root

YouTube Content Creation System Overview

youtube tubeforge content-creation automation video-production

YouTube Content Creation System (TubeForge)

Version: 1.0.0
Status: Production Ready
Created: 2026-01-11
Category: YouTube System

What is TubeForge?

TubeForge is Nexus's comprehensive YouTube content creation system that automates 80-90% of the script generation workflow. It transforms competitor research into production-ready video scripts in minutes instead of hours.

Core Value Proposition: - Time Savings: 80-90% reduction in script creation time - Data-Driven: Real engagement metrics guide content decisions - Scalable: Multi-channel support built-in - Integrated: Leverages existing Nexus infrastructure

System Architecture

Hybrid Approach

TubeForge uses a hybrid architecture that combines:

  1. NEW: YouTube Metadata Environment (Ports 6695/6696)
  2. Channels and episode management
  3. Metadata hub (hashtags, keywords, status)

  4. Links to other environments

  5. EXISTING: Nexus Infrastructure

  6. Transcripts (6670/6671): Competitor video transcripts
  7. Corpus (6660/6661): Script storage with version control
  8. Web (6680/6681): Search history
  9. CDN: Final script files (/data/cdn/users/chris/scripts/)

Why Hybrid? - Leverages existing proven infrastructure (Transcripts MCP already built!) - Clean separation of YouTube-specific metadata - Built-in version control via Corpus hierarchy - Minimal new code (~300 lines for YouTube MCP) - Unified search via Search MCP (Quadfecta)

Key Components

1. Discovery Module

Purpose: Find trending topics and analyze competitors

Location: /opt/mcp-servers/youtube/discovery.py

Capabilities: - Search trending videos within date ranges - Get top videos by views/relevance/recency - Analyze competitor channel strategies - Extract detailed metadata (tags, descriptions, engagement)

Technology: scrapetube 2.6.0 + yt-dlp (no API keys required!)

Performance: - 5-8 seconds per 10-video search - No rate limits (self-managed) - 100% test success rate

2. YouTube MCP Server

Purpose: Central metadata hub and orchestration

Location: /opt/mcp-servers/youtube/mcp_youtube_server.py

Ports: 6695 (vault) / 6696 (operational)

Tools (11 total): - Channel Management (3): create, list, get - Episode Management (4): create, update, get, list - Cross-Environment Linking (3): link_script, link_search, link_transcript - Aggregation (1): episode_full (get complete context)

Integration: Auto-discovered by Gateway MCP

3. Transcript Integration

Existing Tool: Transcripts MCP already supports YouTube!

Capabilities: - Fetch competitor video transcripts - No API limits - Fast retrieval (2-4 seconds) - Automatic storage in Transcripts environment

Usage: Research competitors, extract successful patterns

4. Script Generation Pipeline

Purpose: Transform research into production scripts

Templates: 3 duration formats - 3-minute video format - 5-minute video format - 10-minute video format

Features: - Automatic timing calculations - Section headers with timestamps - Production notes - Metadata packages (hashtags, keywords, description)

Version Control: Corpus hierarchy (parent-child links)

5. CDN Integration

Storage Location: /data/cdn/users/chris/scripts/YYYY-MM-DD/

Pattern: Date-based folders for organization

Example:

/data/cdn/users/chris/scripts/2026-01-11/
├── ep_xyz1_v3.md
├── ep_abc2_v1.md
└── ep_def3_v2.md

Complete Workflow

Step 1: Discovery

# Search for trending topics
videos = youtube.get_top_videos("AI agents", limit=10)
# Result: 1,893,494 avg views per video (HOT TOPIC!)

Step 2: Create Episode

# Create episode in YouTube environment
episode = youtube.episode_create(
    channel_id="@chrisfoustdev",
    title="How to Build an MCP Server",
    description="Complete tutorial...",
    hashtags=["#MCP", "#AI"],
    keywords=["MCP server", "Claude"]
)
# Returns: ep_xyz1

Step 3: Competitor Research

# Get search results (Web environment)
search = web.search_youtube("how to build MCP server")
youtube.episode_link_search("ep_xyz1", search.id)

# Get competitor transcripts (Transcripts environment)
for result in search.results[:3]:
    transcript = transcript.youtube_get(result.video_id)
    youtube.episode_link_transcript("ep_xyz1", transcript.id)

Step 4: Generate Script

# Generate script v1 (Corpus environment)
script_v1 = corpus.create(
    category="youtube_script",
    title="MCP Tutorial - Script v1",
    content="...",  # Generated from research
    tags=["youtube", "@chrisfoustdev", "ep_xyz1"]
)
youtube.episode_link_script("ep_xyz1", script_v1.stable_id)

Step 5: Iterate & Refine

# Create v2 (links to v1 via parent_id)
script_v2 = corpus.create(
    category="youtube_script",
    title="MCP Tutorial - Script v2",
    content="...",  # Refined version
    parent_id=script_v1.id,  # Version control!
    tags=["youtube", "@chrisfoustdev", "ep_xyz1"]
)
youtube.episode_update("ep_xyz1", current_script_id=script_v2.stable_id)

Step 6: Finalize to CDN

# Export to CDN for production
cdn_path = f"/data/cdn/users/chris/scripts/2026-01-11/ep_xyz1_v2.md"
with open(cdn_path, 'w') as f:
    f.write(script_v2.content)

youtube.episode_update(
    "ep_xyz1",
    cdn_script_path=cdn_path,
    status="recording"
)

Step 7: Get Full Context Anytime

# Retrieve complete episode with all linked resources
full = youtube.episode_full("ep_xyz1")
# Returns:
# - Episode metadata
# - All script versions (Corpus IDs)
# - Search results (Web ID)
# - Competitor transcripts (Transcript IDs)

Market Insights

Real engagement data from testing 5 AI topics:

  1. Latest AI News - 2,354,840 avg views
  2. AI Agents Tutorial - 1,893,494 avg views (🔥 HOTTEST)
  3. n8n Automation - 1,215,402 avg views
  4. Claude Code Tutorial - 1,033,809 avg views
  5. MCP Server Development - 741,164 avg views

Finding: ALL topics show strong engagement (700K+ avg views)

Opportunity: MCP and Claude Code are BRAND NEW (late 2025 launch) - less competition, massive interest

Content Gaps (Opportunities)

  1. Comprehensive MCP Server Tutorials
  2. Most content is overview/announcement

  3. Lack of "building from scratch" content

  4. Claude Code Advanced Use Cases

  5. Official content covers basics

  6. Missing: advanced workflows, MCP integration

  7. n8n + MCP Integration

  8. No major content combining these hot topics
  9. Perfect opportunity for first-mover advantage

Time Savings Analysis

Traditional Manual Workflow

  1. Research competitors: 2-3 hours
  2. Analyze engagement: 1 hour
  3. Draft outline: 1 hour
  4. Write full script: 3-4 hours
  5. Revise and refine: 1-2 hours

Total: 8-11 hours per video

TubeForge Automated Workflow

  1. Discovery search: 30 seconds
  2. Competitor analysis: 2 minutes
  3. Script generation: 5 minutes
  4. Human review/refinement: 30-60 minutes

Total: 40-70 minutes per video

Time Savings: 80-90% reduction

Multi-Channel Support

TubeForge is designed for managing multiple YouTube channels:

Channel Examples: - @chrisfoustdev - Developer tutorials - @corleraai - AI automation, enterprise solutions

Per-Channel Tracking: - Episode count - Total views (future) - Engagement metrics (future) - Publishing schedule (future)

Episode Status Workflow:

draft → recording → editing → published

Status updates tracked per episode for production pipeline visibility.

Integration with Nexus Ecosystem

Search MCP (Quadfecta)

Benefit: Unified search across all environments

Use Case: Find all episodes related to "MCP" across YouTube, Corpus, Transcripts, Web

results = search.search("MCP server")
# Returns episodes, scripts, transcripts, searches

Track Projects

Integration: Link Track projects to episodes

Use Case: Video production as Track project tasks

track.create(
    title="MCP Tutorial Video Production",
    description="Episode ep_xyz1",
    category="youtube_production"
)

Voice Announcements

Integration: Status updates via Voice MCP

Use Case: Announce when script ready for recording

voice.update("Script v3 for MCP Tutorial is ready for recording")

Production Status

Completed ✅

  • YouTube Metadata MCP (11 tools)
  • Discovery module (5 functions)
  • Gateway integration (auto-discovered)
  • CLI wrapper for discovery
  • Comprehensive testing (100% success)
  • Production deployment complete

In Development 🚧

  • Script generation templates
  • Conversation protocol (intent recognition)
  • Workflow documentation
  • Metadata research finalization

Planned 📋

  • Analytics dashboard
  • Publishing automation (YouTube API)
  • Scheduling system
  • Thumbnail generation
  • Multi-channel dashboards

Quick Reference

Environment Ports

  • YouTube: 6695 (vault), 6696 (operational)
  • Transcripts: 6670 (vault), 6671 (operational)
  • Corpus: 6660 (vault), 6661 (operational)
  • Web: 6680 (vault), 6681 (operational)

Key Files

  • YouTube MCP: /opt/mcp-servers/youtube/mcp_youtube_server.py
  • Discovery Module: /opt/mcp-servers/youtube/discovery.py
  • CLI Tool: /opt/mcp-servers/youtube/discover.py
  • Scripts CDN: /data/cdn/users/chris/scripts/YYYY-MM-DD/

Tool Prefixes

  • YouTube env: ytube:u_z1p5:channel: or ytube:u_z1p5:episode:
  • Corpus scripts: corp:u_z1p5:timestamp:youtube_script
  • Web searches: web:u_z1p5:search:timestamp
  • Transcripts: transcript:u_z1p5:video_id

Next Steps for Users

  1. Explore trending topics: Use discover.py --search "your topic" CLI
  2. Create your first channel: youtube.channel_create(@yourhandle, "Name")
  3. Generate episode idea: Research competitors, analyze engagement
  4. Create episode: youtube.episode_create with metadata
  5. Link research: Connect search results and competitor transcripts
  6. Generate script: Use script templates based on target duration
  7. Iterate: Create v2, v3 using Corpus hierarchy
  8. Export to CDN: Finalize for production recording

Support & Documentation

Related KB Articles: - YouTube Discovery and Trending Analysis - YouTube Script Generation Complete Guide - YouTube MCP Server Technical Reference - YouTube Conversational Workflows

Files Location: /opt/mcp-servers/youtube/

Gateway Integration: All tools accessible via gateway.run([{server:'youtube', tool:'...', args:{...}}])

System Codename: TubeForge
Operation: g_vmie
Primary Channel: @chrisfoustdev
Created: 2026-01-11

Contents

ID: 1b191f94
Path: YouTube Content Creation System Overview
Updated: 2026-01-13T12:51:50
🌳 View Tree