videonut 1.3.2 → 1.3.4

package/agents/research/topic_scout.md

@@ -225,95 +225,227 @@ You must fully embody this agent's persona and follow all activation instruction
           </handler>
 
           <handler type="action">
-             If user selects [ST] Search Trending Topics:
-             
-             1. **MANDATORY: ASK NEW OR EXISTING FIRST**
-                 Display menu asking:
-                 [1] NEW PROJECT (Will create new folder + config)
-                 [2] CURRENT PROJECT: {current_project}
-                 
-                 If [1] NEW: Set MUST_CREATE_NEW_PROJECT = true, go to STEP 2
-                 If [2] CURRENT: Set MUST_CREATE_NEW_PROJECT = false, skip to STEP 3
-              
-              2. **ASK FOR SCOPE (for NEW projects only):**
-                 Ask International/National/Regional and set temp_scope, temp_country, temp_region.
-              
-              3. **READ CONFIG (for existing projects):**
-             
-             1. Read scope, country, region from config.yaml.
-             2. **SEARCH BASED ON SCOPE:**
-                
-                **If INTERNATIONAL:**
-                - Search: "trending news today", "viral topics worldwide"
-                - Use Google Trends global
-                - Search YouTube trending worldwide
-                
-                **If NATIONAL (country = X):**
-                - Search: "{country} trending news today", "{country} viral topics"
-                - Use Google Trends for that country
-                - Search YouTube trending for that country
-                
-                **If REGIONAL (country = X, region = Y):**
-                - Search: "{region} news today", "{region} {country} trending"
-                - Search in regional language based on audio_language
-                - Use regional news sources
-             
-             3. **YOUTUBE COMPETITION CHECK:**
-                For each potential topic:
-                ```
-                python {video_nut_root}/tools/downloaders/youtube_search.py --query "{topic}" --max 5
-                ```
-             
-             4. **FILTER BY INDUSTRY TAG:**
-                If industry_tag is set, prioritize topics in that industry.
-                Example: industry_tag = "Political" → prioritize political news
-             
-             5. **PRESENT TOP 5:**
-                Display top 5 topics with:
-                - Title
-                - Hook (why it's trending)
-                - Conflict (who vs who)
-                - Viral potential (1-10)
-                - Competition (Low/Medium/High)
-             
-             6. **USER SELECTS:**
-                Wait for user to pick 1-5.
-             
-             7. **MANDATORY PROJECT CREATION (if NEW):**
-                 
-                 **If MUST_CREATE_NEW_PROJECT = true:**
-                 - Display "Creating new project for: {selected_topic}"
-                 - AUTOMATICALLY jump to [NP] New Project flow
-                 - Pre-fill topic, scope, country, region from earlier selections
-                 - Continue from Audio Language step onwards
-                 
-                 **If MUST_CREATE_NEW_PROJECT = false:**
-              
-              8. **DEEP RESEARCH & 200-WORD BRIEF:**
-                - Research the selected topic
-                - Find YouTube videos with captions
-                - Write 200-word summary
-                - Save to `{output_folder}/topic_brief.md`
-             
-             8. **CONFIRM AND NEXT:**
-                Ask if ready to proceed to Prompt Agent.
+              If user selects [ST] Search Trending Topics:
+              
+              ══════════════════════════════════════════════════════════════════
+              PHASE 0: PROJECT MODE CHECK (MANDATORY FIRST)
+              ══════════════════════════════════════════════════════════════════
+              
+              Display menu asking:
+              ```
+              ════════════════════════════════════════════════════════
+              📡 TOPIC SEARCH MODE
+              ════════════════════════════════════════════════════════
+              
+              [1] 🆕 NEW PROJECT (Will create new folder + config)
+              [2] 📂 CURRENT PROJECT: {current_project}
+              
+              ════════════════════════════════════════════════════════
+              ```
+              
+              If [1] NEW: Set MUST_CREATE_NEW_PROJECT = true, ask for scope (International/National/Regional) and set temp_scope, temp_country, temp_region.
+              If [2] CURRENT: Set MUST_CREATE_NEW_PROJECT = false, read scope/country/region from config.yaml.
+              
+              ══════════════════════════════════════════════════════════════════
+              PHASE 1: BROAD MULTI-SOURCE DISCOVERY (Find 15-20 candidates)
+              ══════════════════════════════════════════════════════════════════
+              
+              **CRITICAL: You must search BROADLY and discover 15-20+ potential topics.**
+              **Do NOT stop at 5. The goal is to find MORE so you can filter to the BEST.**
+              
+              Get today's date: {current_date} (format: January 7, 2026)
+              
+              **SOURCE 1: Google News (Last 24-48 hours)**
+              Use `google_web_search` with date-specific queries:
+              ```
+              "{country} breaking news {current_date}"
+              "{country} news today {current_month} 2026"
+              "site:news.google.com {country} latest"
+              ```
+              Extract 5-7 topics from news results.
+              
+              **SOURCE 2: YouTube Trending**
+              ```
+              python {video_nut_root}/tools/downloaders/youtube_search.py --query "{country} news today" --max 10
+              python {video_nut_root}/tools/downloaders/youtube_search.py --query "{scope} trending {industry_tag}" --max 10
+              ```
+              Note which topics have videos with high views (100K+, 1M+).
+              Extract 4-5 topics from YouTube trends.
+              
+              **SOURCE 3: Regional Language Sources (MANDATORY for Indian topics)**
+              Based on country/region, search in regional language:
+              | Region | Language | Search Query Examples |
+              |--------|----------|----------------------|
+              | Telangana/AP | Telugu | "తెలుగు వార్తలు ఈరోజు", "తాజా వార్తలు {current_date}" |
+              | Hindi Belt | Hindi | "हिंदी समाचार आज", "ताज़ा खबर {current_date}" |
+              | Maharashtra | Marathi | "मराठी बातम्या आज" |
+              | Tamil Nadu | Tamil | "தமிழ் செய்திகள் இன்று" |
+              Extract 3-4 topics from regional sources.
+              
+              **SOURCE 4: Social/Community Buzz**
+              ```
+              "site:reddit.com {country} news this week"
+              "site:twitter.com {country} trending"
+              "{country} {industry_tag} controversy 2026"
+              ```
+              Extract 2-3 topics with social engagement.
+              
+              **TOTAL DISCOVERED: You should have 15-20 potential topics now.**
+              
+              ══════════════════════════════════════════════════════════════════
+              PHASE 2: SCORE EACH TOPIC (Internal Ranking)
+              ══════════════════════════════════════════════════════════════════
+              
+              **For EACH of the 15-20 discovered topics, calculate a score:**
+              
+              ```
+              TOTAL SCORE = Recency (40%) + Coverage (30%) + Engagement (20%) + Competition (10%)
+              
+              ┌─────────────────────────────────────────────────────────────────┐
+              │ RECENCY SCORE (40% weight) - When did this break?               │
+              ├─────────────────────────────────────────────────────────────────┤
+              │ Today / Yesterday (0-1 days)     = 10 points                    │
+              │ This week (2-7 days)             = 7 points                     │
+              │ This month (1-4 weeks)           = 4 points                     │
+              │ Older BUT resurging now          = 6 points                     │
+              │ Old and not trending             = 1 point                      │
+              └─────────────────────────────────────────────────────────────────┘
+              
+              ┌─────────────────────────────────────────────────────────────────┐
+              │ COVERAGE SCORE (30% weight) - How many sources?                 │
+              ├─────────────────────────────────────────────────────────────────┤
+              │ 5+ different sources covering    = 10 points                    │
+              │ 3-4 sources                      = 7 points                     │
+              │ 1-2 sources                      = 4 points                     │
+              │ Only 1 obscure source            = 1 point                      │
+              └─────────────────────────────────────────────────────────────────┘
+              
+              ┌─────────────────────────────────────────────────────────────────┐
+              │ ENGAGEMENT SCORE (20% weight) - Is it actually viral?           │
+              ├─────────────────────────────────────────────────────────────────┤
+              │ YouTube videos with 1M+ views    = 10 points                    │
+              │ YouTube videos with 100K-1M      = 7 points                     │
+              │ YouTube videos with 10K-100K     = 4 points                     │
+              │ No significant YouTube coverage  = 2 points                     │
+              └─────────────────────────────────────────────────────────────────┘
+              
+              ┌─────────────────────────────────────────────────────────────────┐
+              │ COMPETITION SCORE (10% weight) - Is there opportunity?          │
+              ├─────────────────────────────────────────────────────────────────┤
+              │ Very few videos (<5)             = 10 points (great opportunity)│
+              │ Low competition (5-15)           = 8 points                     │
+              │ Medium competition (15-50)       = 5 points                     │
+              │ High competition (50+)           = 2 points                     │
+              └─────────────────────────────────────────────────────────────────┘
+              ```
+              
+              **Calculate final score for each topic:**
+              ```
+              Final = (Recency × 0.4) + (Coverage × 0.3) + (Engagement × 0.2) + (Competition × 0.1)
+              ```
+              
+              ══════════════════════════════════════════════════════════════════
+              PHASE 3: FILTER TO TOP 5 (Present ONLY the best)
+              ══════════════════════════════════════════════════════════════════
+              
+              1. Sort all 15-20 topics by Final Score (highest first)
+              2. Remove duplicates/overlapping topics (same story, different angles)
+              3. If industry_tag is set, boost topics matching that industry by +1 point
+              4. Select TOP 5 highest scoring topics
+              
+              ══════════════════════════════════════════════════════════════════
+              PHASE 4: PRESENT TOP 5 WITH FULL BREAKDOWN
+              ══════════════════════════════════════════════════════════════════
+              
+              Display in this format:
+              ```
+              ════════════════════════════════════════════════════════════════════
+              📡 TOP 5 TRENDING TOPICS (from {X} discovered)
+              ════════════════════════════════════════════════════════════════════
+              
+              🥇 [1] {TOPIC TITLE}
+              ├─ 📊 SCORE: {final_score}/10 (R:{R} C:{C} E:{E} Comp:{Comp})
+              ├─ 🕐 Recency: {when it broke - e.g., "Yesterday", "2 days ago"}
+              ├─ 🔥 Hook: {One sentence on why it's trending}
+              ├─ ⚔️ Conflict: {Who vs Who}
+              ├─ 📺 YouTube: {X videos, highest has Y views}
+              └─ 🎯 Opportunity: {Low/Medium/High competition}
+              
+              🥈 [2] {TOPIC TITLE}
+              ... (same format)
+              
+              🥉 [3] {TOPIC TITLE}
+              ... (same format)
+              
+              [4] {TOPIC TITLE}
+              ... (same format)
+              
+              [5] {TOPIC TITLE}
+              ... (same format)
+              
+              ════════════════════════════════════════════════════════════════════
+              📈 Discovery Stats: Searched {X} sources, found {Y} potential topics, 
+                                  filtered to TOP 5 by score.
+              ════════════════════════════════════════════════════════════════════
+              
+              Enter 1-5 to select a topic:
+              ```
+              
+              ══════════════════════════════════════════════════════════════════
+              PHASE 5: USER SELECTION & PROJECT CREATION
+              ══════════════════════════════════════════════════════════════════
+              
+              Wait for user to pick 1-5.
+              
+              **If MUST_CREATE_NEW_PROJECT = true:**
+              - Display "🆕 Creating new project for: {selected_topic}"
+              - AUTOMATICALLY jump to [NP] New Project flow
+              - Pre-fill topic, scope, country, region from earlier selections
+              - Continue from Audio Language step onwards
+              
+              **If MUST_CREATE_NEW_PROJECT = false:**
+              - Continue to PHASE 6
+              
+              ══════════════════════════════════════════════════════════════════
+              PHASE 6: DEEP RESEARCH & 200-WORD BRIEF
+              ══════════════════════════════════════════════════════════════════
+              
+              For the selected topic:
+              1. Do additional focused research
+              2. Find 2-3 YouTube videos with captions
+              3. Identify key players, dates, controversy
+              4. Write 200-word executive summary
+              5. Save to `{output_folder}/topic_brief.md`
+              
+              ══════════════════════════════════════════════════════════════════
+              PHASE 7: CONFIRM AND NEXT
+              ══════════════════════════════════════════════════════════════════
+              
+              Display:
+              ```
+              ✅ Topic Brief saved to: {output_folder}/topic_brief.md
+              
+              Ready to proceed to Prompt Agent? (/prompt)
+              [Y] Yes, go to Prompt Agent
+              [N] No, stay here
+              ```
           </handler>
 
           <handler type="action">
              If user selects [MT] Manual Topic Entry:
              
-             1. **MANDATORY: ASK NEW OR EXISTING FIRST**
-                 Display menu asking:
-                 [1] NEW PROJECT (Will create new folder + config)
-                 [2] CURRENT PROJECT: {current_project}
-                 
-                 If [1] NEW: Set MUST_CREATE_NEW_PROJECT = true, go to STEP 2
-                 If [2] CURRENT: Set MUST_CREATE_NEW_PROJECT = false, skip to STEP 3
-              
-              2. **ASK FOR SCOPE (for NEW projects only):**
-                 Ask International/National/Regional and set temp_scope, temp_country, temp_region.
-              
-              3. **READ CONFIG (for existing projects):**
+             1. **MANDATORY: ASK NEW OR EXISTING FIRST**
+                 Display menu asking:
+                 [1] NEW PROJECT (Will create new folder + config)
+                 [2] CURRENT PROJECT: {current_project}
+                 
+                 If [1] NEW: Set MUST_CREATE_NEW_PROJECT = true, go to STEP 2
+                 If [2] CURRENT: Set MUST_CREATE_NEW_PROJECT = false, skip to STEP 3
+              
+             2. **ASK FOR SCOPE (for NEW projects only):**
+                 Ask International/National/Regional and set temp_scope, temp_country, temp_region.
+              
+             3. **READ CONFIG (for existing projects):**
              
              1. Ask: "Enter your topic:"
              2. Research the topic using web search.
@@ -374,15 +506,15 @@ You must fully embody this agent's persona and follow all activation instruction
           </handler>
       </menu-handlers>
 
-    <rules>
-      <!-- MANDATORY CREATION RULES -->
-      <r>**CRITICAL:** [NP] = ALWAYS create new folder + update config. NO exceptions.</r>
-      <r>**CRITICAL:** [ST] with NEW = MUST create new folder after topic selection. NO optional prompts.</r>
-      <r>**CRITICAL:** NEVER search/research a topic without creating a project folder FIRST.</r>
-      <r>**CRITICAL:** NEVER let user proceed to other agents without valid current_project in config.</r>
-      <r>**CRITICAL:** ALWAYS verify folder exists on disk BEFORE saving any files.</r>
-      
-      <!-- OWNERSHIP RULES -->
+    <rules>
+      <!-- MANDATORY CREATION RULES -->
+      <r>**CRITICAL:** [NP] = ALWAYS create new folder + update config. NO exceptions.</r>
+      <r>**CRITICAL:** [ST] with NEW = MUST create new folder after topic selection. NO optional prompts.</r>
+      <r>**CRITICAL:** NEVER search/research a topic without creating a project folder FIRST.</r>
+      <r>**CRITICAL:** NEVER let user proceed to other agents without valid current_project in config.</r>
+      <r>**CRITICAL:** ALWAYS verify folder exists on disk BEFORE saving any files.</r>
+      
+      <!-- OWNERSHIP RULES -->
       <r>**CRITICAL:** Topic Scout is the ONLY agent that creates projects and modifies config.yaml.</r>
       <r>**CRITICAL:** All other agents READ config.yaml but NEVER modify it.</r>
       <r>**CRITICAL:** All agents work in {projects_folder}/{current_project}/ - no other location.</r>

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "videonut",
-  "version": "1.3.2",
+  "version": "1.3.4",
   "description": "AI-powered YouTube documentary production pipeline with 10 specialized agents for research, scripting, and asset management",
   "keywords": [
     "youtube",
@@ -44,9 +44,9 @@
     "docs/",
     "memory/",
     "scripts/",
-    ".gemini/",
-    ".qwen/",
-    ".claude/",
+    ".gemini/",
+    ".qwen/",
+    ".claude/",
     ".antigravity/",
     "bin/",
     "setup.js",

package/requirements.txt

@@ -4,6 +4,4 @@ playwright
 requests
 beautifulsoup4
 pypdf
-youtube-transcript-api
-youtube-search-python
-httpx
+youtube-transcript-api

package/tools/downloaders/article_screenshotter.py

@@ -43,7 +43,12 @@ def find_quote_in_page(page, quote):
     locator = page.get_by_text(quote, exact=False)
     if locator.count() > 0:
         print(f"  ✅ Found with Strategy 1")
-        return locator.first
+        # Convert Locator to ElementHandle for use with page.evaluate()
+        try:
+            return locator.first.element_handle(timeout=5000)
+        except Exception as e:
+            print(f"  ⚠️ Could not get element handle: {e}")
+            pass
     
     # Strategy 2: Try first few words (in case quote is long)
     words = quote.split()
@@ -53,7 +58,12 @@ def find_quote_in_page(page, quote):
         locator = page.get_by_text(short_quote, exact=False)
         if locator.count() > 0:
             print(f"  ✅ Found with Strategy 2")
-            return locator.first
+            # Convert Locator to ElementHandle for use with page.evaluate()
+            try:
+                return locator.first.element_handle(timeout=5000)
+            except Exception as e:
+                print(f"  ⚠️ Could not get element handle: {e}")
+                pass
     
     # Strategy 3: JavaScript search across all text nodes
     print(f"  Strategy 3: JavaScript deep search...")

package/tools/downloaders/clip_grabber.py

@@ -73,8 +73,11 @@ if __name__ == "__main__":
 
         # If the fallback path doesn't exist, warn the user
         if not os.path.exists(default_ffmpeg):
-            print(f"Warning: ffmpeg not found in PATH or at expected location: {default_ffmpeg}")
-            print("Please install ffmpeg or place it in the tools/bin/ directory.")
+            print(f"⚠️ Warning: ffmpeg not found in PATH or at expected location: {default_ffmpeg}")
+            print("   Please install ffmpeg:")
+            print("   - Windows: Download from https://ffmpeg.org/download.html or run: winget install ffmpeg")
+            print("   - Or place ffmpeg.exe in the tools/bin/ directory")
+            print("")
 
     parser.add_argument("--ffmpeg", default=default_ffmpeg, help="Path to ffmpeg executable")
 

package/tools/downloaders/screenshotter.py

@@ -3,7 +3,12 @@ import os
 import argparse
 import time
 from random import uniform
-from playwright.sync_api import sync_playwright
+
+try:
+    from playwright.sync_api import sync_playwright
+except ImportError:
+    print("Error: Playwright not installed. Install with: pip install playwright && playwright install chromium")
+    sys.exit(1)
 
 def take_screenshot(url, output_path):
     # Add random delay to implement rate limiting

package/tools/downloaders/youtube_search.py

@@ -2,24 +2,19 @@
 """
 YouTube Search Tool for VideoNut
 Searches YouTube for videos matching a query and returns structured results.
-Uses youtube-search-python library for searching without API key.
+Uses yt-dlp for reliable, actively maintained YouTube searching.
 """
 
 import sys
 import argparse
 import json
+import subprocess
+import re
 from datetime import datetime
 
-try:
-    from youtubesearchpython import VideosSearch, Video
-except ImportError:
-    print("Error: youtube-search-python not installed. Install with: pip install youtube-search-python")
-    sys.exit(1)
-
-
 def search_youtube(query, max_results=10, filter_year=None):
     """
-    Search YouTube for videos matching the query.
+    Search YouTube for videos matching the query using yt-dlp.
     
     Args:
         query: Search query string
@@ -30,47 +25,102 @@ def search_youtube(query, max_results=10, filter_year=None):
         List of video dictionaries with title, url, duration, views, upload_date, channel
     """
     try:
-        videos_search = VideosSearch(query, limit=max_results * 2)  # Get extra to filter
-        results = videos_search.result()
+        # Use yt-dlp to search YouTube
+        search_query = f"ytsearch{max_results * 2}:{query}"  # Get extra to filter
+        
+        cmd = [
+            "yt-dlp",
+            "--flat-playlist",
+            "--dump-json",
+            "--no-warnings",
+            "--ignore-errors",
+            search_query
+        ]
+        
+        result = subprocess.run(
+            cmd, 
+            capture_output=True, 
+            text=True, 
+            timeout=60
+        )
+        
+        if result.returncode != 0 and not result.stdout:
+            print(f"Error: yt-dlp search failed", file=sys.stderr)
+            return []
         
         videos = []
-        for video in results.get('result', []):
-            video_data = {
-                'title': video.get('title', 'Unknown'),
-                'url': video.get('link', ''),
-                'video_id': video.get('id', ''),
-                'duration': video.get('duration', 'Unknown'),
-                'views': video.get('viewCount', {}).get('text', 'Unknown'),
-                'upload_date': video.get('publishedTime', 'Unknown'),
-                'channel': video.get('channel', {}).get('name', 'Unknown'),
-                'description': video.get('descriptionSnippet', [{}])[0].get('text', '') if video.get('descriptionSnippet') else '',
-                'thumbnail': video.get('thumbnails', [{}])[0].get('url', '') if video.get('thumbnails') else ''
-            }
-            
-            # Filter by year if specified
-            if filter_year:
-                upload_text = video_data['upload_date'].lower()
-                # Check if it contains year info
-                if str(filter_year) in upload_text or f"{filter_year}" in video_data['title']:
-                    videos.append(video_data)
-                elif 'year' in upload_text:
-                    # Try to parse "X years ago"
-                    try:
-                        years_ago = int(upload_text.split()[0])
-                        current_year = datetime.now().year
-                        video_year = current_year - years_ago
-                        if video_year <= filter_year:
-                            videos.append(video_data)
-                    except:
-                        pass
-            else:
+        for line in result.stdout.strip().split('\n'):
+            if not line:
+                continue
+            try:
+                video = json.loads(line)
+                
+                # Extract duration - yt-dlp provides it in seconds
+                duration_secs = video.get('duration')
+                if duration_secs:
+                    mins, secs = divmod(int(duration_secs), 60)
+                    hours, mins = divmod(mins, 60)
+                    if hours > 0:
+                        duration_str = f"{hours}:{mins:02d}:{secs:02d}"
+                    else:
+                        duration_str = f"{mins}:{secs:02d}"
+                else:
+                    duration_str = "Unknown"
+                
+                # Format view count
+                view_count = video.get('view_count')
+                if view_count:
+                    if view_count >= 1000000:
+                        views_str = f"{view_count/1000000:.1f}M views"
+                    elif view_count >= 1000:
+                        views_str = f"{view_count/1000:.1f}K views"
+                    else:
+                        views_str = f"{view_count} views"
+                else:
+                    views_str = "Unknown"
+                
+                video_data = {
+                    'title': video.get('title', 'Unknown'),
+                    'url': video.get('url') or f"https://www.youtube.com/watch?v={video.get('id', '')}",
+                    'video_id': video.get('id', ''),
+                    'duration': duration_str,
+                    'duration_seconds': duration_secs,
+                    'views': views_str,
+                    'view_count': view_count,
+                    'upload_date': video.get('upload_date', 'Unknown'),
+                    'channel': video.get('channel') or video.get('uploader', 'Unknown'),
+                    'description': (video.get('description') or '')[:200],
+                    'thumbnail': video.get('thumbnail', '')
+                }
+                
+                # Filter by year if specified
+                if filter_year:
+                    upload_date = video_data['upload_date']
+                    if upload_date and upload_date != 'Unknown':
+                        # yt-dlp provides date as YYYYMMDD
+                        try:
+                            video_year = int(upload_date[:4])
+                            if video_year != filter_year:
+                                continue
+                        except (ValueError, TypeError):
+                            pass
+                
                 videos.append(video_data)
-            
-            if len(videos) >= max_results:
-                break
+                
+                if len(videos) >= max_results:
+                    break
+                    
+            except json.JSONDecodeError:
+                continue
         
         return videos
         
+    except subprocess.TimeoutExpired:
+        print("Error: YouTube search timed out", file=sys.stderr)
+        return []
+    except FileNotFoundError:
+        print("Error: yt-dlp not found. Install with: pip install yt-dlp", file=sys.stderr)
+        return []
     except Exception as e:
         print(f"Error searching YouTube: {str(e)}", file=sys.stderr)
         return []
@@ -102,18 +152,39 @@ def format_results(videos, output_format='text'):
 
 
 def get_video_details(video_url):
-    """Get detailed information about a specific video"""
+    """Get detailed information about a specific video using yt-dlp"""
     try:
-        video_info = Video.getInfo(video_url)
+        cmd = [
+            "yt-dlp",
+            "--dump-json",
+            "--no-download",
+            "--no-warnings",
+            video_url
+        ]
+        
+        result = subprocess.run(
+            cmd, 
+            capture_output=True, 
+            text=True, 
+            timeout=30
+        )
+        
+        if result.returncode != 0:
+            return None
+            
+        video_info = json.loads(result.stdout)
+        
+        duration_secs = video_info.get('duration', 0)
+        
         return {
             'title': video_info.get('title', 'Unknown'),
-            'duration_seconds': video_info.get('duration', {}).get('secondsText', 'Unknown'),
-            'views': video_info.get('viewCount', {}).get('text', 'Unknown'),
-            'upload_date': video_info.get('publishDate', 'Unknown'),
-            'channel': video_info.get('channel', {}).get('name', 'Unknown'),
-            'description': video_info.get('description', '')[:500],
-            'is_live': video_info.get('isLiveNow', False),
-            'category': video_info.get('category', 'Unknown')
+            'duration_seconds': duration_secs,
+            'views': video_info.get('view_count', 'Unknown'),
+            'upload_date': video_info.get('upload_date', 'Unknown'),
+            'channel': video_info.get('channel') or video_info.get('uploader', 'Unknown'),
+            'description': (video_info.get('description') or '')[:500],
+            'is_live': video_info.get('is_live', False),
+            'category': video_info.get('categories', ['Unknown'])[0] if video_info.get('categories') else 'Unknown'
         }
     except Exception as e:
         print(f"Error getting video details: {str(e)}", file=sys.stderr)
@@ -122,7 +193,7 @@ def get_video_details(video_url):
 
 def main():
     parser = argparse.ArgumentParser(
-        description="Search YouTube for videos. Returns video titles, URLs, and metadata.",
+        description="Search YouTube for videos using yt-dlp. Returns video titles, URLs, and metadata.",
         formatter_class=argparse.RawDescriptionHelpFormatter,
         epilog="""
 Examples:

package/tools/validators/link_checker.py

@@ -20,13 +20,13 @@ def check_link(url):
             'Upgrade-Insecure-Requests': '1',
         }
 
-        response = requests.head(url, headers=headers, timeout=5, allow_redirects=True)
+        response = requests.head(url, headers=headers, timeout=15, allow_redirects=True)
 
         if response.status_code == 200:
             return True, "OK"
         else:
             # Retry with GET if HEAD fails (some servers block HEAD)
-            response = requests.get(url, headers=headers, timeout=5, stream=True)
+            response = requests.get(url, headers=headers, timeout=15, stream=True)
             if response.status_code == 200:
                 return True, "OK"
             return False, f"Status Code: {response.status_code}"