brave-real-browser-mcp-server 2.24.2 → 2.24.4

package/dist/handlers/advanced-tools.js

@@ -2367,18 +2367,19 @@ export async function handleCloudflareBypass(page, args) {
 }
 /**
  * Master tool: Extract direct stream/download URLs
+ * ULTRA POWERFUL: Handles packed JS, JW Player, Video.js, HLS.js, obfuscated scripts
  */
 export async function handleStreamExtractor(page, args) {
-    const formats = args.formats || ['mp4', 'mkv', 'm3u8', 'mp3', 'webm'];
+    const formats = args.formats || ['mp4', 'mkv', 'm3u8', 'mp3', 'webm', 'flv', 'avi'];
     const maxRedirects = args.maxRedirects || 10;
     const directUrls = [];
     // Navigate if URL provided
     if (args.url) {
         await page.goto(args.url, { waitUntil: 'domcontentloaded', timeout: 30000 });
     }
-    // Handle Cloudflare if enabled (inline logic)
+    // Handle Cloudflare if enabled
     if (args.bypassCloudflare) {
-        const cfPatterns = ['Checking your browser', 'Just a moment', 'cf-browser-verification'];
+        const cfPatterns = ['Checking your browser', 'Just a moment', 'cf-browser-verification', 'cf_chl_opt'];
         const isCloudflare = async () => {
             try {
                 const content = await page.content();
@@ -2388,72 +2389,212 @@ export async function handleStreamExtractor(page, args) {
                 return false;
             }
         };
-        // Wait up to 15 seconds for Cloudflare to pass
         const startCf = Date.now();
-        while (await isCloudflare() && Date.now() - startCf < 15000) {
+        while (await isCloudflare() && Date.now() - startCf < 20000) {
             await new Promise(r => setTimeout(r, 1000));
         }
     }
-    // Handle countdown if enabled (inline logic)
+    // Handle countdown if enabled
     if (args.waitForCountdown) {
-        const maxWait = 60;
+        const maxWait = 120;
         const startTime = Date.now();
         while ((Date.now() - startTime) / 1000 < maxWait) {
             const hasCountdown = await page.evaluate(() => {
                 const text = document.body?.innerText || '';
-                return /\d+\s*seconds?|wait\s*\d+|please\s*wait|countdown/gi.test(text);
+                return /\d+\s*seconds?|wait\s*\d+|please\s*wait|countdown|getting link/gi.test(text);
             });
             if (!hasCountdown)
                 break;
             await new Promise(r => setTimeout(r, 1000));
         }
     }
-    // Extract URLs from page
-    const extractedUrls = await page.evaluate((fmts) => {
+    // ULTRA POWERFUL: Extract from all sources
+    const extractedData = await page.evaluate((fmts) => {
         const urls = [];
-        const patterns = fmts.map(f => new RegExp(`https?://[^"'\\s]+\\.${f}([?#][^"'\\s]*)?`, 'gi'));
-        // Check page HTML
         const html = document.documentElement.innerHTML;
-        patterns.forEach(pattern => {
-            const matches = html.match(pattern);
-            if (matches)
-                urls.push(...matches);
+        // ============================================================
+        // 1. PACKED JS UNPACKING (p,a,c,k,e,d)
+        // ============================================================
+        const unpackPackedJS = (packed) => {
+            try {
+                // Find packed function pattern
+                const match = packed.match(/eval\(function\(p,a,c,k,e,[rd]\)\{[^}]+\}[^)]+\('[^']+'/);
+                if (!match)
+                    return '';
+                // Extract the encoded string and dictionary
+                const stringsMatch = packed.match(/'([^']+)'\.split\('\|'\)/);
+                if (!stringsMatch)
+                    return '';
+                const dict = stringsMatch[1].split('|');
+                let result = packed;
+                // Replace placeholders with actual values
+                for (let i = 0; i < dict.length; i++) {
+                    if (dict[i]) {
+                        const base36 = i.toString(36);
+                        result = result.replace(new RegExp(`\\b${base36}\\b`, 'g'), dict[i]);
+                    }
+                }
+                return result;
+            }
+            catch {
+                return '';
+            }
+        };
+        // Find and unpack all packed scripts
+        const scripts = document.querySelectorAll('script');
+        scripts.forEach(script => {
+            const content = script.textContent || '';
+            if (content.includes('eval(function(p,a,c,k,e,')) {
+                const unpacked = unpackPackedJS(content);
+                // Extract URLs from unpacked content
+                fmts.forEach(fmt => {
+                    const regex = new RegExp(`https?://[^"'\\s]+\\.${fmt}[^"'\\s]*`, 'gi');
+                    const matches = unpacked.match(regex);
+                    if (matches)
+                        matches.forEach(url => urls.push({ url, source: 'packed_js' }));
+                });
+            }
         });
-        // Check video/audio sources
-        document.querySelectorAll('video source, audio source, video, audio').forEach(el => {
-            const src = el.getAttribute('src');
-            if (src && fmts.some(f => src.includes(`.${f}`))) {
-                urls.push(src);
+        // ============================================================
+        // 2. JW PLAYER DETECTION
+        // ============================================================
+        if (window.jwplayer) {
+            try {
+                const player = window.jwplayer();
+                if (player && player.getPlaylistItem) {
+                    const item = player.getPlaylistItem();
+                    if (item) {
+                        if (item.file)
+                            urls.push({ url: item.file, source: 'jwplayer' });
+                        if (item.sources) {
+                            item.sources.forEach((s) => {
+                                if (s.file)
+                                    urls.push({ url: s.file, source: 'jwplayer' });
+                            });
+                        }
+                    }
+                }
+            }
+            catch { /* ignore */ }
+        }
+        // JW Player setup patterns in scripts
+        const jwPatterns = [
+            /file:\s*["']([^"']+\.m3u8[^"']*?)["']/gi,
+            /file:\s*["']([^"']+\.mp4[^"']*?)["']/gi,
+            /sources:\s*\[\s*\{[^}]*file:\s*["']([^"']+)["']/gi,
+            /setup\([^)]*file:\s*["']([^"']+)["']/gi,
+        ];
+        jwPatterns.forEach(pattern => {
+            let match;
+            while ((match = pattern.exec(html)) !== null) {
+                urls.push({ url: match[1], source: 'jwplayer_setup' });
+            }
+        });
+        // ============================================================
+        // 3. VIDEO.JS DETECTION
+        // ============================================================
+        const videoJsPlayers = document.querySelectorAll('.video-js, [data-setup], video[id^="vjs"]');
+        videoJsPlayers.forEach(player => {
+            const video = player.querySelector('source') || player;
+            const src = video.getAttribute('src') || player.src;
+            if (src)
+                urls.push({ url: src, source: 'videojs' });
+        });
+        // ============================================================
+        // 4. HLS.JS DETECTION
+        // ============================================================
+        const hlsPatterns = [
+            /hls\.loadSource\(["']([^"']+)["']\)/gi,
+            /Hls\.loadSource\(["']([^"']+)["']\)/gi,
+            /source:\s*["']([^"']+\.m3u8[^"']*)["']/gi,
+            /src:\s*["']([^"']+\.m3u8[^"']*)["']/gi,
+        ];
+        hlsPatterns.forEach(pattern => {
+            let match;
+            while ((match = pattern.exec(html)) !== null) {
+                urls.push({ url: match[1], source: 'hlsjs' });
             }
         });
-        // Check links
+        // ============================================================
+        // 5. PLYR DETECTION
+        // ============================================================
+        if (window.Plyr) {
+            try {
+                const plyrPlayer = window.player;
+                if (plyrPlayer && plyrPlayer.source) {
+                    urls.push({ url: plyrPlayer.source, source: 'plyr' });
+                }
+            }
+            catch { /* ignore */ }
+        }
+        // ============================================================
+        // 6. DATA ATTRIBUTES
+        // ============================================================
+        document.querySelectorAll('[data-src], [data-video], [data-file], [data-stream]').forEach(el => {
+            const attrs = ['data-src', 'data-video', 'data-file', 'data-stream', 'data-link'];
+            attrs.forEach(attr => {
+                const val = el.getAttribute(attr);
+                if (val && fmts.some(f => val.includes(`.${f}`))) {
+                    urls.push({ url: val, source: 'data_attr' });
+                }
+            });
+        });
+        // ============================================================
+        // 7. STANDARD VIDEO/AUDIO ELEMENTS
+        // ============================================================
+        document.querySelectorAll('video, audio, source').forEach(el => {
+            const src = el.getAttribute('src');
+            if (src)
+                urls.push({ url: src, source: 'html_media' });
+        });
+        // ============================================================
+        // 8. DIRECT LINKS
+        // ============================================================
         document.querySelectorAll('a[href]').forEach(el => {
             const href = el.href;
             if (href && fmts.some(f => href.includes(`.${f}`))) {
-                urls.push(href);
+                urls.push({ url: href, source: 'direct_link' });
             }
         });
-        // Check iframes for embedded players
+        // ============================================================
+        // 9. IFRAME PLAYERS
+        // ============================================================
         document.querySelectorAll('iframe').forEach(iframe => {
             const src = iframe.src;
-            if (src && (src.includes('player') || src.includes('embed'))) {
-                urls.push(`iframe:${src}`);
+            if (src && (src.includes('player') || src.includes('embed') || src.includes('video'))) {
+                urls.push({ url: `iframe:${src}`, source: 'iframe' });
             }
         });
-        return [...new Set(urls)];
+        // ============================================================
+        // 10. REGEX SCAN OF ENTIRE HTML
+        // ============================================================
+        fmts.forEach(fmt => {
+            const pattern = new RegExp(`https?://[^"'\\s<>]+\\.${fmt}[^"'\\s<>]*`, 'gi');
+            const matches = html.match(pattern);
+            if (matches)
+                matches.forEach(url => urls.push({ url, source: 'regex_scan' }));
+        });
+        // Deduplicate
+        const seen = new Set();
+        return urls.filter(u => {
+            if (seen.has(u.url))
+                return false;
+            seen.add(u.url);
+            return true;
+        });
     }, formats);
     // Process found URLs
-    for (const url of extractedUrls) {
-        const format = formats.find(f => url.includes(`.${f}`)) || 'unknown';
+    for (const item of extractedData) {
+        const format = formats.find(f => item.url.includes(`.${f}`)) || 'unknown';
         directUrls.push({
-            url,
+            url: item.url,
             format,
             quality: args.quality || 'auto',
+            source: item.source,
         });
     }
     // Check network requests for media URLs
     const networkUrls = await page.evaluate((fmts) => {
-        // Check performance entries for loaded resources
         const resources = performance.getEntriesByType('resource');
         return resources
             .filter(r => fmts.some(f => r.name.includes(`.${f}`)))
@@ -2462,14 +2603,14 @@ export async function handleStreamExtractor(page, args) {
     for (const url of networkUrls) {
         if (!directUrls.some(d => d.url === url)) {
             const format = formats.find(f => url.includes(`.${f}`)) || 'unknown';
-            directUrls.push({ url, format });
+            directUrls.push({ url, format, source: 'network' });
         }
     }
     return {
         success: directUrls.length > 0,
         directUrls,
         message: directUrls.length > 0
-            ? `Found ${directUrls.length} direct URL(s)`
+            ? `🎬 Found ${directUrls.length} direct URL(s) from ${new Set(directUrls.map(d => d.source)).size} sources`
             : 'No direct URLs found',
     };
 }

package/dist/handlers/content-handlers.js

@@ -80,6 +80,40 @@ export async function handleFindSelector(args) {
             // Ensure elementType has a fallback value
             const elementType = args?.elementType || '*';
             tracker.setProgress(10, '🔧 Preparing search strategies...');
+            // ============================================================
+            // FUZZY MATCHING: Levenshtein distance for typo tolerance
+            // ============================================================
+            const fuzzyMatch = (str1, str2, threshold = 0.7) => {
+                const s1 = str1.toLowerCase();
+                const s2 = str2.toLowerCase();
+                // Exact match
+                if (s1 === s2)
+                    return { match: true, score: 1 };
+                // Contains match
+                if (s1.includes(s2) || s2.includes(s1))
+                    return { match: true, score: 0.9 };
+                // Levenshtein distance
+                const len1 = s1.length;
+                const len2 = s2.length;
+                const matrix = [];
+                for (let i = 0; i <= len1; i++)
+                    matrix[i] = [i];
+                for (let j = 0; j <= len2; j++)
+                    matrix[0][j] = j;
+                for (let i = 1; i <= len1; i++) {
+                    for (let j = 1; j <= len2; j++) {
+                        const cost = s1[i - 1] === s2[j - 1] ? 0 : 1;
+                        matrix[i][j] = Math.min(matrix[i - 1][j] + 1, // deletion
+                        matrix[i][j - 1] + 1, // insertion
+                        matrix[i - 1][j - 1] + cost // substitution
+                        );
+                    }
+                }
+                const distance = matrix[len1][len2];
+                const maxLen = Math.max(len1, len2);
+                const score = 1 - (distance / maxLen);
+                return { match: score >= threshold, score };
+            };
             // Helper: Search in Shadow DOM
             const searchInShadowDOM = async (sel) => {
                 return await pageInstance.evaluate((selector) => {

package/dist/index.js

@@ -57,9 +57,7 @@ import { handleClick, handleType, handleSolveCaptcha, handleRandomScroll } from
 import { handleGetContent, handleFindSelector } from './handlers/content-handlers.js';
 import { handleSaveContentAsMarkdown } from './handlers/file-handlers.js';
 // Import advanced tools handlers
-import { handleBreadcrumbNavigator, handleUrlRedirectTracer, handleSearchContent, handleExtractJson, handleScrapeMetaTags, handlePressKey, handleProgressTracker, handleDeepAnalysis, handleNetworkRecorder, handleApiFinder, handleAjaxContentWaiter, handleMediaExtractor, handleElementScreenshot, handleLinkHarvester, handleBatchElementScraper, handleExtractSchema, 
-// Streaming tools
-handleM3u8Parser, handleCookieManager, 
+import { handleBreadcrumbNavigator, handleUrlRedirectTracer, handleSearchContent, handleExtractJson, handleScrapeMetaTags, handlePressKey, handleProgressTracker, handleDeepAnalysis, handleNetworkRecorder, handleApiFinder, handleAjaxContentWaiter, handleElementScreenshot, handleLinkHarvester, handleBatchElementScraper, handleExtractSchema, handleCookieManager, 
 // Download tools
 handleFileDownloader, 
 // Enhanced streaming/download tools
@@ -220,10 +218,7 @@ server.setRequestHandler(CallToolRequestSchema, async (request) => {
                 if (!page)
                     throw new Error('Browser not initialized. Call browser_init first.');
                 return { content: [{ type: 'text', text: JSON.stringify(await handleAjaxContentWaiter(page, args || {})) }] };
-            case TOOL_NAMES.MEDIA_EXTRACTOR:
-                if (!page)
-                    throw new Error('Browser not initialized. Call browser_init first.');
-                return { content: [{ type: 'text', text: JSON.stringify(await handleMediaExtractor(page, args || {})) }] };
+            // MEDIA_EXTRACTOR case REMOVED - merged into STREAM_EXTRACTOR
             case TOOL_NAMES.ELEMENT_SCREENSHOT:
                 if (!page)
                     throw new Error('Browser not initialized. Call browser_init first.');
@@ -240,11 +235,7 @@ server.setRequestHandler(CallToolRequestSchema, async (request) => {
                 if (!page)
                     throw new Error('Browser not initialized. Call browser_init first.');
                 return { content: [{ type: 'text', text: JSON.stringify(await handleExtractSchema(page, args || {})) }] };
-            // Streaming & Media Tools
-            case TOOL_NAMES.M3U8_PARSER:
-                if (!page)
-                    throw new Error('Browser not initialized. Call browser_init first.');
-                return { content: [{ type: 'text', text: JSON.stringify(await handleM3u8Parser(page, args || {})) }] };
+            // M3U8_PARSER case REMOVED - merged into STREAM_EXTRACTOR
             case TOOL_NAMES.COOKIE_MANAGER:
                 if (!page)
                     throw new Error('Browser not initialized. Call browser_init first.');

package/dist/tool-definitions.js

@@ -485,21 +485,7 @@ export const TOOLS = [
             },
         },
     },
-    {
-        name: 'media_extractor',
-        description: 'Extract media (audio/video) from page with quality options and ad-bypass',
-        inputSchema: {
-            type: 'object',
-            additionalProperties: false,
-            properties: {
-                mediaType: { type: 'string', enum: ['video', 'audio', 'all'], description: 'Type of media to extract', default: 'all' },
-                includeEmbedded: { type: 'boolean', description: 'Include embedded iframes', default: true },
-                quality: { type: 'string', description: 'Preferred quality (highest, lowest, 1080p, 720p)' },
-                format: { type: 'string', description: 'Preferred format (mp4, webm, m3u8)' },
-                bypassAds: { type: 'boolean', description: 'Attempt to bypass video ads', default: false },
-            },
-        },
-    },
+    // media_extractor REMOVED - functionality merged into stream_extractor
     {
         name: 'element_screenshot',
         description: 'Capture screenshot of a specific element',
@@ -554,23 +540,7 @@ export const TOOLS = [
             },
         },
     },
-    // ============================================================
-    // STREAMING & MEDIA TOOLS (3 new tools)
-    // ============================================================
-    {
-        name: 'm3u8_parser',
-        description: 'Parse and extract HLS/m3u8 streaming URLs with quality options',
-        inputSchema: {
-            type: 'object',
-            additionalProperties: false,
-            properties: {
-                url: { type: 'string', description: 'URL of the page or m3u8 file' },
-                extractAll: { type: 'boolean', description: 'Extract all quality variants', default: true },
-                preferQuality: { type: 'string', description: 'Preferred quality (1080p, 720p, 480p, best, worst)', default: 'best' },
-                includeAudio: { type: 'boolean', description: 'Include audio-only streams', default: true },
-            },
-        },
-    },
+    // m3u8_parser REMOVED - functionality merged into stream_extractor
     {
         name: 'cookie_manager',
         description: 'Manage browser cookies for premium accounts and sessions',
@@ -678,12 +648,12 @@ export const TOOL_NAMES = {
     NETWORK_RECORDER: 'network_recorder',
     API_FINDER: 'api_finder',
     AJAX_CONTENT_WAITER: 'ajax_content_waiter',
-    MEDIA_EXTRACTOR: 'media_extractor',
+    // MEDIA_EXTRACTOR: 'media_extractor', // REMOVED - merged into STREAM_EXTRACTOR
     ELEMENT_SCREENSHOT: 'element_screenshot',
     LINK_HARVESTER: 'link_harvester',
     BATCH_ELEMENT_SCRAPER: 'batch_element_scraper',
     EXTRACT_SCHEMA: 'extract_schema',
-    M3U8_PARSER: 'm3u8_parser',
+    // M3U8_PARSER: 'm3u8_parser', // REMOVED - merged into STREAM_EXTRACTOR
     COOKIE_MANAGER: 'cookie_manager',
     FILE_DOWNLOADER: 'file_downloader',
     // Enhanced tools

package/dist/unified-server.js

@@ -132,11 +132,7 @@ mcpServer.setRequestHandler(CallToolRequestSchema, async (request) => {
                     throw new Error('Browser not initialized');
                 result = { content: [{ type: 'text', text: JSON.stringify(await advancedTools.handleSearchContent(page, args)) }] };
                 break;
-            case TOOL_NAMES.MEDIA_EXTRACTOR:
-                if (!page)
-                    throw new Error('Browser not initialized');
-                result = { content: [{ type: 'text', text: JSON.stringify(await advancedTools.handleMediaExtractor(page, args || {})) }] };
-                break;
+            // MEDIA_EXTRACTOR case REMOVED - merged into STREAM_EXTRACTOR
             case TOOL_NAMES.STREAM_EXTRACTOR:
                 if (!page)
                     throw new Error('Browser not initialized');
@@ -513,10 +509,9 @@ async function main() {
             console.error('   📜 random_scroll         - Natural scrolling');
             console.error('   🤖 solve_captcha         - Solve CAPTCHAs');
             console.error('');
-            console.error('   Media Extraction:');
-            console.error('   🎬 media_extractor       - Extract video/audio');
-            console.error('   📺 m3u8_parser           - Parse HLS streams');
-            console.error('   🎥 stream_extractor      - Master stream extraction');
+            console.error('   Media & Streaming:');
+            console.error('   🎬 stream_extractor      - Master: Extract video/audio/m3u8/mp4');
+            console.error('   🖼️  iframe_handler        - Handle nested iframes (deep_scrape)');
             console.error('');
             console.error('   Advanced Tools:');
             console.error('   🔎 search_content        - Search patterns in page');

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "brave-real-browser-mcp-server",
-  "version": "2.24.2",
+  "version": "2.24.4",
   "description": "🦁 MCP server for Brave Real Browser - NPM Workspaces Monorepo with anti-detection features, SSE streaming, and LSP compatibility",
   "type": "module",
   "main": "dist/index.js",
@@ -50,7 +50,7 @@
   "dependencies": {
     "@modelcontextprotocol/sdk": "latest",
     "@types/turndown": "latest",
-    "brave-real-browser": "^2.5.2",
+    "brave-real-browser": "^2.5.4",
     "turndown": "latest",
     "vscode-languageserver": "^9.0.1",
     "vscode-languageserver-textdocument": "^1.0.12"