m as matchBot,G as ontoMiddleware};
 //# sourceMappingURL=middleware.mjs.map

package/dist/middleware.mjs.map CHANGED Viewed

	@@ -1 +1 @@
1	- {"version":3,"sources":["../src/middleware.ts","../src/bots.ts","../src/config.ts"],"sourcesContent":["import { NextRequest, NextResponse } from 'next/server';\r\nimport { matchBot } from './bots';\r\nimport { loadOntoConfig, generateLlmsTxt, OntoConfig } from './config';\r\n\r\nexport async function ontoMiddleware(request: NextRequest, config?: OntoConfig) {\r\n const userAgent = request.headers.get('user-agent');\r\n const url = request.nextUrl.clone();\r\n const matched = matchBot(userAgent);\r\n\r\n const accept = request.headers.get('accept') \|\| '';\r\n\r\n const isAiBot = !!matched;\r\n const isMarkdownRequested = accept.includes('text/markdown');\r\n\r\n // If traffic is identified as an AI Bot or markdown is requested\r\n if (isAiBot \|\| isMarkdownRequested) {\r\n\r\n // Ignore internal next.js requests & static assets (but not llms.txt)\r\n if (url.pathname.startsWith('/_next')) {\r\n return NextResponse.next();\r\n }\r\n\r\n // --- llms.txt Auto-Discovery ---\r\n // Serve the llms.txt manifest to AI agents\r\n if (url.pathname === '/llms.txt') {\r\n try {\r\n // Use provided config or try to load dynamic one\r\n const activeConfig = config \|\| await loadOntoConfig();\r\n\r\n if (activeConfig) {\r\n // Generate llms.txt dynamically from config\r\n const llmsTxtContent = generateLlmsTxt(activeConfig);\r\n const response = new NextResponse(llmsTxtContent, {\r\n headers: {\r\n 'Content-Type': 'text/plain; charset=utf-8',\r\n 'Cache-Control': 'public, max-age=3600, s-maxage=3600, stale-while-revalidate=86400',\r\n }\r\n });\r\n\r\n if (matched) {\r\n response.headers.set('X-Onto-Bot', `${matched.name} (${matched.company})`);\r\n }\r\n\r\n return response;\r\n } else {\r\n // Fallback: try to serve static llms.txt from public folder\r\n url.pathname = '/llms.txt';\r\n const response = NextResponse.rewrite(url);\r\n response.headers.set('Content-Type', 'text/plain; charset=utf-8');\r\n response.headers.set('Cache-Control', 'public, max-age=3600, s-maxage=3600, stale-while-revalidate=86400');\r\n if (matched) {\r\n response.headers.set('X-Onto-Bot', `${matched.name} (${matched.company})`);\r\n }\r\n return response;\r\n }\r\n } catch (error) {\r\n console.error('[Onto] Failed to generate llms.txt:', error);\r\n // Fallback to static file on error\r\n url.pathname = '/llms.txt';\r\n const response = NextResponse.rewrite(url);\r\n response.headers.set('Content-Type', 'text/plain; charset=utf-8');\r\n if (matched) {\r\n response.headers.set('X-Onto-Bot', `${matched.name} (${matched.company})`);\r\n }\r\n return response;\r\n }\r\n }\r\n\r\n // Skip other static assets\r\n if (url.pathname.includes('.')) {\r\n return NextResponse.next();\r\n }\r\n\r\n // Determine the corresponding payload path\r\n let payloadPath = url.pathname;\r\n if (payloadPath === '/' \|\| payloadPath === '') {\r\n payloadPath = '/index';\r\n }\r\n\r\n // Strip trailing slash if present\r\n if (payloadPath.endsWith('/') && payloadPath !== '/') {\r\n payloadPath = payloadPath.slice(0, -1);\r\n }\r\n\r\n // Common response headers for all bot responses\r\n const botHeaders: Record<string, string> = {\r\n 'Content-Type': 'text/markdown; charset=utf-8',\r\n 'Cache-Control': 'public, max-age=3600, s-maxage=3600, stale-while-revalidate=86400',\r\n };\r\n if (matched) {\r\n botHeaders['X-Onto-Bot'] = `${matched.name} (${matched.company})`;\r\n }\r\n\r\n // --- Onto Control Plane Integration (Premium) ---\r\n const ONTO_API_KEY = process.env.ONTO_API_KEY;\r\n const DASHBOARD_URL = process.env.ONTO_DASHBOARD_URL \|\| 'https://app.buildonto.dev';\r\n\r\n if (ONTO_API_KEY) {\r\n // 1. Fire-and-forget tracking — includes structured bot info\r\n fetch(`${DASHBOARD_URL}/api/track`, {\r\n method: 'POST',\r\n headers: {\r\n 'x-onto-key': ONTO_API_KEY,\r\n 'Content-Type': 'application/json'\r\n },\r\n body: JSON.stringify({\r\n route: url.pathname,\r\n userAgent: userAgent,\r\n bot: matched ? matched.name : null,\r\n company: matched ? matched.company : null,\r\n })\r\n }).catch(() => {});\r\n\r\n // 2. Dynamic Context Injection\r\n try {\r\n const injectRes = await fetch(`${DASHBOARD_URL}/api/sdk/inject?route=${url.pathname}`, {\r\n headers: { 'x-onto-key': ONTO_API_KEY },\r\n signal: AbortSignal.timeout(1500)\r\n });\r\n\r\n if (injectRes.ok) {\r\n const { injection } = await injectRes.json();\r\n \r\n if (injection) {\r\n const localMdUrl = `${url.origin}/.onto${payloadPath}.md`;\r\n const mdRes = await fetch(localMdUrl);\r\n \r\n if (mdRes.ok) {\r\n const baseMarkdown = await mdRes.text();\r\n const finalMarkdown = `${baseMarkdown}\\n\\n---\\n\\n${injection}`;\r\n \r\n return new NextResponse(finalMarkdown, {\r\n headers: {\r\n ...botHeaders,\r\n 'X-Onto-Injected': 'true'\r\n }\r\n });\r\n }\r\n }\r\n }\r\n } catch (err) {\r\n console.error('[Onto] Injection failed', err);\r\n }\r\n }\r\n // ------------------------------------------------\r\n\r\n url.pathname = `/.onto${payloadPath}.md`;\r\n\r\n // Rewrite implicitly serves the target URL transparently to the client.\r\n const response = NextResponse.rewrite(url);\r\n if (matched) {\r\n response.headers.set('X-Onto-Bot', `${matched.name} (${matched.company})`);\r\n }\r\n return response;\r\n }\r\n\r\n return NextResponse.next();\r\n}\r\n\r\n// Re-export the bot registry for consumers who want to extend or inspect it\r\nexport { AI_BOT_USER_AGENTS, matchBot } from './bots';\r\nexport type { AiBot } from './bots';\r\n\r\n\r\n","/*\n Comprehensive registry of AI bot user-agent strings.\n * The middleware uses this list to detect AI crawlers and serve optimized markdown.\n /\n\nexport interface AiBot {\n /* The user-agent substring to match against /\n name: string;\n /* The company operating this bot /\n company: string;\n}\n\n/\n Structured registry of all known AI bots, grouped by company.\n * Useful for analytics and the Control Plane dashboard.\n /\nexport const AI_BOTS: AiBot[] = [\n // OpenAI\n { name: 'GPTBot', company: 'OpenAI' },\n { name: 'ChatGPT-User', company: 'OpenAI' },\n { name: 'OAI-SearchBot', company: 'OpenAI' },\n\n // Google\n { name: 'Googlebot', company: 'Google' },\n { name: 'Google-CloudVertexBot', company: 'Google' },\n { name: 'Google-Extended', company: 'Google' },\n { name: 'GoogleOther', company: 'Google' },\n\n // Anthropic\n { name: 'ClaudeBot', company: 'Anthropic' },\n { name: 'Claude-User', company: 'Anthropic' },\n { name: 'anthropic-ai', company: 'Anthropic' },\n\n // Perplexity\n { name: 'PerplexityBot', company: 'Perplexity' },\n { name: 'Perplexity-User', company: 'Perplexity' },\n\n // Meta\n { name: 'Meta-ExternalAgent', company: 'Meta' },\n { name: 'Meta-ExternalFetcher', company: 'Meta' },\n { name: 'FacebookBot', company: 'Meta' },\n\n // Common Crawl (used by most smaller AI companies)\n { name: 'CCBot', company: 'Common Crawl' },\n\n // Other notable AI crawlers\n { name: 'Bytespider', company: 'ByteDance' },\n { name: 'Applebot-Extended', company: 'Apple' },\n { name: 'cohere-ai', company: 'Cohere' },\n { name: 'YouBot', company: 'You.com' },\n];\n\n/\n Flat list of user-agent substrings for fast matching in the middleware.\n /\nexport const AI_BOT_USER_AGENTS: string[] = AI_BOTS.map(bot => bot.name);\n\n/\n Given a raw user-agent string, returns the matched AiBot entry or undefined.\n * Comparison is case-insensitive to handle inconsistent agent casing.\n /\nexport function matchBot(userAgent: string \| null): AiBot \| undefined {\n if (!userAgent) return undefined;\n const lowerUA = userAgent.toLowerCase();\n return AI_BOTS.find(bot => \n lowerUA.includes(bot.name.toLowerCase())\n );\n}\n","/\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":"AAAA,OAAsB,gBAAAA,MAAoB,cCgBnC,IAAMC,EAAmB,CAE5B,CAAE,KAAM,SAAqB,QAAS,QAAS,EAC/C,CAAE,KAAM,eAAoB,QAAS,QAAS,EAC9C,CAAE,KAAM,gBAAoB,QAAS,QAAS,EAG9C,CAAE,KAAM,YAA0B,QAAS,QAAS,EACpD,CAAE,KAAM,wBAA2B,QAAS,QAAS,EACrD,CAAE,KAAM,kBAA2B,QAAS,QAAS,EACrD,CAAE,KAAM,cAA2B,QAAS,QAAS,EAGrD,CAAE,KAAM,YAAmB,QAAS,WAAY,EAChD,CAAE,KAAM,cAAkB,QAAS,WAAY,EAC/C,CAAE,KAAM,eAAkB,QAAS,WAAY,EAG/C,CAAE,KAAM,gBAAmB,QAAS,YAAa,EACjD,CAAE,KAAM,kBAAmB,QAAS,YAAa,EAGjD,CAAE,KAAM,qBAAwB,QAAS,MAAO,EAChD,CAAE,KAAM,uBAAwB,QAAS,MAAO,EAChD,CAAE,KAAM,cAAuB,QAAS,MAAO,EAG/C,CAAE,KAAM,QAAS,QAAS,cAAe,EAGzC,CAAE,KAAM,aAAqB,QAAS,WAAY,EAClD,CAAE,KAAM,oBAAqB,QAAS,OAAQ,EAC9C,CAAE,KAAM,YAAoB,QAAS,QAAS,EAC9C,CAAE,KAAM,SAAoB,QAAS,SAAU,CACnD,EAKaC,EAA+BD,EAAQ,IAAIE,GAAOA,EAAI,IAAI,EAMhE,SAASC,EAASC,EAA6C,CAClE,GAAI,CAACA,EAAW,OAChB,IAAMC,EAAUD,EAAU,YAAY,EACtC,OAAOJ,EAAQ,KAAKE,GAChBG,EAAQ,SAASH,EAAI,KAAK,YAAY,CAAC,CAC3C,CACJ,CCiBA,eAAsBI,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,CFlJA,eAAsBK,EAAeC,EAAsBC,EAAqB,CAC5E,IAAMC,EAAYF,EAAQ,QAAQ,IAAI,YAAY,EAC5CG,EAAMH,EAAQ,QAAQ,MAAM,EAC5BI,EAAUC,EAASH,CAAS,EAE5BI,EAASN,EAAQ,QAAQ,IAAI,QAAQ,GAAK,GAE1CO,EAAU,CAAC,CAACH,EACZI,EAAsBF,EAAO,SAAS,eAAe,EAG3D,GAAIC,GAAWC,EAAqB,CAGhC,GAAIL,EAAI,SAAS,WAAW,QAAQ,EAChC,OAAOM,EAAa,KAAK,EAK7B,GAAIN,EAAI,WAAa,YACjB,GAAI,CAEA,IAAMO,EAAeT,GAAU,MAAMU,EAAe,EAEpD,GAAID,EAAc,CAEd,IAAME,EAAiBC,EAAgBH,CAAY,EAC7CI,EAAW,IAAIL,EAAaG,EAAgB,CAC9C,QAAS,CACL,eAAgB,4BAChB,gBAAiB,mEACrB,CACJ,CAAC,EAED,OAAIR,GACAU,EAAS,QAAQ,IAAI,aAAc,GAAGV,EAAQ,IAAI,KAAKA,EAAQ,OAAO,GAAG,EAGtEU,CACX,KAAO,CAEHX,EAAI,SAAW,YACf,IAAMW,EAAWL,EAAa,QAAQN,CAAG,EACzC,OAAAW,EAAS,QAAQ,IAAI,eAAgB,2BAA2B,EAChEA,EAAS,QAAQ,IAAI,gBAAiB,mEAAmE,EACrGV,GACAU,EAAS,QAAQ,IAAI,aAAc,GAAGV,EAAQ,IAAI,KAAKA,EAAQ,OAAO,GAAG,EAEtEU,CACX,CACJ,OAASC,EAAO,CACZ,QAAQ,MAAM,sCAAuCA,CAAK,EAE1DZ,EAAI,SAAW,YACf,IAAMW,EAAWL,EAAa,QAAQN,CAAG,EACzC,OAAAW,EAAS,QAAQ,IAAI,eAAgB,2BAA2B,EAC5DV,GACAU,EAAS,QAAQ,IAAI,aAAc,GAAGV,EAAQ,IAAI,KAAKA,EAAQ,OAAO,GAAG,EAEtEU,CACX,CAIJ,GAAIX,EAAI,SAAS,SAAS,GAAG,EACzB,OAAOM,EAAa,KAAK,EAI7B,IAAIO,EAAcb,EAAI,UAClBa,IAAgB,KAAOA,IAAgB,MACvCA,EAAc,UAIdA,EAAY,SAAS,GAAG,GAAKA,IAAgB,MAC7CA,EAAcA,EAAY,MAAM,EAAG,EAAE,GAIzC,IAAMC,EAAqC,CACvC,eAAgB,+BAChB,gBAAiB,mEACrB,EACIb,IACAa,EAAW,YAAY,EAAI,GAAGb,EAAQ,IAAI,KAAKA,EAAQ,OAAO,KAIlE,IAAMc,EAAe,QAAQ,IAAI,aAC3BC,EAAgB,QAAQ,IAAI,oBAAsB,4BAExD,GAAID,EAAc,CAEd,MAAM,GAAGC,CAAa,aAAc,CAChC,OAAQ,OACR,QAAS,CACL,aAAcD,EACd,eAAgB,kBACpB,EACA,KAAM,KAAK,UAAU,CACjB,MAAOf,EAAI,SACX,UAAWD,EACX,IAAKE,EAAUA,EAAQ,KAAO,KAC9B,QAASA,EAAUA,EAAQ,QAAU,IACzC,CAAC,CACL,CAAC,EAAE,MAAM,IAAM,CAAC,CAAC,EAGjB,GAAI,CACA,IAAMgB,EAAY,MAAM,MAAM,GAAGD,CAAa,yBAAyBhB,EAAI,QAAQ,GAAI,CACnF,QAAS,CAAE,aAAce,CAAa,EACtC,OAAQ,YAAY,QAAQ,IAAI,CACpC,CAAC,EAED,GAAIE,EAAU,GAAI,CACd,GAAM,CAAE,UAAAC,CAAU,EAAI,MAAMD,EAAU,KAAK,EAE3C,GAAIC,EAAW,CACX,IAAMC,EAAa,GAAGnB,EAAI,MAAM,SAASa,CAAW,MAC9CO,EAAQ,MAAM,MAAMD,CAAU,EAEpC,GAAIC,EAAM,GAAI,CAEV,IAAMC,EAAgB,GADD,MAAMD,EAAM,KAAK,CACD;AAAA;AAAA;AAAA;AAAA,EAAcF,CAAS,GAE5D,OAAO,IAAIZ,EAAae,EAAe,CACnC,QAAS,CACL,GAAGP,EACH,kBAAmB,MACvB,CACJ,CAAC,CACL,CACJ,CACJ,CACJ,OAASQ,EAAK,CACV,QAAQ,MAAM,0BAA2BA,CAAG,CAChD,CACJ,CAGAtB,EAAI,SAAW,SAASa,CAAW,MAGnC,IAAMF,EAAWL,EAAa,QAAQN,CAAG,EACzC,OAAIC,GACAU,EAAS,QAAQ,IAAI,aAAc,GAAGV,EAAQ,IAAI,KAAKA,EAAQ,OAAO,GAAG,EAEtEU,CACX,CAEA,OAAOL,EAAa,KAAK,CAC7B","names":["NextResponse","AI_BOTS","AI_BOT_USER_AGENTS","bot","matchBot","userAgent","lowerUA","loadOntoConfig","config","generateLlmsTxt","lines","route","fullUrl","link","section","ontoMiddleware","request","config","userAgent","url","matched","matchBot","accept","isAiBot","isMarkdownRequested","NextResponse","activeConfig","loadOntoConfig","llmsTxtContent","generateLlmsTxt","response","error","payloadPath","botHeaders","ONTO_API_KEY","DASHBOARD_URL","injectRes","injection","localMdUrl","mdRes","finalMarkdown","err"]}
1	+ {"version":3,"sources":["../src/middleware.ts","../src/bots.ts","../src/config.ts"],"sourcesContent":["import { NextRequest, NextResponse } from 'next/server';\r\nimport { matchBot } from './bots';\r\nimport { generateLlmsTxt, OntoConfig } from './config';\r\n\r\nexport async function ontoMiddleware(request: NextRequest, config?: OntoConfig) {\r\n const userAgent = request.headers.get('user-agent');\r\n const url = request.nextUrl.clone();\r\n const matched = matchBot(userAgent);\r\n\r\n const accept = request.headers.get('accept') \|\| '';\r\n\r\n const isAiBot = !!matched;\r\n const isMarkdownRequested = accept.includes('text/markdown');\r\n\r\n // If traffic is identified as an AI Bot or markdown is requested\r\n if (isAiBot \|\| isMarkdownRequested) {\r\n\r\n // Ignore internal next.js requests & static assets (but not llms.txt)\r\n if (url.pathname.startsWith('/_next')) {\r\n return NextResponse.next();\r\n }\r\n\r\n // --- llms.txt Auto-Discovery ---\r\n // Serve the llms.txt manifest to AI agents\r\n if (url.pathname === '/llms.txt') {\r\n try {\r\n if (config) {\r\n // Generate llms.txt dynamically from config\r\n const llmsTxtContent = generateLlmsTxt(config);\r\n const response = new NextResponse(llmsTxtContent, {\r\n headers: {\r\n 'Content-Type': 'text/plain; charset=utf-8',\r\n 'Cache-Control': 'public, max-age=3600, s-maxage=3600, stale-while-revalidate=86400',\r\n }\r\n });\r\n\r\n if (matched) {\r\n response.headers.set('X-Onto-Bot', `${matched.name} (${matched.company})`);\r\n }\r\n\r\n return response;\r\n } else {\r\n // Fallback: try to serve static llms.txt from public folder\r\n url.pathname = '/llms.txt';\r\n const response = NextResponse.rewrite(url);\r\n response.headers.set('Content-Type', 'text/plain; charset=utf-8');\r\n response.headers.set('Cache-Control', 'public, max-age=3600, s-maxage=3600, stale-while-revalidate=86400');\r\n if (matched) {\r\n response.headers.set('X-Onto-Bot', `${matched.name} (${matched.company})`);\r\n }\r\n return response;\r\n }\r\n } catch (error) {\r\n console.error('[Onto] Failed to generate llms.txt:', error);\r\n // Fallback to static file on error\r\n url.pathname = '/llms.txt';\r\n const response = NextResponse.rewrite(url);\r\n response.headers.set('Content-Type', 'text/plain; charset=utf-8');\r\n if (matched) {\r\n response.headers.set('X-Onto-Bot', `${matched.name} (${matched.company})`);\r\n }\r\n return response;\r\n }\r\n }\r\n\r\n // Skip other static assets\r\n if (url.pathname.includes('.')) {\r\n return NextResponse.next();\r\n }\r\n\r\n // Determine the corresponding payload path\r\n let payloadPath = url.pathname;\r\n if (payloadPath === '/' \|\| payloadPath === '') {\r\n payloadPath = '/index';\r\n }\r\n\r\n // Strip trailing slash if present\r\n if (payloadPath.endsWith('/') && payloadPath !== '/') {\r\n payloadPath = payloadPath.slice(0, -1);\r\n }\r\n\r\n // Common response headers for all bot responses\r\n const botHeaders: Record<string, string> = {\r\n 'Content-Type': 'text/markdown; charset=utf-8',\r\n 'Cache-Control': 'public, max-age=3600, s-maxage=3600, stale-while-revalidate=86400',\r\n };\r\n if (matched) {\r\n botHeaders['X-Onto-Bot'] = `${matched.name} (${matched.company})`;\r\n }\r\n\r\n // --- Onto Control Plane Integration (Premium) ---\r\n const ONTO_API_KEY = process.env.ONTO_API_KEY;\r\n const DASHBOARD_URL = process.env.ONTO_DASHBOARD_URL \|\| 'https://app.buildonto.dev';\r\n\r\n if (ONTO_API_KEY) {\r\n // 1. Fire-and-forget tracking — includes structured bot info\r\n fetch(`${DASHBOARD_URL}/api/track`, {\r\n method: 'POST',\r\n headers: {\r\n 'x-onto-key': ONTO_API_KEY,\r\n 'Content-Type': 'application/json'\r\n },\r\n body: JSON.stringify({\r\n route: url.pathname,\r\n userAgent: userAgent,\r\n bot: matched ? matched.name : null,\r\n company: matched ? matched.company : null,\r\n })\r\n }).catch(() => {});\r\n\r\n // 2. Dynamic Context Injection\r\n try {\r\n const injectRes = await fetch(`${DASHBOARD_URL}/api/sdk/inject?route=${url.pathname}`, {\r\n headers: { 'x-onto-key': ONTO_API_KEY },\r\n signal: AbortSignal.timeout(1500)\r\n });\r\n\r\n if (injectRes.ok) {\r\n const { injection } = await injectRes.json();\r\n \r\n if (injection) {\r\n const localMdUrl = `${url.origin}/.onto${payloadPath}.md`;\r\n const mdRes = await fetch(localMdUrl);\r\n \r\n if (mdRes.ok) {\r\n const baseMarkdown = await mdRes.text();\r\n const finalMarkdown = `${baseMarkdown}\\n\\n---\\n\\n${injection}`;\r\n \r\n return new NextResponse(finalMarkdown, {\r\n headers: {\r\n ...botHeaders,\r\n 'X-Onto-Injected': 'true'\r\n }\r\n });\r\n }\r\n }\r\n }\r\n } catch (err) {\r\n console.error('[Onto] Injection failed', err);\r\n }\r\n }\r\n // ------------------------------------------------\r\n\r\n url.pathname = `/.onto${payloadPath}.md`;\r\n\r\n // Rewrite implicitly serves the target URL transparently to the client.\r\n const response = NextResponse.rewrite(url);\r\n if (matched) {\r\n response.headers.set('X-Onto-Bot', `${matched.name} (${matched.company})`);\r\n }\r\n return response;\r\n }\r\n\r\n return NextResponse.next();\r\n}\r\n\r\n// Re-export the bot registry for consumers who want to extend or inspect it\r\nexport { AI_BOT_USER_AGENTS, matchBot } from './bots';\r\nexport type { AiBot } from './bots';\r\n\r\n\r\n","/*\n Comprehensive registry of AI bot user-agent strings.\n * The middleware uses this list to detect AI crawlers and serve optimized markdown.\n /\n\nexport interface AiBot {\n /* The user-agent substring to match against /\n name: string;\n /* The company operating this bot /\n company: string;\n}\n\n/\n Structured registry of all known AI bots, grouped by company.\n * Useful for analytics and the Control Plane dashboard.\n /\nexport const AI_BOTS: AiBot[] = [\n // OpenAI\n { name: 'GPTBot', company: 'OpenAI' },\n { name: 'ChatGPT-User', company: 'OpenAI' },\n { name: 'OAI-SearchBot', company: 'OpenAI' },\n\n // Google\n { name: 'Googlebot', company: 'Google' },\n { name: 'Google-CloudVertexBot', company: 'Google' },\n { name: 'Google-Extended', company: 'Google' },\n { name: 'GoogleOther', company: 'Google' },\n\n // Anthropic\n { name: 'ClaudeBot', company: 'Anthropic' },\n { name: 'Claude-User', company: 'Anthropic' },\n { name: 'anthropic-ai', company: 'Anthropic' },\n\n // Perplexity\n { name: 'PerplexityBot', company: 'Perplexity' },\n { name: 'Perplexity-User', company: 'Perplexity' },\n\n // Meta\n { name: 'Meta-ExternalAgent', company: 'Meta' },\n { name: 'Meta-ExternalFetcher', company: 'Meta' },\n { name: 'FacebookBot', company: 'Meta' },\n\n // Common Crawl (used by most smaller AI companies)\n { name: 'CCBot', company: 'Common Crawl' },\n\n // Other notable AI crawlers\n { name: 'Bytespider', company: 'ByteDance' },\n { name: 'Applebot-Extended', company: 'Apple' },\n { name: 'cohere-ai', company: 'Cohere' },\n { name: 'YouBot', company: 'You.com' },\n];\n\n/\n Flat list of user-agent substrings for fast matching in the middleware.\n /\nexport const AI_BOT_USER_AGENTS: string[] = AI_BOTS.map(bot => bot.name);\n\n/\n Given a raw user-agent string, returns the matched AiBot entry or undefined.\n * Comparison is case-insensitive to handle inconsistent agent casing.\n /\nexport function matchBot(userAgent: string \| null): AiBot \| undefined {\n if (!userAgent) return undefined;\n const lowerUA = userAgent.toLowerCase();\n return AI_BOTS.find(bot => \n lowerUA.includes(bot.name.toLowerCase())\n );\n}\n","/\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":"AAAA,OAAsB,gBAAAA,MAAoB,cCgBnC,IAAMC,EAAmB,CAE5B,CAAE,KAAM,SAAqB,QAAS,QAAS,EAC/C,CAAE,KAAM,eAAoB,QAAS,QAAS,EAC9C,CAAE,KAAM,gBAAoB,QAAS,QAAS,EAG9C,CAAE,KAAM,YAA0B,QAAS,QAAS,EACpD,CAAE,KAAM,wBAA2B,QAAS,QAAS,EACrD,CAAE,KAAM,kBAA2B,QAAS,QAAS,EACrD,CAAE,KAAM,cAA2B,QAAS,QAAS,EAGrD,CAAE,KAAM,YAAmB,QAAS,WAAY,EAChD,CAAE,KAAM,cAAkB,QAAS,WAAY,EAC/C,CAAE,KAAM,eAAkB,QAAS,WAAY,EAG/C,CAAE,KAAM,gBAAmB,QAAS,YAAa,EACjD,CAAE,KAAM,kBAAmB,QAAS,YAAa,EAGjD,CAAE,KAAM,qBAAwB,QAAS,MAAO,EAChD,CAAE,KAAM,uBAAwB,QAAS,MAAO,EAChD,CAAE,KAAM,cAAuB,QAAS,MAAO,EAG/C,CAAE,KAAM,QAAS,QAAS,cAAe,EAGzC,CAAE,KAAM,aAAqB,QAAS,WAAY,EAClD,CAAE,KAAM,oBAAqB,QAAS,OAAQ,EAC9C,CAAE,KAAM,YAAoB,QAAS,QAAS,EAC9C,CAAE,KAAM,SAAoB,QAAS,SAAU,CACnD,EAKaC,EAA+BD,EAAQ,IAAIE,GAAOA,EAAI,IAAI,EAMhE,SAASC,EAASC,EAA6C,CAClE,GAAI,CAACA,EAAW,OAChB,IAAMC,EAAUD,EAAU,YAAY,EACtC,OAAOJ,EAAQ,KAAKE,GAChBG,EAAQ,SAASH,EAAI,KAAK,YAAY,CAAC,CAC3C,CACJ,CCoBO,SAASI,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,CFlIA,eAAsBK,EAAeC,EAAsBC,EAAqB,CAC5E,IAAMC,EAAYF,EAAQ,QAAQ,IAAI,YAAY,EAC5CG,EAAMH,EAAQ,QAAQ,MAAM,EAC5BI,EAAUC,EAASH,CAAS,EAE5BI,EAASN,EAAQ,QAAQ,IAAI,QAAQ,GAAK,GAE1CO,EAAU,CAAC,CAACH,EACZI,EAAsBF,EAAO,SAAS,eAAe,EAG3D,GAAIC,GAAWC,EAAqB,CAGhC,GAAIL,EAAI,SAAS,WAAW,QAAQ,EAChC,OAAOM,EAAa,KAAK,EAK7B,GAAIN,EAAI,WAAa,YACjB,GAAI,CACA,GAAIF,EAAQ,CAER,IAAMS,EAAiBC,EAAgBV,CAAM,EACvCW,EAAW,IAAIH,EAAaC,EAAgB,CAC9C,QAAS,CACL,eAAgB,4BAChB,gBAAiB,mEACrB,CACJ,CAAC,EAED,OAAIN,GACAQ,EAAS,QAAQ,IAAI,aAAc,GAAGR,EAAQ,IAAI,KAAKA,EAAQ,OAAO,GAAG,EAGtEQ,CACX,KAAO,CAEHT,EAAI,SAAW,YACf,IAAMS,EAAWH,EAAa,QAAQN,CAAG,EACzC,OAAAS,EAAS,QAAQ,IAAI,eAAgB,2BAA2B,EAChEA,EAAS,QAAQ,IAAI,gBAAiB,mEAAmE,EACrGR,GACAQ,EAAS,QAAQ,IAAI,aAAc,GAAGR,EAAQ,IAAI,KAAKA,EAAQ,OAAO,GAAG,EAEtEQ,CACX,CACJ,OAASC,EAAO,CACZ,QAAQ,MAAM,sCAAuCA,CAAK,EAE1DV,EAAI,SAAW,YACf,IAAMS,EAAWH,EAAa,QAAQN,CAAG,EACzC,OAAAS,EAAS,QAAQ,IAAI,eAAgB,2BAA2B,EAC5DR,GACAQ,EAAS,QAAQ,IAAI,aAAc,GAAGR,EAAQ,IAAI,KAAKA,EAAQ,OAAO,GAAG,EAEtEQ,CACX,CAIJ,GAAIT,EAAI,SAAS,SAAS,GAAG,EACzB,OAAOM,EAAa,KAAK,EAI7B,IAAIK,EAAcX,EAAI,UAClBW,IAAgB,KAAOA,IAAgB,MACvCA,EAAc,UAIdA,EAAY,SAAS,GAAG,GAAKA,IAAgB,MAC7CA,EAAcA,EAAY,MAAM,EAAG,EAAE,GAIzC,IAAMC,EAAqC,CACvC,eAAgB,+BAChB,gBAAiB,mEACrB,EACIX,IACAW,EAAW,YAAY,EAAI,GAAGX,EAAQ,IAAI,KAAKA,EAAQ,OAAO,KAIlE,IAAMY,EAAe,QAAQ,IAAI,aAC3BC,EAAgB,QAAQ,IAAI,oBAAsB,4BAExD,GAAID,EAAc,CAEd,MAAM,GAAGC,CAAa,aAAc,CAChC,OAAQ,OACR,QAAS,CACL,aAAcD,EACd,eAAgB,kBACpB,EACA,KAAM,KAAK,UAAU,CACjB,MAAOb,EAAI,SACX,UAAWD,EACX,IAAKE,EAAUA,EAAQ,KAAO,KAC9B,QAASA,EAAUA,EAAQ,QAAU,IACzC,CAAC,CACL,CAAC,EAAE,MAAM,IAAM,CAAC,CAAC,EAGjB,GAAI,CACA,IAAMc,EAAY,MAAM,MAAM,GAAGD,CAAa,yBAAyBd,EAAI,QAAQ,GAAI,CACnF,QAAS,CAAE,aAAca,CAAa,EACtC,OAAQ,YAAY,QAAQ,IAAI,CACpC,CAAC,EAED,GAAIE,EAAU,GAAI,CACd,GAAM,CAAE,UAAAC,CAAU,EAAI,MAAMD,EAAU,KAAK,EAE3C,GAAIC,EAAW,CACX,IAAMC,EAAa,GAAGjB,EAAI,MAAM,SAASW,CAAW,MAC9CO,EAAQ,MAAM,MAAMD,CAAU,EAEpC,GAAIC,EAAM,GAAI,CAEV,IAAMC,EAAgB,GADD,MAAMD,EAAM,KAAK,CACD;AAAA;AAAA;AAAA;AAAA,EAAcF,CAAS,GAE5D,OAAO,IAAIV,EAAaa,EAAe,CACnC,QAAS,CACL,GAAGP,EACH,kBAAmB,MACvB,CACJ,CAAC,CACL,CACJ,CACJ,CACJ,OAASQ,EAAK,CACV,QAAQ,MAAM,0BAA2BA,CAAG,CAChD,CACJ,CAGApB,EAAI,SAAW,SAASW,CAAW,MAGnC,IAAMF,EAAWH,EAAa,QAAQN,CAAG,EACzC,OAAIC,GACAQ,EAAS,QAAQ,IAAI,aAAc,GAAGR,EAAQ,IAAI,KAAKA,EAAQ,OAAO,GAAG,EAEtEQ,CACX,CAEA,OAAOH,EAAa,KAAK,CAC7B","names":["NextResponse","AI_BOTS","AI_BOT_USER_AGENTS","bot","matchBot","userAgent","lowerUA","generateLlmsTxt","config","lines","route","fullUrl","link","section","ontoMiddleware","request","config","userAgent","url","matched","matchBot","accept","isAiBot","isMarkdownRequested","NextResponse","llmsTxtContent","generateLlmsTxt","response","error","payloadPath","botHeaders","ONTO_API_KEY","DASHBOARD_URL","injectRes","injection","localMdUrl","mdRes","finalMarkdown","err"]}

package/package.json CHANGED Viewed

@@ -1,6 +1,6 @@
 {
     "name": "@ontosdk/next",
-    "version": "1.3.1",
+    "version": "1.3.2",
     "description": "Extracts semantic Markdown from React/Next.js pages for AI Agents",
     "main": "dist/index.js",
     "module": "dist/index.mjs",

package/src/cli.ts CHANGED Viewed

@@ -4,6 +4,24 @@ import fs from 'fs';
 import path from 'path';
 import pc from 'picocolors';
 import { extractContent } from './extractor';
+import { generateLlmsTxt, OntoConfig } from './config';
+async function loadOntoConfig(): Promise<OntoConfig | null> {
+    try {
+        const configPath = path.join(process.cwd(), 'onto.config');
+        // Node.js dynamic import
+        const config = await import('file://' + configPath.replace(/\\/g, '/') + '.ts');
+        return config.default || config;
+    } catch (error) {
+        try {
+            const configPath = path.join(process.cwd(), 'onto.config');
+            const config = await import('file://' + configPath.replace(/\\/g, '/') + '.js');
+            return config.default || config;
+        } catch (e) {
+            return null;
+        }
+    }
+}
 // Simple helper to load .env.local from the current working directory
 function loadEnv() {
@@ -139,6 +157,22 @@ async function main() {
         }
     }
+    // --- Generate llms.txt manifest ---
+    const config = await loadOntoConfig();
+    if (config) {
+        const llmsTxtContent = generateLlmsTxt(config);
+        const llmsTxtPath = path.join(cwd, 'public/llms.txt');
+        // Ensure public dir exists
+        const publicDir = path.join(cwd, 'public');
+        if (!fs.existsSync(publicDir)) {
+            fs.mkdirSync(publicDir, { recursive: true });
+        }
+        fs.writeFileSync(llmsTxtPath, llmsTxtContent, 'utf8');
+        console.log(pc.green('✓ Generated') + pc.dim(' /llms.txt'));
+    }
     console.log(pc.dim(`Edge payloads are ready at /public/.onto/*\n`));
 }

package/src/config.ts CHANGED Viewed

@@ -78,22 +78,6 @@ export interface OntoConfig {
   };
 }
-/**
- * Load the onto.config.ts file from the user's project
- * This is used by the middleware to dynamically generate llms.txt
- */
-export async function loadOntoConfig(): Promise<OntoConfig | null> {
-  try {
-    // Try to dynamically import the config file from the user's project root
-    // This runs in the middleware context, so we look in the project root
-    const config = await import(process.cwd() + '/onto.config');
-    return config.default || config;
-  } catch (error) {
-    // Config file doesn't exist or failed to load
-    return null;
-  }
-}
 /**
  * Generate llms.txt content from OntoConfig
  * Follows the llms.txt specification:

package/src/middleware.ts CHANGED Viewed

@@ -1,6 +1,6 @@
 import { NextRequest, NextResponse } from 'next/server';
 import { matchBot } from './bots';
-import { loadOntoConfig, generateLlmsTxt, OntoConfig } from './config';
+import { generateLlmsTxt, OntoConfig } from './config';
 export async function ontoMiddleware(request: NextRequest, config?: OntoConfig) {
     const userAgent = request.headers.get('user-agent');
@@ -24,12 +24,9 @@ export async function ontoMiddleware(request: NextRequest, config?: OntoConfig)
         // Serve the llms.txt manifest to AI agents
         if (url.pathname === '/llms.txt') {
             try {
-                // Use provided config or try to load dynamic one
-                const activeConfig = config || await loadOntoConfig();
-                if (activeConfig) {
+                if (config) {
                     // Generate llms.txt dynamically from config
-                    const llmsTxtContent = generateLlmsTxt(activeConfig);
+                    const llmsTxtContent = generateLlmsTxt(config);
                     const response = new NextResponse(llmsTxtContent, {
                         headers: {
                             'Content-Type': 'text/plain; charset=utf-8',

package/dist/OntoHead.d.mts DELETED Viewed

@@ -1,27 +0,0 @@
-import * as react_jsx_runtime from 'react/jsx-runtime';
-/**
- * OntoHead — Auto-Discovery component for AI agents.
- *
- * Injects `<link rel="alternate">` tags into the page `<head>` so AI crawlers
- * can discover the optimized markdown endpoint for the current route.
- *
- * Usage in a Next.js App Router layout:
- * ```tsx
- * import { OntoHead } from '@ontosdk/next/components';
- *
- * export default function RootLayout({ children }) {
- *   return (
- *     <html>
- *       <head>
- *         <OntoHead />
- *       </head>
- *       <body>{children}</body>
- *     </html>
- *   );
- * }
- * ```
- */
-declare function OntoHead(): react_jsx_runtime.JSX.Element;
-export { OntoHead };

package/dist/OntoHead.d.ts DELETED Viewed

@@ -1,27 +0,0 @@
-import * as react_jsx_runtime from 'react/jsx-runtime';
-/**
- * OntoHead — Auto-Discovery component for AI agents.
- *
- * Injects `<link rel="alternate">` tags into the page `<head>` so AI crawlers
- * can discover the optimized markdown endpoint for the current route.
- *
- * Usage in a Next.js App Router layout:
- * ```tsx
- * import { OntoHead } from '@ontosdk/next/components';
- *
- * export default function RootLayout({ children }) {
- *   return (
- *     <html>
- *       <head>
- *         <OntoHead />
- *       </head>
- *       <body>{children}</body>
- *     </html>
- *   );
- * }
- * ```
- */
-declare function OntoHead(): react_jsx_runtime.JSX.Element;
-export { OntoHead };

package/dist/OntoProvider.d.mts DELETED Viewed

@@ -1,52 +0,0 @@
-import * as react_jsx_runtime from 'react/jsx-runtime';
-import { ReactNode } from 'react';
-import { OntoConfig } from './config.mjs';
-interface OntoProviderProps {
-    /**
-     * The base URL of your site (e.g., 'https://example.com')
-     * Used to construct the full href for the AI discovery link tag.
-     */
-    baseUrl: string;
-    /**
-     * Child components to render
-     */
-    children: ReactNode;
-    /**
-     * Optional: Onto configuration for automatic JSON-LD schema injection
-     * If provided, the provider will automatically inject JSON-LD schemas
-     * based on the page type configuration
-     */
-    config?: OntoConfig;
-}
-/**
- * OntoProvider — Automatic AI Discovery Provider
- *
- * Wraps your application and automatically injects:
- * 1. `<link rel="alternate">` tags for AI discovery
- * 2. JSON-LD structured data schemas based on page type
- *
- * With config, automatically generates JSON-LD schemas:
- * - 'scoring' pages get Methodology schema with AIO weights (40/35/25)
- * - 'about' pages get Organization/AboutPage schema
- *
- * Usage in a Next.js App Router layout:
- * ```tsx
- * import { OntoProvider } from '@ontosdk/next/provider';
- * import config from '../onto.config';
- *
- * export default function RootLayout({ children }) {
- *   return (
- *     <OntoProvider baseUrl="https://example.com" config={config}>
- *       <html>
- *         <head />
- *         <body>{children}</body>
- *       </html>
- *     </OntoProvider>
- *   );
- * }
- * ```
- */
-declare function OntoProvider({ baseUrl, children, config }: OntoProviderProps): react_jsx_runtime.JSX.Element;
-export { OntoProvider, type OntoProviderProps };