@ontosdk/next 1.4.2 → 1.4.3
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.
- package/dist/middleware.js +3 -3
- package/dist/middleware.js.map +1 -1
- package/dist/middleware.mjs +3 -3
- package/dist/middleware.mjs.map +1 -1
- package/package.json +1 -1
- package/src/middleware.ts +15 -2
package/dist/middleware.js
CHANGED
|
@@ -1,8 +1,8 @@
|
|
|
1
|
-
"use strict";var
|
|
1
|
+
"use strict";var d=Object.defineProperty;var T=Object.getOwnPropertyDescriptor;var b=Object.getOwnPropertyNames;var k=Object.prototype.hasOwnProperty;var R=(t,e)=>{for(var o in e)d(t,o,{get:e[o],enumerable:!0})},_=(t,e,o,a)=>{if(e&&typeof e=="object"||typeof e=="function")for(let n of b(e))!k.call(t,n)&&n!==o&&d(t,n,{get:()=>e[n],enumerable:!(a=T(e,n))||a.enumerable});return t};var P=t=>_(d({},"__esModule",{value:!0}),t);var G={};R(G,{AI_BOT_USER_AGENTS:()=>f,matchBot:()=>m,ontoMiddleware:()=>U});module.exports=P(G);var p=require("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"}],f=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 O(t){let e=[];if(e.push(`# ${t.name}`),e.push(""),e.push(`> ${t.summary}`),e.push(""),t.routes&&t.routes.length>0){e.push("## Key Routes"),e.push("");for(let o of t.routes){let a=`${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(`## ${o.heading}`),e.push(""),e.push(o.content),e.push("");return e.join(`
|
|
2
2
|
`).trim()+`
|
|
3
|
-
`}async function U(t,e){let o=t.headers.get("user-agent"),a=t.nextUrl.clone(),n=
|
|
3
|
+
`}async function U(t,e){let o=t.headers.get("user-agent"),a=t.nextUrl.clone(),n=m(o),$=t.headers.get("accept")||"",l=t.nextUrl.searchParams.has("onto"),C=!!n,A=$.includes("text/markdown")||l;if(C||A){if(a.pathname.startsWith("/_next"))return p.NextResponse.next();if(a.pathname==="/llms.txt")try{if(e){let r=O(e),i=new p.NextResponse(r,{headers:{"Content-Type":"text/plain; charset=utf-8","Cache-Control":"public, max-age=3600, s-maxage=3600, stale-while-revalidate=86400"}});return n&&i.headers.set("X-Onto-Bot",`${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} (${n.company})`),l&&(h["X-Onto-Debug"]="true");let u=process.env.ONTO_API_KEY,g=process.env.ONTO_DASHBOARD_URL||"https://app.buildonto.dev";if(u){fetch(`${g}/api/track`,{method:"POST",headers:{"x-onto-key":u,"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(`${g}/api/sdk/inject?route=${a.pathname}`,{headers:{"x-onto-key":u},signal:AbortSignal.timeout(1500)});if(r.ok){let{injection:i}=await r.json();if(i){let B=`${a.origin}/.onto${s}.md`,x=await fetch(B);if(x.ok){let w=`${await x.text()}
|
|
4
4
|
|
|
5
5
|
---
|
|
6
6
|
|
|
7
|
-
${i}`;return new p.NextResponse(
|
|
7
|
+
${i}`;return new p.NextResponse(w,{headers:{...h,"X-Onto-Injected":"true"}})}}}}catch(r){console.error("[Onto] Injection failed",r)}}a.pathname=`/.onto${s}.md`;let c=p.NextResponse.rewrite(a);return c.headers.set("Content-Type","text/markdown; charset=utf-8"),c.headers.set("Cache-Control","public, max-age=3600, s-maxage=3600, stale-while-revalidate=86400"),n&&c.headers.set("X-Onto-Bot",`${n.name} (${n.company})`),l&&c.headers.set("X-Onto-Debug","true"),c}return p.NextResponse.next()}0&&(module.exports={AI_BOT_USER_AGENTS,matchBot,ontoMiddleware});
|
|
8
8
|
//# sourceMappingURL=middleware.js.map
|
package/dist/middleware.js.map
CHANGED
|
@@ -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 { generateLlmsTxt, OntoConfig } from './config';\r\n\r\nexport async function ontoMiddleware(request: any, config?: OntoConfig) {\r\n const userAgent = request.headers.get('user-agent');\r\n const url = request.nextUrl.clone();\r\n const matched = matchBot(userAgent);\r\n\r\n const accept = request.headers.get('accept') || '';\r\n\r\n const isAiBot = !!matched;\r\n const isMarkdownRequested = accept.includes('text/markdown');\r\n\r\n // If traffic is identified as an AI Bot or markdown is requested\r\n if (isAiBot || isMarkdownRequested) {\r\n\r\n // Ignore internal next.js requests & static assets (but not llms.txt)\r\n if (url.pathname.startsWith('/_next')) {\r\n return NextResponse.next();\r\n }\r\n\r\n // --- llms.txt Auto-Discovery ---\r\n // Serve the llms.txt manifest to AI agents\r\n if (url.pathname === '/llms.txt') {\r\n try {\r\n if (config) {\r\n // Generate llms.txt dynamically from config\r\n const llmsTxtContent = generateLlmsTxt(config);\r\n const response = new NextResponse(llmsTxtContent, {\r\n headers: {\r\n 'Content-Type': 'text/plain; charset=utf-8',\r\n 'Cache-Control': 'public, max-age=3600, s-maxage=3600, stale-while-revalidate=86400',\r\n }\r\n });\r\n\r\n if (matched) {\r\n response.headers.set('X-Onto-Bot', `${matched.name} (${matched.company})`);\r\n }\r\n\r\n return response;\r\n } else {\r\n // Fallback: try to serve static llms.txt from public folder\r\n url.pathname = '/llms.txt';\r\n const response = NextResponse.rewrite(url);\r\n response.headers.set('Content-Type', 'text/plain; charset=utf-8');\r\n response.headers.set('Cache-Control', 'public, max-age=3600, s-maxage=3600, stale-while-revalidate=86400');\r\n if (matched) {\r\n response.headers.set('X-Onto-Bot', `${matched.name} (${matched.company})`);\r\n }\r\n return response;\r\n }\r\n } catch (error) {\r\n console.error('[Onto] Failed to generate llms.txt:', error);\r\n // Fallback to static file on error\r\n url.pathname = '/llms.txt';\r\n const response = NextResponse.rewrite(url);\r\n response.headers.set('Content-Type', 'text/plain; charset=utf-8');\r\n if (matched) {\r\n response.headers.set('X-Onto-Bot', `${matched.name} (${matched.company})`);\r\n }\r\n return response;\r\n }\r\n }\r\n\r\n // Skip other static assets\r\n if (url.pathname.includes('.')) {\r\n return NextResponse.next();\r\n }\r\n\r\n // Determine the corresponding payload path\r\n let payloadPath = url.pathname;\r\n if (payloadPath === '/' || payloadPath === '') {\r\n payloadPath = '/index';\r\n }\r\n\r\n // Strip trailing slash if present\r\n if (payloadPath.endsWith('/') && payloadPath !== '/') {\r\n payloadPath = payloadPath.slice(0, -1);\r\n }\r\n\r\n // Common response headers for all bot responses\r\n const botHeaders: Record<string, string> = {\r\n 'Content-Type': 'text/markdown; charset=utf-8',\r\n 'Cache-Control': 'public, max-age=3600, s-maxage=3600, stale-while-revalidate=86400',\r\n };\r\n if (matched) {\r\n botHeaders['X-Onto-Bot'] = `${matched.name} (${matched.company})`;\r\n }\r\n\r\n // --- Onto Control Plane Integration (Premium) ---\r\n const ONTO_API_KEY = process.env.ONTO_API_KEY;\r\n const DASHBOARD_URL = process.env.ONTO_DASHBOARD_URL || 'https://app.buildonto.dev';\r\n\r\n if (ONTO_API_KEY) {\r\n // 1. Fire-and-forget tracking — includes structured bot info\r\n fetch(`${DASHBOARD_URL}/api/track`, {\r\n method: 'POST',\r\n headers: {\r\n 'x-onto-key': ONTO_API_KEY,\r\n 'Content-Type': 'application/json'\r\n },\r\n body: JSON.stringify({\r\n route: url.pathname,\r\n userAgent: userAgent,\r\n bot: matched ? matched.name : null,\r\n company: matched ? matched.company : null,\r\n })\r\n }).catch(() => {});\r\n\r\n // 2. Dynamic Context Injection\r\n try {\r\n const injectRes = await fetch(`${DASHBOARD_URL}/api/sdk/inject?route=${url.pathname}`, {\r\n headers: { 'x-onto-key': ONTO_API_KEY },\r\n signal: AbortSignal.timeout(1500)\r\n });\r\n\r\n if (injectRes.ok) {\r\n const { injection } = await injectRes.json();\r\n \r\n if (injection) {\r\n const localMdUrl = `${url.origin}/.onto${payloadPath}.md`;\r\n const mdRes = await fetch(localMdUrl);\r\n \r\n if (mdRes.ok) {\r\n const baseMarkdown = await mdRes.text();\r\n const finalMarkdown = `${baseMarkdown}\\n\\n---\\n\\n${injection}`;\r\n \r\n return new NextResponse(finalMarkdown, {\r\n headers: {\r\n ...botHeaders,\r\n 'X-Onto-Injected': 'true'\r\n }\r\n });\r\n }\r\n }\r\n }\r\n } catch (err) {\r\n console.error('[Onto] Injection failed', err);\r\n }\r\n }\r\n // ------------------------------------------------\r\n\r\n url.pathname = `/.onto${payloadPath}.md`;\r\n\r\n // Rewrite implicitly serves the target URL transparently to the client.\r\n const response = NextResponse.rewrite(url);\r\n if (matched) {\r\n response.headers.set('X-Onto-Bot', `${matched.name} (${matched.company})`);\r\n }\r\n return response;\r\n }\r\n\r\n return NextResponse.next();\r\n}\r\n\r\n// Re-export the bot registry for consumers who want to extend or inspect it\r\nexport { AI_BOT_USER_AGENTS, matchBot } from './bots';\r\nexport type { AiBot } from './bots';\r\n\r\n\r\n","/**\n * Comprehensive registry of AI bot user-agent strings.\n * The middleware uses this list to detect AI crawlers and serve optimized markdown.\n */\n\nexport interface AiBot {\n /** The user-agent substring to match against */\n name: string;\n /** The company operating this bot */\n company: string;\n}\n\n/**\n * Structured registry of all known AI bots, grouped by company.\n * Useful for analytics and the Control Plane dashboard.\n */\nexport const AI_BOTS: AiBot[] = [\n // OpenAI\n { name: 'GPTBot', company: 'OpenAI' },\n { name: 'ChatGPT-User', company: 'OpenAI' },\n { name: 'OAI-SearchBot', company: 'OpenAI' },\n\n // Google\n { name: 'Googlebot', company: 'Google' },\n { name: 'Google-CloudVertexBot', company: 'Google' },\n { name: 'Google-Extended', company: 'Google' },\n { name: 'GoogleOther', company: 'Google' },\n\n // Anthropic\n { name: 'ClaudeBot', company: 'Anthropic' },\n { name: 'Claude-User', company: 'Anthropic' },\n { name: 'anthropic-ai', company: 'Anthropic' },\n\n // Perplexity\n { name: 'PerplexityBot', company: 'Perplexity' },\n { name: 'Perplexity-User', company: 'Perplexity' },\n\n // Meta\n { name: 'Meta-ExternalAgent', company: 'Meta' },\n { name: 'Meta-ExternalFetcher', company: 'Meta' },\n { name: 'FacebookBot', company: 'Meta' },\n\n // Common Crawl (used by most smaller AI companies)\n { name: 'CCBot', company: 'Common Crawl' },\n\n // Other notable AI crawlers\n { name: 'Bytespider', company: 'ByteDance' },\n { name: 'Applebot-Extended', company: 'Apple' },\n { name: 'cohere-ai', company: 'Cohere' },\n { name: 'YouBot', company: 'You.com' },\n];\n\n/**\n * Flat list of user-agent substrings for fast matching in the middleware.\n */\nexport const AI_BOT_USER_AGENTS: string[] = AI_BOTS.map(bot => bot.name);\n\n/**\n * Given a raw user-agent string, returns the matched AiBot entry or undefined.\n * Comparison is case-insensitive to handle inconsistent agent casing.\n */\nexport function matchBot(userAgent: string | null): AiBot | undefined {\n if (!userAgent) return undefined;\n const lowerUA = userAgent.toLowerCase();\n return AI_BOTS.find(bot => \n lowerUA.includes(bot.name.toLowerCase())\n );\n}\n","/**\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,EAAcC,EAAqB,CACpE,IAAMC,EAAYF,EAAQ,QAAQ,IAAI,YAAY,EAC5CG,EAAMH,EAAQ,QAAQ,MAAM,EAC5BI,EAAUC,EAASH,CAAS,EAE5BI,EAASN,EAAQ,QAAQ,IAAI,QAAQ,GAAK,GAE1CO,EAAU,CAAC,CAACH,EACZI,EAAsBF,EAAO,SAAS,eAAe,EAG3D,GAAIC,GAAWC,EAAqB,CAGhC,GAAIL,EAAI,SAAS,WAAW,QAAQ,EAChC,OAAO,eAAa,KAAK,EAK7B,GAAIA,EAAI,WAAa,YACjB,GAAI,CACA,GAAIF,EAAQ,CAER,IAAMQ,EAAiBC,EAAgBT,CAAM,EACvCU,EAAW,IAAI,eAAaF,EAAgB,CAC9C,QAAS,CACL,eAAgB,4BAChB,gBAAiB,mEACrB,CACJ,CAAC,EAED,OAAIL,GACAO,EAAS,QAAQ,IAAI,aAAc,GAAGP,EAAQ,IAAI,KAAKA,EAAQ,OAAO,GAAG,EAGtEO,CACX,KAAO,CAEHR,EAAI,SAAW,YACf,IAAMQ,EAAW,eAAa,QAAQR,CAAG,EACzC,OAAAQ,EAAS,QAAQ,IAAI,eAAgB,2BAA2B,EAChEA,EAAS,QAAQ,IAAI,gBAAiB,mEAAmE,EACrGP,GACAO,EAAS,QAAQ,IAAI,aAAc,GAAGP,EAAQ,IAAI,KAAKA,EAAQ,OAAO,GAAG,EAEtEO,CACX,CACJ,OAASC,EAAO,CACZ,QAAQ,MAAM,sCAAuCA,CAAK,EAE1DT,EAAI,SAAW,YACf,IAAMQ,EAAW,eAAa,QAAQR,CAAG,EACzC,OAAAQ,EAAS,QAAQ,IAAI,eAAgB,2BAA2B,EAC5DP,GACAO,EAAS,QAAQ,IAAI,aAAc,GAAGP,EAAQ,IAAI,KAAKA,EAAQ,OAAO,GAAG,EAEtEO,CACX,CAIJ,GAAIR,EAAI,SAAS,SAAS,GAAG,EACzB,OAAO,eAAa,KAAK,EAI7B,IAAIU,EAAcV,EAAI,UAClBU,IAAgB,KAAOA,IAAgB,MACvCA,EAAc,UAIdA,EAAY,SAAS,GAAG,GAAKA,IAAgB,MAC7CA,EAAcA,EAAY,MAAM,EAAG,EAAE,GAIzC,IAAMC,EAAqC,CACvC,eAAgB,+BAChB,gBAAiB,mEACrB,EACIV,IACAU,EAAW,YAAY,EAAI,GAAGV,EAAQ,IAAI,KAAKA,EAAQ,OAAO,KAIlE,IAAMW,EAAe,QAAQ,IAAI,aAC3BC,EAAgB,QAAQ,IAAI,oBAAsB,4BAExD,GAAID,EAAc,CAEd,MAAM,GAAGC,CAAa,aAAc,CAChC,OAAQ,OACR,QAAS,CACL,aAAcD,EACd,eAAgB,kBACpB,EACA,KAAM,KAAK,UAAU,CACjB,MAAOZ,EAAI,SACX,UAAWD,EACX,IAAKE,EAAUA,EAAQ,KAAO,KAC9B,QAASA,EAAUA,EAAQ,QAAU,IACzC,CAAC,CACL,CAAC,EAAE,MAAM,IAAM,CAAC,CAAC,EAGjB,GAAI,CACA,IAAMa,EAAY,MAAM,MAAM,GAAGD,CAAa,yBAAyBb,EAAI,QAAQ,GAAI,CACnF,QAAS,CAAE,aAAcY,CAAa,EACtC,OAAQ,YAAY,QAAQ,IAAI,CACpC,CAAC,EAED,GAAIE,EAAU,GAAI,CACd,GAAM,CAAE,UAAAC,CAAU,EAAI,MAAMD,EAAU,KAAK,EAE3C,GAAIC,EAAW,CACX,IAAMC,EAAa,GAAGhB,EAAI,MAAM,SAASU,CAAW,MAC9CO,EAAQ,MAAM,MAAMD,CAAU,EAEpC,GAAIC,EAAM,GAAI,CAEV,IAAMC,EAAgB,GADD,MAAMD,EAAM,KAAK,CACD;AAAA;AAAA;AAAA;AAAA,EAAcF,CAAS,GAE5D,OAAO,IAAI,eAAaG,EAAe,CACnC,QAAS,CACL,GAAGP,EACH,kBAAmB,MACvB,CACJ,CAAC,CACL,CACJ,CACJ,CACJ,OAASQ,EAAK,CACV,QAAQ,MAAM,0BAA2BA,CAAG,CAChD,CACJ,CAGAnB,EAAI,SAAW,SAASU,CAAW,MAGnC,IAAMF,EAAW,eAAa,QAAQR,CAAG,EACzC,OAAIC,GACAO,EAAS,QAAQ,IAAI,aAAc,GAAGP,EAAQ,IAAI,KAAKA,EAAQ,OAAO,GAAG,EAEtEO,CACX,CAEA,OAAO,eAAa,KAAK,CAC7B","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"]}
|
|
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: any, config?: OntoConfig) {\r\n const userAgent = request.headers.get('user-agent');\r\n const url = request.nextUrl.clone();\r\n const matched = matchBot(userAgent);\r\n\r\n const accept = request.headers.get('accept') || '';\r\n const hasDebugParam = request.nextUrl.searchParams.has('onto');\r\n\r\n const isAiBot = !!matched;\r\n const isMarkdownRequested = accept.includes('text/markdown') || hasDebugParam;\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} (${matched.company})`;\r\n }\r\n if (hasDebugParam) {\r\n botHeaders['X-Onto-Debug'] = 'true';\r\n }\r\n\r\n // --- Onto Control Plane Integration (Premium) ---\r\n const ONTO_API_KEY = process.env.ONTO_API_KEY;\r\n const DASHBOARD_URL = process.env.ONTO_DASHBOARD_URL || 'https://app.buildonto.dev';\r\n\r\n if (ONTO_API_KEY) {\r\n // 1. Fire-and-forget tracking\r\n fetch(`${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${payloadPath}.md`;\r\n\r\n // Rewrite implicitly serves the target URL transparently to the client.\r\n const response = NextResponse.rewrite(url);\r\n \r\n // Explicitly set headers on the rewrite response\r\n response.headers.set('Content-Type', 'text/markdown; charset=utf-8');\r\n response.headers.set('Cache-Control', 'public, max-age=3600, s-maxage=3600, stale-while-revalidate=86400');\r\n \r\n if (matched) {\r\n response.headers.set('X-Onto-Bot', `${matched.name} (${matched.company})`);\r\n }\r\n if (hasDebugParam) {\r\n response.headers.set('X-Onto-Debug', 'true');\r\n }\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,EAAcC,EAAqB,CACpE,IAAMC,EAAYF,EAAQ,QAAQ,IAAI,YAAY,EAC5CG,EAAMH,EAAQ,QAAQ,MAAM,EAC5BI,EAAUC,EAASH,CAAS,EAE5BI,EAASN,EAAQ,QAAQ,IAAI,QAAQ,GAAK,GAC1CO,EAAgBP,EAAQ,QAAQ,aAAa,IAAI,MAAM,EAEvDQ,EAAU,CAAC,CAACJ,EACZK,EAAsBH,EAAO,SAAS,eAAe,GAAKC,EAGhE,GAAIC,GAAWC,EAAqB,CAGhC,GAAIN,EAAI,SAAS,WAAW,QAAQ,EAChC,OAAO,eAAa,KAAK,EAK7B,GAAIA,EAAI,WAAa,YACjB,GAAI,CACA,GAAIF,EAAQ,CAER,IAAMS,EAAiBC,EAAgBV,CAAM,EACvCW,EAAW,IAAI,eAAaF,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,EAAW,eAAa,QAAQT,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,EAAW,eAAa,QAAQT,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,OAAO,eAAa,KAAK,EAI7B,IAAIW,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,KAE9DG,IACAQ,EAAW,cAAc,EAAI,QAIjC,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,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,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,CAGApB,EAAI,SAAW,SAASW,CAAW,MAGnC,IAAMF,EAAW,eAAa,QAAQT,CAAG,EAGzC,OAAAS,EAAS,QAAQ,IAAI,eAAgB,8BAA8B,EACnEA,EAAS,QAAQ,IAAI,gBAAiB,mEAAmE,EAErGR,GACAQ,EAAS,QAAQ,IAAI,aAAc,GAAGR,EAAQ,IAAI,KAAKA,EAAQ,OAAO,GAAG,EAEzEG,GACAK,EAAS,QAAQ,IAAI,eAAgB,MAAM,EAGxCA,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","hasDebugParam","isAiBot","isMarkdownRequested","llmsTxtContent","generateLlmsTxt","response","error","payloadPath","botHeaders","ONTO_API_KEY","DASHBOARD_URL","injectRes","injection","localMdUrl","mdRes","finalMarkdown","err"]}
|
package/dist/middleware.mjs
CHANGED
|
@@ -1,8 +1,8 @@
|
|
|
1
|
-
import{NextResponse as p}from"next/server";var
|
|
1
|
+
import{NextResponse as p}from"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"}],B=x.map(t=>t.name);function u(t){if(!t)return;let e=t.toLowerCase();return x.find(o=>e.includes(o.name.toLowerCase()))}function y(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(`## ${o.heading}`),e.push(""),e.push(o.content),e.push("");return e.join(`
|
|
2
2
|
`).trim()+`
|
|
3
|
-
`}async function G(t,e){let o=t.headers.get("user-agent"),a=t.nextUrl.clone(),n=
|
|
3
|
+
`}async function G(t,e){let o=t.headers.get("user-agent"),a=t.nextUrl.clone(),n=u(o),f=t.headers.get("accept")||"",m=t.nextUrl.searchParams.has("onto"),O=!!n,$=f.includes("text/markdown")||m;if(O||$){if(a.pathname.startsWith("/_next"))return p.next();if(a.pathname==="/llms.txt")try{if(e){let r=y(e),i=new p(r,{headers:{"Content-Type":"text/plain; charset=utf-8","Cache-Control":"public, max-age=3600, s-maxage=3600, stale-while-revalidate=86400"}});return n&&i.headers.set("X-Onto-Bot",`${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} (${n.company})`),m&&(l["X-Onto-Debug"]="true");let h=process.env.ONTO_API_KEY,d=process.env.ONTO_DASHBOARD_URL||"https://app.buildonto.dev";if(h){fetch(`${d}/api/track`,{method:"POST",headers:{"x-onto-key":h,"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(`${d}/api/sdk/inject?route=${a.pathname}`,{headers:{"x-onto-key":h},signal:AbortSignal.timeout(1500)});if(r.ok){let{injection:i}=await r.json();if(i){let C=`${a.origin}/.onto${s}.md`,g=await fetch(C);if(g.ok){let A=`${await g.text()}
|
|
4
4
|
|
|
5
5
|
---
|
|
6
6
|
|
|
7
|
-
${i}`;return new p(A,{headers:{...l,"X-Onto-Injected":"true"}})}}}}catch(r){console.error("[Onto] Injection failed",r)}}a.pathname=`/.onto${s}.md`;let
|
|
7
|
+
${i}`;return new p(A,{headers:{...l,"X-Onto-Injected":"true"}})}}}}catch(r){console.error("[Onto] Injection failed",r)}}a.pathname=`/.onto${s}.md`;let c=p.rewrite(a);return c.headers.set("Content-Type","text/markdown; charset=utf-8"),c.headers.set("Cache-Control","public, max-age=3600, s-maxage=3600, stale-while-revalidate=86400"),n&&c.headers.set("X-Onto-Bot",`${n.name} (${n.company})`),m&&c.headers.set("X-Onto-Debug","true"),c}return p.next()}export{B as AI_BOT_USER_AGENTS,u as matchBot,G as ontoMiddleware};
|
|
8
8
|
//# sourceMappingURL=middleware.mjs.map
|
package/dist/middleware.mjs.map
CHANGED
|
@@ -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 { generateLlmsTxt, OntoConfig } from './config';\r\n\r\nexport async function ontoMiddleware(request: any, config?: OntoConfig) {\r\n const userAgent = request.headers.get('user-agent');\r\n const url = request.nextUrl.clone();\r\n const matched = matchBot(userAgent);\r\n\r\n const accept = request.headers.get('accept') || '';\r\n\r\n const isAiBot = !!matched;\r\n const isMarkdownRequested = accept.includes('text/markdown');\r\n\r\n // If traffic is identified as an AI Bot or markdown is requested\r\n if (isAiBot || isMarkdownRequested) {\r\n\r\n // Ignore internal next.js requests & static assets (but not llms.txt)\r\n if (url.pathname.startsWith('/_next')) {\r\n return NextResponse.next();\r\n }\r\n\r\n // --- llms.txt Auto-Discovery ---\r\n // Serve the llms.txt manifest to AI agents\r\n if (url.pathname === '/llms.txt') {\r\n try {\r\n if (config) {\r\n // Generate llms.txt dynamically from config\r\n const llmsTxtContent = generateLlmsTxt(config);\r\n const response = new NextResponse(llmsTxtContent, {\r\n headers: {\r\n 'Content-Type': 'text/plain; charset=utf-8',\r\n 'Cache-Control': 'public, max-age=3600, s-maxage=3600, stale-while-revalidate=86400',\r\n }\r\n });\r\n\r\n if (matched) {\r\n response.headers.set('X-Onto-Bot', `${matched.name} (${matched.company})`);\r\n }\r\n\r\n return response;\r\n } else {\r\n // Fallback: try to serve static llms.txt from public folder\r\n url.pathname = '/llms.txt';\r\n const response = NextResponse.rewrite(url);\r\n response.headers.set('Content-Type', 'text/plain; charset=utf-8');\r\n response.headers.set('Cache-Control', 'public, max-age=3600, s-maxage=3600, stale-while-revalidate=86400');\r\n if (matched) {\r\n response.headers.set('X-Onto-Bot', `${matched.name} (${matched.company})`);\r\n }\r\n return response;\r\n }\r\n } catch (error) {\r\n console.error('[Onto] Failed to generate llms.txt:', error);\r\n // Fallback to static file on error\r\n url.pathname = '/llms.txt';\r\n const response = NextResponse.rewrite(url);\r\n response.headers.set('Content-Type', 'text/plain; charset=utf-8');\r\n if (matched) {\r\n response.headers.set('X-Onto-Bot', `${matched.name} (${matched.company})`);\r\n }\r\n return response;\r\n }\r\n }\r\n\r\n // Skip other static assets\r\n if (url.pathname.includes('.')) {\r\n return NextResponse.next();\r\n }\r\n\r\n // Determine the corresponding payload path\r\n let payloadPath = url.pathname;\r\n if (payloadPath === '/' || payloadPath === '') {\r\n payloadPath = '/index';\r\n }\r\n\r\n // Strip trailing slash if present\r\n if (payloadPath.endsWith('/') && payloadPath !== '/') {\r\n payloadPath = payloadPath.slice(0, -1);\r\n }\r\n\r\n // Common response headers for all bot responses\r\n const botHeaders: Record<string, string> = {\r\n 'Content-Type': 'text/markdown; charset=utf-8',\r\n 'Cache-Control': 'public, max-age=3600, s-maxage=3600, stale-while-revalidate=86400',\r\n };\r\n if (matched) {\r\n botHeaders['X-Onto-Bot'] = `${matched.name} (${matched.company})`;\r\n }\r\n\r\n // --- Onto Control Plane Integration (Premium) ---\r\n const ONTO_API_KEY = process.env.ONTO_API_KEY;\r\n const DASHBOARD_URL = process.env.ONTO_DASHBOARD_URL || 'https://app.buildonto.dev';\r\n\r\n if (ONTO_API_KEY) {\r\n // 1. Fire-and-forget tracking — includes structured bot info\r\n fetch(`${DASHBOARD_URL}/api/track`, {\r\n method: 'POST',\r\n headers: {\r\n 'x-onto-key': ONTO_API_KEY,\r\n 'Content-Type': 'application/json'\r\n },\r\n body: JSON.stringify({\r\n route: url.pathname,\r\n userAgent: userAgent,\r\n bot: matched ? matched.name : null,\r\n company: matched ? matched.company : null,\r\n })\r\n }).catch(() => {});\r\n\r\n // 2. Dynamic Context Injection\r\n try {\r\n const injectRes = await fetch(`${DASHBOARD_URL}/api/sdk/inject?route=${url.pathname}`, {\r\n headers: { 'x-onto-key': ONTO_API_KEY },\r\n signal: AbortSignal.timeout(1500)\r\n });\r\n\r\n if (injectRes.ok) {\r\n const { injection } = await injectRes.json();\r\n \r\n if (injection) {\r\n const localMdUrl = `${url.origin}/.onto${payloadPath}.md`;\r\n const mdRes = await fetch(localMdUrl);\r\n \r\n if (mdRes.ok) {\r\n const baseMarkdown = await mdRes.text();\r\n const finalMarkdown = `${baseMarkdown}\\n\\n---\\n\\n${injection}`;\r\n \r\n return new NextResponse(finalMarkdown, {\r\n headers: {\r\n ...botHeaders,\r\n 'X-Onto-Injected': 'true'\r\n }\r\n });\r\n }\r\n }\r\n }\r\n } catch (err) {\r\n console.error('[Onto] Injection failed', err);\r\n }\r\n }\r\n // ------------------------------------------------\r\n\r\n url.pathname = `/.onto${payloadPath}.md`;\r\n\r\n // Rewrite implicitly serves the target URL transparently to the client.\r\n const response = NextResponse.rewrite(url);\r\n if (matched) {\r\n response.headers.set('X-Onto-Bot', `${matched.name} (${matched.company})`);\r\n }\r\n return response;\r\n }\r\n\r\n return NextResponse.next();\r\n}\r\n\r\n// Re-export the bot registry for consumers who want to extend or inspect it\r\nexport { AI_BOT_USER_AGENTS, matchBot } from './bots';\r\nexport type { AiBot } from './bots';\r\n\r\n\r\n","/**\n * Comprehensive registry of AI bot user-agent strings.\n * The middleware uses this list to detect AI crawlers and serve optimized markdown.\n */\n\nexport interface AiBot {\n /** The user-agent substring to match against */\n name: string;\n /** The company operating this bot */\n company: string;\n}\n\n/**\n * Structured registry of all known AI bots, grouped by company.\n * Useful for analytics and the Control Plane dashboard.\n */\nexport const AI_BOTS: AiBot[] = [\n // OpenAI\n { name: 'GPTBot', company: 'OpenAI' },\n { name: 'ChatGPT-User', company: 'OpenAI' },\n { name: 'OAI-SearchBot', company: 'OpenAI' },\n\n // Google\n { name: 'Googlebot', company: 'Google' },\n { name: 'Google-CloudVertexBot', company: 'Google' },\n { name: 'Google-Extended', company: 'Google' },\n { name: 'GoogleOther', company: 'Google' },\n\n // Anthropic\n { name: 'ClaudeBot', company: 'Anthropic' },\n { name: 'Claude-User', company: 'Anthropic' },\n { name: 'anthropic-ai', company: 'Anthropic' },\n\n // Perplexity\n { name: 'PerplexityBot', company: 'Perplexity' },\n { name: 'Perplexity-User', company: 'Perplexity' },\n\n // Meta\n { name: 'Meta-ExternalAgent', company: 'Meta' },\n { name: 'Meta-ExternalFetcher', company: 'Meta' },\n { name: 'FacebookBot', company: 'Meta' },\n\n // Common Crawl (used by most smaller AI companies)\n { name: 'CCBot', company: 'Common Crawl' },\n\n // Other notable AI crawlers\n { name: 'Bytespider', company: 'ByteDance' },\n { name: 'Applebot-Extended', company: 'Apple' },\n { name: 'cohere-ai', company: 'Cohere' },\n { name: 'YouBot', company: 'You.com' },\n];\n\n/**\n * Flat list of user-agent substrings for fast matching in the middleware.\n */\nexport const AI_BOT_USER_AGENTS: string[] = AI_BOTS.map(bot => bot.name);\n\n/**\n * Given a raw user-agent string, returns the matched AiBot entry or undefined.\n * Comparison is case-insensitive to handle inconsistent agent casing.\n */\nexport function matchBot(userAgent: string | null): AiBot | undefined {\n if (!userAgent) return undefined;\n const lowerUA = userAgent.toLowerCase();\n return AI_BOTS.find(bot => \n lowerUA.includes(bot.name.toLowerCase())\n );\n}\n","/**\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,EAAcC,EAAqB,CACpE,IAAMC,EAAYF,EAAQ,QAAQ,IAAI,YAAY,EAC5CG,EAAMH,EAAQ,QAAQ,MAAM,EAC5BI,EAAUC,EAASH,CAAS,EAE5BI,EAASN,EAAQ,QAAQ,IAAI,QAAQ,GAAK,GAE1CO,EAAU,CAAC,CAACH,EACZI,EAAsBF,EAAO,SAAS,eAAe,EAG3D,GAAIC,GAAWC,EAAqB,CAGhC,GAAIL,EAAI,SAAS,WAAW,QAAQ,EAChC,OAAOM,EAAa,KAAK,EAK7B,GAAIN,EAAI,WAAa,YACjB,GAAI,CACA,GAAIF,EAAQ,CAER,IAAMS,EAAiBC,EAAgBV,CAAM,EACvCW,EAAW,IAAIH,EAAaC,EAAgB,CAC9C,QAAS,CACL,eAAgB,4BAChB,gBAAiB,mEACrB,CACJ,CAAC,EAED,OAAIN,GACAQ,EAAS,QAAQ,IAAI,aAAc,GAAGR,EAAQ,IAAI,KAAKA,EAAQ,OAAO,GAAG,EAGtEQ,CACX,KAAO,CAEHT,EAAI,SAAW,YACf,IAAMS,EAAWH,EAAa,QAAQN,CAAG,EACzC,OAAAS,EAAS,QAAQ,IAAI,eAAgB,2BAA2B,EAChEA,EAAS,QAAQ,IAAI,gBAAiB,mEAAmE,EACrGR,GACAQ,EAAS,QAAQ,IAAI,aAAc,GAAGR,EAAQ,IAAI,KAAKA,EAAQ,OAAO,GAAG,EAEtEQ,CACX,CACJ,OAASC,EAAO,CACZ,QAAQ,MAAM,sCAAuCA,CAAK,EAE1DV,EAAI,SAAW,YACf,IAAMS,EAAWH,EAAa,QAAQN,CAAG,EACzC,OAAAS,EAAS,QAAQ,IAAI,eAAgB,2BAA2B,EAC5DR,GACAQ,EAAS,QAAQ,IAAI,aAAc,GAAGR,EAAQ,IAAI,KAAKA,EAAQ,OAAO,GAAG,EAEtEQ,CACX,CAIJ,GAAIT,EAAI,SAAS,SAAS,GAAG,EACzB,OAAOM,EAAa,KAAK,EAI7B,IAAIK,EAAcX,EAAI,UAClBW,IAAgB,KAAOA,IAAgB,MACvCA,EAAc,UAIdA,EAAY,SAAS,GAAG,GAAKA,IAAgB,MAC7CA,EAAcA,EAAY,MAAM,EAAG,EAAE,GAIzC,IAAMC,EAAqC,CACvC,eAAgB,+BAChB,gBAAiB,mEACrB,EACIX,IACAW,EAAW,YAAY,EAAI,GAAGX,EAAQ,IAAI,KAAKA,EAAQ,OAAO,KAIlE,IAAMY,EAAe,QAAQ,IAAI,aAC3BC,EAAgB,QAAQ,IAAI,oBAAsB,4BAExD,GAAID,EAAc,CAEd,MAAM,GAAGC,CAAa,aAAc,CAChC,OAAQ,OACR,QAAS,CACL,aAAcD,EACd,eAAgB,kBACpB,EACA,KAAM,KAAK,UAAU,CACjB,MAAOb,EAAI,SACX,UAAWD,EACX,IAAKE,EAAUA,EAAQ,KAAO,KAC9B,QAASA,EAAUA,EAAQ,QAAU,IACzC,CAAC,CACL,CAAC,EAAE,MAAM,IAAM,CAAC,CAAC,EAGjB,GAAI,CACA,IAAMc,EAAY,MAAM,MAAM,GAAGD,CAAa,yBAAyBd,EAAI,QAAQ,GAAI,CACnF,QAAS,CAAE,aAAca,CAAa,EACtC,OAAQ,YAAY,QAAQ,IAAI,CACpC,CAAC,EAED,GAAIE,EAAU,GAAI,CACd,GAAM,CAAE,UAAAC,CAAU,EAAI,MAAMD,EAAU,KAAK,EAE3C,GAAIC,EAAW,CACX,IAAMC,EAAa,GAAGjB,EAAI,MAAM,SAASW,CAAW,MAC9CO,EAAQ,MAAM,MAAMD,CAAU,EAEpC,GAAIC,EAAM,GAAI,CAEV,IAAMC,EAAgB,GADD,MAAMD,EAAM,KAAK,CACD;AAAA;AAAA;AAAA;AAAA,EAAcF,CAAS,GAE5D,OAAO,IAAIV,EAAaa,EAAe,CACnC,QAAS,CACL,GAAGP,EACH,kBAAmB,MACvB,CACJ,CAAC,CACL,CACJ,CACJ,CACJ,OAASQ,EAAK,CACV,QAAQ,MAAM,0BAA2BA,CAAG,CAChD,CACJ,CAGApB,EAAI,SAAW,SAASW,CAAW,MAGnC,IAAMF,EAAWH,EAAa,QAAQN,CAAG,EACzC,OAAIC,GACAQ,EAAS,QAAQ,IAAI,aAAc,GAAGR,EAAQ,IAAI,KAAKA,EAAQ,OAAO,GAAG,EAEtEQ,CACX,CAEA,OAAOH,EAAa,KAAK,CAC7B","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"]}
|
|
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: any, config?: OntoConfig) {\r\n const userAgent = request.headers.get('user-agent');\r\n const url = request.nextUrl.clone();\r\n const matched = matchBot(userAgent);\r\n\r\n const accept = request.headers.get('accept') || '';\r\n const hasDebugParam = request.nextUrl.searchParams.has('onto');\r\n\r\n const isAiBot = !!matched;\r\n const isMarkdownRequested = accept.includes('text/markdown') || hasDebugParam;\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} (${matched.company})`;\r\n }\r\n if (hasDebugParam) {\r\n botHeaders['X-Onto-Debug'] = 'true';\r\n }\r\n\r\n // --- Onto Control Plane Integration (Premium) ---\r\n const ONTO_API_KEY = process.env.ONTO_API_KEY;\r\n const DASHBOARD_URL = process.env.ONTO_DASHBOARD_URL || 'https://app.buildonto.dev';\r\n\r\n if (ONTO_API_KEY) {\r\n // 1. Fire-and-forget tracking\r\n fetch(`${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${payloadPath}.md`;\r\n\r\n // Rewrite implicitly serves the target URL transparently to the client.\r\n const response = NextResponse.rewrite(url);\r\n \r\n // Explicitly set headers on the rewrite response\r\n response.headers.set('Content-Type', 'text/markdown; charset=utf-8');\r\n response.headers.set('Cache-Control', 'public, max-age=3600, s-maxage=3600, stale-while-revalidate=86400');\r\n \r\n if (matched) {\r\n response.headers.set('X-Onto-Bot', `${matched.name} (${matched.company})`);\r\n }\r\n if (hasDebugParam) {\r\n response.headers.set('X-Onto-Debug', 'true');\r\n }\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,EAAcC,EAAqB,CACpE,IAAMC,EAAYF,EAAQ,QAAQ,IAAI,YAAY,EAC5CG,EAAMH,EAAQ,QAAQ,MAAM,EAC5BI,EAAUC,EAASH,CAAS,EAE5BI,EAASN,EAAQ,QAAQ,IAAI,QAAQ,GAAK,GAC1CO,EAAgBP,EAAQ,QAAQ,aAAa,IAAI,MAAM,EAEvDQ,EAAU,CAAC,CAACJ,EACZK,EAAsBH,EAAO,SAAS,eAAe,GAAKC,EAGhE,GAAIC,GAAWC,EAAqB,CAGhC,GAAIN,EAAI,SAAS,WAAW,QAAQ,EAChC,OAAOO,EAAa,KAAK,EAK7B,GAAIP,EAAI,WAAa,YACjB,GAAI,CACA,GAAIF,EAAQ,CAER,IAAMU,EAAiBC,EAAgBX,CAAM,EACvCY,EAAW,IAAIH,EAAaC,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,EAAWH,EAAa,QAAQP,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,EAAWH,EAAa,QAAQP,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,OAAOO,EAAa,KAAK,EAI7B,IAAIK,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,KAE9DG,IACAS,EAAW,cAAc,EAAI,QAIjC,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,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,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,CAGArB,EAAI,SAAW,SAASY,CAAW,MAGnC,IAAMF,EAAWH,EAAa,QAAQP,CAAG,EAGzC,OAAAU,EAAS,QAAQ,IAAI,eAAgB,8BAA8B,EACnEA,EAAS,QAAQ,IAAI,gBAAiB,mEAAmE,EAErGT,GACAS,EAAS,QAAQ,IAAI,aAAc,GAAGT,EAAQ,IAAI,KAAKA,EAAQ,OAAO,GAAG,EAEzEG,GACAM,EAAS,QAAQ,IAAI,eAAgB,MAAM,EAGxCA,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","hasDebugParam","isAiBot","isMarkdownRequested","NextResponse","llmsTxtContent","generateLlmsTxt","response","error","payloadPath","botHeaders","ONTO_API_KEY","DASHBOARD_URL","injectRes","injection","localMdUrl","mdRes","finalMarkdown","err"]}
|
package/package.json
CHANGED
package/src/middleware.ts
CHANGED
|
@@ -8,9 +8,10 @@ export async function ontoMiddleware(request: any, config?: OntoConfig) {
|
|
|
8
8
|
const matched = matchBot(userAgent);
|
|
9
9
|
|
|
10
10
|
const accept = request.headers.get('accept') || '';
|
|
11
|
+
const hasDebugParam = request.nextUrl.searchParams.has('onto');
|
|
11
12
|
|
|
12
13
|
const isAiBot = !!matched;
|
|
13
|
-
const isMarkdownRequested = accept.includes('text/markdown');
|
|
14
|
+
const isMarkdownRequested = accept.includes('text/markdown') || hasDebugParam;
|
|
14
15
|
|
|
15
16
|
// If traffic is identified as an AI Bot or markdown is requested
|
|
16
17
|
if (isAiBot || isMarkdownRequested) {
|
|
@@ -87,13 +88,16 @@ export async function ontoMiddleware(request: any, config?: OntoConfig) {
|
|
|
87
88
|
if (matched) {
|
|
88
89
|
botHeaders['X-Onto-Bot'] = `${matched.name} (${matched.company})`;
|
|
89
90
|
}
|
|
91
|
+
if (hasDebugParam) {
|
|
92
|
+
botHeaders['X-Onto-Debug'] = 'true';
|
|
93
|
+
}
|
|
90
94
|
|
|
91
95
|
// --- Onto Control Plane Integration (Premium) ---
|
|
92
96
|
const ONTO_API_KEY = process.env.ONTO_API_KEY;
|
|
93
97
|
const DASHBOARD_URL = process.env.ONTO_DASHBOARD_URL || 'https://app.buildonto.dev';
|
|
94
98
|
|
|
95
99
|
if (ONTO_API_KEY) {
|
|
96
|
-
// 1. Fire-and-forget tracking
|
|
100
|
+
// 1. Fire-and-forget tracking
|
|
97
101
|
fetch(`${DASHBOARD_URL}/api/track`, {
|
|
98
102
|
method: 'POST',
|
|
99
103
|
headers: {
|
|
@@ -145,9 +149,18 @@ export async function ontoMiddleware(request: any, config?: OntoConfig) {
|
|
|
145
149
|
|
|
146
150
|
// Rewrite implicitly serves the target URL transparently to the client.
|
|
147
151
|
const response = NextResponse.rewrite(url);
|
|
152
|
+
|
|
153
|
+
// Explicitly set headers on the rewrite response
|
|
154
|
+
response.headers.set('Content-Type', 'text/markdown; charset=utf-8');
|
|
155
|
+
response.headers.set('Cache-Control', 'public, max-age=3600, s-maxage=3600, stale-while-revalidate=86400');
|
|
156
|
+
|
|
148
157
|
if (matched) {
|
|
149
158
|
response.headers.set('X-Onto-Bot', `${matched.name} (${matched.company})`);
|
|
150
159
|
}
|
|
160
|
+
if (hasDebugParam) {
|
|
161
|
+
response.headers.set('X-Onto-Debug', 'true');
|
|
162
|
+
}
|
|
163
|
+
|
|
151
164
|
return response;
|
|
152
165
|
}
|
|
153
166
|
|