install
source · Clone the upstream repo
git clone https://github.com/majiayu000/claude-skill-registry
Claude Code · Install into ~/.claude/skills/
T=$(mktemp -d) && git clone --depth=1 https://github.com/majiayu000/claude-skill-registry "$T" && mkdir -p ~/.claude/skills && cp -r "$T/skills/data/consume-youtube" ~/.claude/skills/majiayu000-claude-skill-registry-consume-youtube && rm -rf "$T"
manifest:
skills/data/consume-youtube/SKILL.mdsource content
--- name: consume-youtube description: > Search and annotate YouTube transcripts already ingested by /ingest-youtube. Supports per-channel indexing and Horus notes with /memory integration. triggers: - consume youtube - search youtube - youtube notes - youtube transcript search allowed-tools: - Bash - Python metadata: short-description: Consume YouTube transcripts - search, index, take notes --- # Consume YouTube Skill Search and annotate ingested YouTube transcripts. Works with content already processed by `/ingest-youtube`. ## Quick Start ```bash cd .pi/skills/consume-youtube # Import transcripts into registry ./run.sh sync # Search transcripts ./run.sh search "siege" --channel "luetin09" # Build a channel index ./run.sh index --channel "luetin09" # Add a note at a timestamp ./run.sh note --video <video_id> --timestamp 184.5 --note "Key claim" # List videos ./run.sh list --channel "luetin09" ``` ## Commands ### Sync Transcripts ```bash ./run.sh sync [--ingest-root <path>] ``` Imports transcript JSON files from the ingest output directory. ### Search Transcripts ```bash ./run.sh search <query> [--channel <name>] [--video <id>] [--context <n>] ``` Searches transcript segments and returns matches with timestamps and context. ### Build Index ```bash ./run.sh index --channel <name> ``` Builds a lightweight inverted index for faster channel searches. ### Add Note ```bash ./run.sh note --video <id> --timestamp <sec> --note <text> [--agent <id>] ``` Adds a Horus note at a timestamp. ### List Videos ```bash ./run.sh list [--json] [--channel <name>] ``` Lists all indexed videos in the registry. ## Data Storage - **Registry**: `~/.pi/consume-youtube/registry.json` - **Notes**: `~/.pi/consume-youtube/notes/<agent_id>/notes.jsonl` - **Indices**: `~/.pi/consume-youtube/indices/<channel>.json` ## Integration with /memory Notes are stored in `/memory` using the consume-common memory bridge: ```bash ./memory/run.sh learn \ --problem "Consumed youtube: <title>" \ --solution "<note>" \ --category emotional_learning \ --tags youtube,horus_lupercal ```