@ontosdk/next 1.2.0 → 1.3.1

package/dist/middleware.js.map

@@ -1 +1 @@
-{"version":3,"sources":["../src/middleware.ts","../src/bots.ts"],"sourcesContent":["import { NextRequest, NextResponse } from 'next/server';\r\nimport { AI_BOT_USER_AGENTS, matchBot } from './bots';\r\n\r\nexport async function ontoMiddleware(request: NextRequest) {\r\n    const userAgent = request.headers.get('user-agent') || '';\r\n    const accept = request.headers.get('accept') || '';\r\n\r\n    const matched = matchBot(userAgent);\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, rewrite the URL\r\n    if (isAiBot || isMarkdownRequested) {\r\n        const url = request.nextUrl.clone();\r\n\r\n        // Ignore internal next.js requests & static assets\r\n        if (url.pathname.startsWith('/_next') || 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(`${DASHBOARD_URL}/api/track`, {\r\n                method: 'POST',\r\n                headers: {\r\n                    'x-onto-key': ONTO_API_KEY,\r\n                    'Content-Type': 'application/json'\r\n                },\r\n                body: JSON.stringify({\r\n                    route: url.pathname,\r\n                    userAgent: userAgent,\r\n                    bot: matched ? matched.name : null,\r\n                    company: matched ? matched.company : null,\r\n                })\r\n            }).catch(() => {});\r\n\r\n            // 2. Dynamic Context Injection\r\n            try {\r\n                // Fetch the injection from the Control Plane\r\n                const injectRes = await fetch(`${DASHBOARD_URL}/api/sdk/inject?route=${url.pathname}`, {\r\n                    headers: { 'x-onto-key': ONTO_API_KEY },\r\n                    // Set a strict timeout to keep edge fast\r\n                    signal: AbortSignal.timeout(1500)\r\n                });\r\n\r\n                if (injectRes.ok) {\r\n                    const { injection } = await injectRes.json();\r\n                    \r\n                    if (injection) {\r\n                        // To inject, we must fetch the local markdown and append\r\n                        const localMdUrl = `${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        // Attach bot identification headers to the rewrite response\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","/**\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): AiBot | undefined {\n    const ua = userAgent.toLowerCase();\n    return AI_BOTS.find(bot => ua.includes(bot.name.toLowerCase()));\n}\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,EAAsC,CAC3D,IAAMC,EAAKD,EAAU,YAAY,EACjC,OAAOJ,EAAQ,KAAKE,GAAOG,EAAG,SAASH,EAAI,KAAK,YAAY,CAAC,CAAC,CAClE,CD7DA,eAAsBI,EAAeC,EAAsB,CACvD,IAAMC,EAAYD,EAAQ,QAAQ,IAAI,YAAY,GAAK,GACjDE,EAASF,EAAQ,QAAQ,IAAI,QAAQ,GAAK,GAE1CG,EAAUC,EAASH,CAAS,EAC5BI,EAAU,CAAC,CAACF,EACZG,EAAsBJ,EAAO,SAAS,eAAe,EAG3D,GAAIG,GAAWC,EAAqB,CAChC,IAAMC,EAAMP,EAAQ,QAAQ,MAAM,EAGlC,GAAIO,EAAI,SAAS,WAAW,QAAQ,GAAKA,EAAI,SAAS,SAAS,GAAG,EAC9D,OAAO,eAAa,KAAK,EAI7B,IAAIC,EAAcD,EAAI,UAClBC,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,EACIN,IACAM,EAAW,YAAY,EAAI,GAAGN,EAAQ,IAAI,KAAKA,EAAQ,OAAO,KAIlE,IAAMO,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,MAAOH,EAAI,SACX,UAAWN,EACX,IAAKE,EAAUA,EAAQ,KAAO,KAC9B,QAASA,EAAUA,EAAQ,QAAU,IACzC,CAAC,CACL,CAAC,EAAE,MAAM,IAAM,CAAC,CAAC,EAGjB,GAAI,CAEA,IAAMS,EAAY,MAAM,MAAM,GAAGD,CAAa,yBAAyBJ,EAAI,QAAQ,GAAI,CACnF,QAAS,CAAE,aAAcG,CAAa,EAEtC,OAAQ,YAAY,QAAQ,IAAI,CACpC,CAAC,EAED,GAAIE,EAAU,GAAI,CACd,GAAM,CAAE,UAAAC,CAAU,EAAI,MAAMD,EAAU,KAAK,EAE3C,GAAIC,EAAW,CAEX,IAAMC,EAAa,GAAGP,EAAI,MAAM,SAASC,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,CAGAV,EAAI,SAAW,SAASC,CAAW,MAGnC,IAAMU,EAAW,eAAa,QAAQX,CAAG,EAEzC,OAAIJ,GACAe,EAAS,QAAQ,IAAI,aAAc,GAAGf,EAAQ,IAAI,KAAKA,EAAQ,OAAO,GAAG,EAEtEe,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","ua","ontoMiddleware","request","userAgent","accept","matched","matchBot","isAiBot","isMarkdownRequested","url","payloadPath","botHeaders","ONTO_API_KEY","DASHBOARD_URL","injectRes","injection","localMdUrl","mdRes","finalMarkdown","err","response"]}
+{"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"]}

package/dist/middleware.mjs

@@ -1,6 +1,8 @@
-import{NextResponse as c}from"next/server";var u=[{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=u.map(o=>o.name);function s(o){let a=o.toLowerCase();return u.find(m=>a.includes(m.name.toLowerCase()))}async function k(o){let a=o.headers.get("user-agent")||"",m=o.headers.get("accept")||"",e=s(a),x=!!e,A=m.includes("text/markdown");if(x||A){let n=o.nextUrl.clone();if(n.pathname.startsWith("/_next")||n.pathname.includes("."))return c.next();let t=n.pathname;(t==="/"||t==="")&&(t="/index"),t.endsWith("/")&&t!=="/"&&(t=t.slice(0,-1));let p={"Content-Type":"text/markdown; charset=utf-8","Cache-Control":"public, max-age=3600, s-maxage=3600, stale-while-revalidate=86400"};e&&(p["X-Onto-Bot"]=`${e.name} (${e.company})`);let i=process.env.ONTO_API_KEY,l=process.env.ONTO_DASHBOARD_URL||"https://app.buildonto.dev";if(i){fetch(`${l}/api/track`,{method:"POST",headers:{"x-onto-key":i,"Content-Type":"application/json"},body:JSON.stringify({route:n.pathname,userAgent:a,bot:e?e.name:null,company:e?e.company:null})}).catch(()=>{});try{let r=await fetch(`${l}/api/sdk/inject?route=${n.pathname}`,{headers:{"x-onto-key":i},signal:AbortSignal.timeout(1500)});if(r.ok){let{injection:y}=await r.json();if(y){let f=`${n.origin}/.onto${t}.md`,h=await fetch(f);if(h.ok){let g=`${await h.text()}
+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(`## ${o.heading}`),t.push(""),t.push(o.content),t.push("");return t.join(`
+`).trim()+`
+`}async function G(e,t){let o=e.headers.get("user-agent"),a=e.nextUrl.clone(),n=l(o),O=e.headers.get("accept")||"",C=!!n,$=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()}
 
 ---
 
-${y}`;return new c(g,{headers:{...p,"X-Onto-Injected":"true"}})}}}}catch(r){console.error("[Onto] Injection failed",r)}}n.pathname=`/.onto${t}.md`;let d=c.rewrite(n);return e&&d.headers.set("X-Onto-Bot",`${e.name} (${e.company})`),d}return c.next()}export{B as AI_BOT_USER_AGENTS,s as matchBot,k as ontoMiddleware};
+${r}`;return new c(A,{headers:{...u,"X-Onto-Injected":"true"}})}}}}catch(i){console.error("[Onto] Injection failed",i)}}a.pathname=`/.onto${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};
 //# sourceMappingURL=middleware.mjs.map

package/dist/middleware.mjs.map

@@ -1 +1 @@
-{"version":3,"sources":["../src/middleware.ts","../src/bots.ts"],"sourcesContent":["import { NextRequest, NextResponse } from 'next/server';\r\nimport { AI_BOT_USER_AGENTS, matchBot } from './bots';\r\n\r\nexport async function ontoMiddleware(request: NextRequest) {\r\n    const userAgent = request.headers.get('user-agent') || '';\r\n    const accept = request.headers.get('accept') || '';\r\n\r\n    const matched = matchBot(userAgent);\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, rewrite the URL\r\n    if (isAiBot || isMarkdownRequested) {\r\n        const url = request.nextUrl.clone();\r\n\r\n        // Ignore internal next.js requests & static assets\r\n        if (url.pathname.startsWith('/_next') || 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(`${DASHBOARD_URL}/api/track`, {\r\n                method: 'POST',\r\n                headers: {\r\n                    'x-onto-key': ONTO_API_KEY,\r\n                    'Content-Type': 'application/json'\r\n                },\r\n                body: JSON.stringify({\r\n                    route: url.pathname,\r\n                    userAgent: userAgent,\r\n                    bot: matched ? matched.name : null,\r\n                    company: matched ? matched.company : null,\r\n                })\r\n            }).catch(() => {});\r\n\r\n            // 2. Dynamic Context Injection\r\n            try {\r\n                // Fetch the injection from the Control Plane\r\n                const injectRes = await fetch(`${DASHBOARD_URL}/api/sdk/inject?route=${url.pathname}`, {\r\n                    headers: { 'x-onto-key': ONTO_API_KEY },\r\n                    // Set a strict timeout to keep edge fast\r\n                    signal: AbortSignal.timeout(1500)\r\n                });\r\n\r\n                if (injectRes.ok) {\r\n                    const { injection } = await injectRes.json();\r\n                    \r\n                    if (injection) {\r\n                        // To inject, we must fetch the local markdown and append\r\n                        const localMdUrl = `${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        // Attach bot identification headers to the rewrite response\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","/**\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): AiBot | undefined {\n    const ua = userAgent.toLowerCase();\n    return AI_BOTS.find(bot => ua.includes(bot.name.toLowerCase()));\n}\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,EAAsC,CAC3D,IAAMC,EAAKD,EAAU,YAAY,EACjC,OAAOJ,EAAQ,KAAKE,GAAOG,EAAG,SAASH,EAAI,KAAK,YAAY,CAAC,CAAC,CAClE,CD7DA,eAAsBI,EAAeC,EAAsB,CACvD,IAAMC,EAAYD,EAAQ,QAAQ,IAAI,YAAY,GAAK,GACjDE,EAASF,EAAQ,QAAQ,IAAI,QAAQ,GAAK,GAE1CG,EAAUC,EAASH,CAAS,EAC5BI,EAAU,CAAC,CAACF,EACZG,EAAsBJ,EAAO,SAAS,eAAe,EAG3D,GAAIG,GAAWC,EAAqB,CAChC,IAAMC,EAAMP,EAAQ,QAAQ,MAAM,EAGlC,GAAIO,EAAI,SAAS,WAAW,QAAQ,GAAKA,EAAI,SAAS,SAAS,GAAG,EAC9D,OAAOC,EAAa,KAAK,EAI7B,IAAIC,EAAcF,EAAI,UAClBE,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,EACIP,IACAO,EAAW,YAAY,EAAI,GAAGP,EAAQ,IAAI,KAAKA,EAAQ,OAAO,KAIlE,IAAMQ,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,MAAOJ,EAAI,SACX,UAAWN,EACX,IAAKE,EAAUA,EAAQ,KAAO,KAC9B,QAASA,EAAUA,EAAQ,QAAU,IACzC,CAAC,CACL,CAAC,EAAE,MAAM,IAAM,CAAC,CAAC,EAGjB,GAAI,CAEA,IAAMU,EAAY,MAAM,MAAM,GAAGD,CAAa,yBAAyBL,EAAI,QAAQ,GAAI,CACnF,QAAS,CAAE,aAAcI,CAAa,EAEtC,OAAQ,YAAY,QAAQ,IAAI,CACpC,CAAC,EAED,GAAIE,EAAU,GAAI,CACd,GAAM,CAAE,UAAAC,CAAU,EAAI,MAAMD,EAAU,KAAK,EAE3C,GAAIC,EAAW,CAEX,IAAMC,EAAa,GAAGR,EAAI,MAAM,SAASE,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,IAAIN,EAAaS,EAAe,CACnC,QAAS,CACL,GAAGP,EACH,kBAAmB,MACvB,CACJ,CAAC,CACL,CACJ,CACJ,CACJ,OAASQ,EAAK,CACV,QAAQ,MAAM,0BAA2BA,CAAG,CAChD,CACJ,CAGAX,EAAI,SAAW,SAASE,CAAW,MAGnC,IAAMU,EAAWX,EAAa,QAAQD,CAAG,EAEzC,OAAIJ,GACAgB,EAAS,QAAQ,IAAI,aAAc,GAAGhB,EAAQ,IAAI,KAAKA,EAAQ,OAAO,GAAG,EAEtEgB,CACX,CAEA,OAAOX,EAAa,KAAK,CAC7B","names":["NextResponse","AI_BOTS","AI_BOT_USER_AGENTS","bot","matchBot","userAgent","ua","ontoMiddleware","request","userAgent","accept","matched","matchBot","isAiBot","isMarkdownRequested","url","NextResponse","payloadPath","botHeaders","ONTO_API_KEY","DASHBOARD_URL","injectRes","injection","localMdUrl","mdRes","finalMarkdown","err","response"]}
+{"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"]}

package/dist/schemas.d.mts

@@ -0,0 +1,73 @@
+import { OntoConfig } from './config.mjs';
+
+/**
+ * JSON-LD Schema generators for automatic structured data injection
+ * Follows Schema.org standards for AI-friendly metadata
+ */
+
+/**
+ * Standard AIO (AI Optimization) scoring methodology
+ * Based on the Onto scoring algorithm:
+ * - Content Negotiation: 40% (30 points)
+ * - React Tax / Token Efficiency: 35% (30 points)
+ * - Structured Data: 25% (25 points)
+ * - Semantic HTML: Bonus (15 points)
+ */
+interface AIOMethodologySchema {
+    '@context': 'https://schema.org';
+    '@type': 'HowTo';
+    name: string;
+    description: string;
+    step: Array<{
+        '@type': 'HowToStep';
+        name: string;
+        text: string;
+        position: number;
+    }>;
+}
+/**
+ * Generate AIO Scoring Methodology JSON-LD schema
+ * This explains to AI agents how the scoring system works
+ */
+declare function generateAIOMethodologySchema(config: OntoConfig, pageUrl: string): AIOMethodologySchema;
+/**
+ * Organization schema for About pages
+ */
+interface OrganizationSchema {
+    '@context': 'https://schema.org';
+    '@type': 'Organization';
+    name: string;
+    url?: string;
+    description?: string;
+    logo?: string;
+    foundingDate?: string;
+}
+/**
+ * Generate Organization JSON-LD schema for About pages
+ */
+declare function generateOrganizationSchema(config: OntoConfig, pageUrl: string): OrganizationSchema | null;
+/**
+ * AboutPage schema combining Organization and WebPage
+ */
+interface AboutPageSchema {
+    '@context': 'https://schema.org';
+    '@type': 'AboutPage';
+    name: string;
+    url: string;
+    description?: string;
+    mainEntity?: OrganizationSchema;
+}
+/**
+ * Generate AboutPage JSON-LD schema
+ */
+declare function generateAboutPageSchema(config: OntoConfig, pageUrl: string): AboutPageSchema;
+/**
+ * Determine which schema to generate based on page type
+ */
+declare function generateSchemaForPageType(pageType: 'scoring' | 'about' | 'default', config: OntoConfig, pageUrl: string): any | null;
+/**
+ * Serialize schema to JSON-LD script tag content
+ */
+declare function serializeSchema(schema: any | null): string | null;
+
+export { type AIOMethodologySchema, type AboutPageSchema, type OrganizationSchema, generateAIOMethodologySchema, generateAboutPageSchema, generateOrganizationSchema, generateSchemaForPageType, serializeSchema };

package/dist/schemas.d.ts

@@ -0,0 +1,73 @@
+import { OntoConfig } from './config.js';
+
+/**
+ * JSON-LD Schema generators for automatic structured data injection
+ * Follows Schema.org standards for AI-friendly metadata
+ */
+
+/**
+ * Standard AIO (AI Optimization) scoring methodology
+ * Based on the Onto scoring algorithm:
+ * - Content Negotiation: 40% (30 points)
+ * - React Tax / Token Efficiency: 35% (30 points)
+ * - Structured Data: 25% (25 points)
+ * - Semantic HTML: Bonus (15 points)
+ */
+interface AIOMethodologySchema {
+    '@context': 'https://schema.org';
+    '@type': 'HowTo';
+    name: string;
+    description: string;
+    step: Array<{
+        '@type': 'HowToStep';
+        name: string;
+        text: string;
+        position: number;
+    }>;
+}
+/**
+ * Generate AIO Scoring Methodology JSON-LD schema
+ * This explains to AI agents how the scoring system works
+ */
+declare function generateAIOMethodologySchema(config: OntoConfig, pageUrl: string): AIOMethodologySchema;
+/**
+ * Organization schema for About pages
+ */
+interface OrganizationSchema {
+    '@context': 'https://schema.org';
+    '@type': 'Organization';
+    name: string;
+    url?: string;
+    description?: string;
+    logo?: string;
+    foundingDate?: string;
+}
+/**
+ * Generate Organization JSON-LD schema for About pages
+ */
+declare function generateOrganizationSchema(config: OntoConfig, pageUrl: string): OrganizationSchema | null;
+/**
+ * AboutPage schema combining Organization and WebPage
+ */
+interface AboutPageSchema {
+    '@context': 'https://schema.org';
+    '@type': 'AboutPage';
+    name: string;
+    url: string;
+    description?: string;
+    mainEntity?: OrganizationSchema;
+}
+/**
+ * Generate AboutPage JSON-LD schema
+ */
+declare function generateAboutPageSchema(config: OntoConfig, pageUrl: string): AboutPageSchema;
+/**
+ * Determine which schema to generate based on page type
+ */
+declare function generateSchemaForPageType(pageType: 'scoring' | 'about' | 'default', config: OntoConfig, pageUrl: string): any | null;
+/**
+ * Serialize schema to JSON-LD script tag content
+ */
+declare function serializeSchema(schema: any | null): string | null;
+
+export { type AIOMethodologySchema, type AboutPageSchema, type OrganizationSchema, generateAIOMethodologySchema, generateAboutPageSchema, generateOrganizationSchema, generateSchemaForPageType, serializeSchema };

package/dist/schemas.js

@@ -0,0 +1,2 @@
+"use strict";var a=Object.defineProperty;var g=Object.getOwnPropertyDescriptor;var p=Object.getOwnPropertyNames;var u=Object.prototype.hasOwnProperty;var m=(t,e)=>{for(var n in e)a(t,n,{get:e[n],enumerable:!0})},l=(t,e,n,o)=>{if(e&&typeof e=="object"||typeof e=="function")for(let i of p(e))!u.call(t,i)&&i!==n&&a(t,i,{get:()=>e[i],enumerable:!(o=g(e,i))||o.enumerable});return t};var h=t=>l(a({},"__esModule",{value:!0}),t);var y={};m(y,{generateAIOMethodologySchema:()=>r,generateAboutPageSchema:()=>c,generateOrganizationSchema:()=>s,generateSchemaForPageType:()=>f,serializeSchema:()=>d});module.exports=h(y);function r(t,e){return{"@context":"https://schema.org","@type":"HowTo",name:"AIO Score Calculation Methodology",description:"AI Optimization (AIO) Score measures how well a website is optimized for AI agents and LLM crawlers. Scored out of 100 points based on four key metrics.",step:[{"@type":"HowToStep",name:"Content Negotiation",text:"Check if the site responds to Accept: text/markdown header. Weight: 40%. Penalty: -30 points if missing. This ensures AI bots receive optimized content instead of heavy HTML.",position:1},{"@type":"HowToStep",name:"Token Efficiency (React Tax)",text:"Measure the ratio of visible text to total HTML size. Weight: 35%. Penalty: -30 points if HTML > 100KB but text < 1KB. Detects JavaScript-heavy sites that are difficult for AI to parse.",position:2},{"@type":"HowToStep",name:"Structured Data",text:"Verify presence of JSON-LD structured data (Schema.org). Weight: 25%. Penalty: -25 points if missing. Enables AI to confidently extract pricing, products, and entities.",position:3},{"@type":"HowToStep",name:"Semantic HTML",text:"Check for semantic tags like <main> and <article>. Bonus: +15 points if present. Helps AI agents separate navigation from core content.",position:4}]}}function s(t,e){if(!t.organization)return null;let n={"@context":"https://schema.org","@type":"Organization",name:t.organization.name};return t.organization.url&&(n.url=t.organization.url),t.organization.description&&(n.description=t.organization.description),t.organization.logo&&(n.logo=t.organization.logo),t.organization.foundingDate&&(n.foundingDate=t.organization.foundingDate),n}function c(t,e){let n=s(t,e),o={"@context":"https://schema.org","@type":"AboutPage",name:`About ${t.name}`,url:e};return t.summary&&(o.description=t.summary),n&&(o.mainEntity=n),o}function f(t,e,n){switch(t){case"scoring":return r(e,n);case"about":return c(e,n);default:return null}}function d(t){return t?JSON.stringify(t,null,2):null}0&&(module.exports={generateAIOMethodologySchema,generateAboutPageSchema,generateOrganizationSchema,generateSchemaForPageType,serializeSchema});
+//# sourceMappingURL=schemas.js.map

package/dist/schemas.js.map

@@ -0,0 +1 @@
+{"version":3,"sources":["../src/schemas.ts"],"sourcesContent":["/**\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":"yaAAA,IAAAA,EAAA,GAAAC,EAAAD,EAAA,kCAAAE,EAAA,4BAAAC,EAAA,+BAAAC,EAAA,8BAAAC,EAAA,oBAAAC,IAAA,eAAAC,EAAAP,GAgCO,SAASE,EACdM,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,SAASL,EACdI,EACAC,EAC2B,CAC3B,GAAI,CAACD,EAAO,aACV,OAAO,KAGT,IAAME,EAA6B,CACjC,WAAY,qBACZ,QAAS,eACT,KAAMF,EAAO,aAAa,IAC5B,EAEA,OAAIA,EAAO,aAAa,MACtBE,EAAO,IAAMF,EAAO,aAAa,KAG/BA,EAAO,aAAa,cACtBE,EAAO,YAAcF,EAAO,aAAa,aAGvCA,EAAO,aAAa,OACtBE,EAAO,KAAOF,EAAO,aAAa,MAGhCA,EAAO,aAAa,eACtBE,EAAO,aAAeF,EAAO,aAAa,cAGrCE,CACT,CAiBO,SAASP,EACdK,EACAC,EACiB,CACjB,IAAME,EAAYP,EAA2BI,EAAQC,CAAO,EAEtDC,EAA0B,CAC9B,WAAY,qBACZ,QAAS,YACT,KAAM,SAASF,EAAO,IAAI,GAC1B,IAAKC,CACP,EAEA,OAAID,EAAO,UACTE,EAAO,YAAcF,EAAO,SAG1BG,IACFD,EAAO,WAAaC,GAGfD,CACT,CAKO,SAASL,EACdO,EACAJ,EACAC,EACY,CACZ,OAAQG,EAAU,CAChB,IAAK,UACH,OAAOV,EAA6BM,EAAQC,CAAO,EACrD,IAAK,QACH,OAAON,EAAwBK,EAAQC,CAAO,EAEhD,QACE,OAAO,IACX,CACF,CAKO,SAASH,EAAgBI,EAAmC,CACjE,OAAKA,EAGE,KAAK,UAAUA,EAAQ,KAAM,CAAC,EAF5B,IAGX","names":["schemas_exports","__export","generateAIOMethodologySchema","generateAboutPageSchema","generateOrganizationSchema","generateSchemaForPageType","serializeSchema","__toCommonJS","config","pageUrl","schema","orgSchema","pageType"]}

package/dist/schemas.mjs

@@ -0,0 +1,2 @@
+function i(t,n){return{"@context":"https://schema.org","@type":"HowTo",name:"AIO Score Calculation Methodology",description:"AI Optimization (AIO) Score measures how well a website is optimized for AI agents and LLM crawlers. Scored out of 100 points based on four key metrics.",step:[{"@type":"HowToStep",name:"Content Negotiation",text:"Check if the site responds to Accept: text/markdown header. Weight: 40%. Penalty: -30 points if missing. This ensures AI bots receive optimized content instead of heavy HTML.",position:1},{"@type":"HowToStep",name:"Token Efficiency (React Tax)",text:"Measure the ratio of visible text to total HTML size. Weight: 35%. Penalty: -30 points if HTML > 100KB but text < 1KB. Detects JavaScript-heavy sites that are difficult for AI to parse.",position:2},{"@type":"HowToStep",name:"Structured Data",text:"Verify presence of JSON-LD structured data (Schema.org). Weight: 25%. Penalty: -25 points if missing. Enables AI to confidently extract pricing, products, and entities.",position:3},{"@type":"HowToStep",name:"Semantic HTML",text:"Check for semantic tags like <main> and <article>. Bonus: +15 points if present. Helps AI agents separate navigation from core content.",position:4}]}}function a(t,n){if(!t.organization)return null;let e={"@context":"https://schema.org","@type":"Organization",name:t.organization.name};return t.organization.url&&(e.url=t.organization.url),t.organization.description&&(e.description=t.organization.description),t.organization.logo&&(e.logo=t.organization.logo),t.organization.foundingDate&&(e.foundingDate=t.organization.foundingDate),e}function r(t,n){let e=a(t,n),o={"@context":"https://schema.org","@type":"AboutPage",name:`About ${t.name}`,url:n};return t.summary&&(o.description=t.summary),e&&(o.mainEntity=e),o}function s(t,n,e){switch(t){case"scoring":return i(n,e);case"about":return r(n,e);default:return null}}function c(t){return t?JSON.stringify(t,null,2):null}export{i as generateAIOMethodologySchema,r as generateAboutPageSchema,a as generateOrganizationSchema,s as generateSchemaForPageType,c as serializeSchema};
+//# sourceMappingURL=schemas.mjs.map

package/dist/schemas.mjs.map

@@ -0,0 +1 @@
+{"version":3,"sources":["../src/schemas.ts"],"sourcesContent":["/**\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":"AAgCO,SAASA,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":["generateAIOMethodologySchema","config","pageUrl","generateOrganizationSchema","schema","generateAboutPageSchema","orgSchema","generateSchemaForPageType","pageType","serializeSchema"]}

package/onto.config.example.ts

@@ -0,0 +1,111 @@
+import { OntoConfig } from './src/config';
+
+/**
+ * Example onto.config.ts file
+ *
+ * Place this file in your project root as `onto.config.ts` to enable
+ * dynamic llms.txt generation.
+ *
+ * The middleware will automatically read this config and generate
+ * /llms.txt on the fly when AI agents request it.
+ */
+const config: OntoConfig = {
+  // Required: The name of your project/site
+  name: 'My Awesome Project',
+
+  // Required: A concise summary for AI agents
+  summary: 'A modern web application for managing tasks and projects. Built with Next.js, TypeScript, and React. Provides a RESTful API and interactive UI.',
+
+  // Required: Base URL of your site
+  baseUrl: 'https://example.com',
+
+  // Optional: Key routes that AI agents should know about
+  routes: [
+    {
+      path: '/',
+      description: 'Homepage with product overview and key features',
+      pageType: 'default'
+    },
+    {
+      path: '/score',
+      description: 'AIO Score Calculator - check your AI optimization score',
+      pageType: 'scoring' // Automatically injects Methodology JSON-LD schema
+    },
+    {
+      path: '/about',
+      description: 'About our company and mission',
+      pageType: 'about' // Automatically injects Organization/AboutPage JSON-LD schema
+    },
+    {
+      path: '/docs',
+      description: 'Complete API documentation and integration guides'
+    },
+    {
+      path: '/docs/getting-started',
+      description: 'Quick start guide for new users'
+    },
+    {
+      path: '/api/reference',
+      description: 'REST API reference with all endpoints and schemas'
+    },
+    {
+      path: '/blog',
+      description: 'Technical blog posts and product updates'
+    },
+    {
+      path: '/pricing',
+      description: 'Pricing plans and feature comparison'
+    }
+  ],
+
+  // Optional: External resources
+  externalLinks: [
+    {
+      title: 'GitHub Repository',
+      url: 'https://github.com/example/project',
+      description: 'Open-source code and issue tracker'
+    },
+    {
+      title: 'API Status',
+      url: 'https://status.example.com',
+      description: 'Real-time API status and incident reports'
+    }
+  ],
+
+  // Optional: Organization info for JSON-LD schemas (used on 'about' pages)
+  organization: {
+    name: 'Example Corp',
+    description: 'Building the future of AI-optimized web applications',
+    url: 'https://example.com',
+    logo: 'https://example.com/logo.png',
+    foundingDate: '2024-01-01'
+  },
+
+  // Optional: Custom sections with any markdown content
+  sections: [
+    {
+      heading: 'About',
+      content: `This project helps teams collaborate more effectively by providing
+a unified workspace for tasks, projects, and documentation. It integrates with
+popular tools like GitHub, Slack, and Jira.`
+    },
+    {
+      heading: 'Key Features',
+      content: `- **Task Management**: Create, assign, and track tasks across teams
+- **Real-time Collaboration**: WebSocket-based updates for instant sync
+- **Rich API**: RESTful API with comprehensive documentation
+- **Integrations**: Connect with 50+ external services
+- **Security**: SOC 2 Type II certified with end-to-end encryption`
+    },
+    {
+      heading: 'Use Cases',
+      content: `**For Development Teams**: Track sprints, manage backlogs, and integrate with CI/CD pipelines.
+
+**For Content Teams**: Plan editorial calendars, collaborate on drafts, and publish content.
+
+**For Product Teams**: Gather feedback, prioritize features, and communicate roadmaps.`
+    }
+  ]
+};
+
+export default config;

package/package.json

@@ -1,6 +1,6 @@
 {
     "name": "@ontosdk/next",
-    "version": "1.2.0",
+    "version": "1.3.1",
     "description": "Extracts semantic Markdown from React/Next.js pages for AI Agents",
     "main": "dist/index.js",
     "module": "dist/index.mjs",
@@ -18,6 +18,21 @@
             "require": "./dist/middleware.js",
             "import": "./dist/middleware.mjs",
             "types": "./dist/middleware.d.ts"
+        },
+        "./components": {
+            "require": "./dist/OntoHead.js",
+            "import": "./dist/OntoHead.mjs",
+            "types": "./dist/OntoHead.d.ts"
+        },
+        "./provider": {
+            "require": "./dist/OntoProvider.js",
+            "import": "./dist/OntoProvider.mjs",
+            "types": "./dist/OntoProvider.d.ts"
+        },
+        "./config": {
+            "require": "./dist/config.js",
+            "import": "./dist/config.mjs",
+            "types": "./dist/config.d.ts"
         }
     },
     "bin": {
@@ -43,11 +58,22 @@
         "picocolors": "^1.0.0",
         "turndown": "^7.1.3"
     },
+    "peerDependencies": {
+        "next": ">=14.0.0",
+        "react": "^18.0.0 || ^19.0.0"
+    },
+    "peerDependenciesMeta": {
+        "react": {
+            "optional": true
+        }
+    },
     "devDependencies": {
         "@types/node": "^20.11.24",
+        "@types/react": "^19.2.14",
         "@types/turndown": "^5.0.4",
         "next": "^16.1.6",
+        "react": "^19.2.4",
         "tsup": "^8.0.2",
         "typescript": "^5.3.3"
     }
-}
+}

package/src/OntoHead.tsx

@@ -0,0 +1,59 @@
+'use client';
+
+import { usePathname } from 'next/navigation';
+
+/**
+ * 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>
+ *   );
+ * }
+ * ```
+ */
+export function OntoHead() {
+    const pathname = usePathname();
+
+    // Map current route to its .onto markdown payload path
+    let payloadPath = pathname;
+    if (payloadPath === '/' || payloadPath === '') {
+        payloadPath = '/index';
+    }
+    if (payloadPath.endsWith('/') && payloadPath !== '/') {
+        payloadPath = payloadPath.slice(0, -1);
+    }
+
+    const markdownHref = `/.onto${payloadPath}.md`;
+
+    return (
+        <>
+            {/* Per-page markdown alternate for AI agents */}
+            <link
+                rel="alternate"
+                type="text/markdown"
+                href={markdownHref}
+                title="AI-optimized Markdown version"
+            />
+            {/* Site-wide llms.txt manifest */}
+            <link
+                rel="alternate"
+                type="text/plain"
+                href="/llms.txt"
+                title="LLM-readable site manifest"
+            />
+        </>
+    );
+}