solo-index-youtube

What This Skill Does

The solo-index-youtube skill is a powerful tool designed to bring the vast knowledge contained within YouTube videos into your local OpenClaw workspace. It enables you to index entire channels or specific videos, extracting transcripts and converting them into a semantic knowledge base. Whether you are tracking industry leaders, tutorials, or long-form podcasts, this skill allows you to perform semantic searches across video content. It operates in two intelligent modes: it defaults to using the solograph MCP (Model Context Protocol) for advanced vector-based indexing and tag management if detected, or falls back to a robust standalone process using yt-dlp to fetch and summarize transcripts locally.

Installation

To add this skill to your OpenClaw agent, ensure you have the necessary dependencies installed on your system. First, verify yt-dlp is available (which yt-dlp). If you wish to leverage the advanced features of the Solograph integration, install it via pip install solograph or uvx solograph. Finally, run the following command in your terminal to initialize the skill within your agent:

clawhub install openclaw/skills/skills/fortunto2/solo-index-youtube

Use Cases

• Research and Education: Index educational channels to quickly query specific concepts discussed in multi-hour lectures.
• Content Monitoring: Keep up with industry influencers by indexing the latest uploads from your favorite creators automatically.
• Knowledge Management: Create searchable documentation from technical tutorials, effectively converting audio-visual information into text-based insights.
• Competitive Intelligence: Analyze multiple channels simultaneously to identify recurring themes and market trends.

Example Prompts

• "Index the last 5 videos from the YCombinator YouTube channel so I can search for startup advice."
• "Add the GregIsenberg channel to my index and give me a summary of his recent videos about AI business models."
• "Update my YouTube transcripts index; I need to find the section in the latest devlog that mentions environment variables."

Tips & Limitations

• Efficiency: Always use the solograph MCP if possible for faster, more accurate semantic results.
• Cleanliness: When running in standalone mode, the skill generates a docs/youtube/ directory. Be mindful of disk space if indexing hundreds of videos.
• Formatting: Standalone mode relies on sed and awk to clean VTT subtitles; ensure these utilities are available in your shell environment for the best experience.
• Dry Runs: Use the --dry-run flag before bulk indexing to verify that the tool is targeting the correct channels.