skills.rest
harperaa
harperaaCOMMUNITY
Explore all skills by this author →
GitHub

youtube-bulk-ingest

Ingest and index YouTube transcripts from a local workspace into an insight system.

30|3|Updated Feb 21, 2026
Search Tags:#insights#youtube#batching#transcripts#timestamps#ingest
One-click install
npx skills add https://github.com/harperaa/bastionclaw --skill youtube-bulk-ingest
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: youtube-bulk-ingest
Source: https://github.com/harperaa/bastionclaw/tree/main/.claude/skills/youtube-bulk-ingest
Command: npx skills add https://github.com/harperaa/bastionclaw --skill youtube-bulk-ingest

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Bulk ingest YouTube transcripts from the local workspace into the insight tracking system. Scans the youtube/ directory for videos with transcripts, skips already-indexed ones, and extracts insights with timestamps. Reusable for future batches. Triggers on "youtube bulk ingest", "ingest all videos", "process youtube transcripts", "bulk ingest".

Core Features & Use Cases

  • Bulk ingest: scans transcripts, builds a manifest, filters indexed transcripts, and extracts insights with timestamps for each video.
  • Metadata-driven processing: uses transcript.json and metadata/*.json to attach source information and context.
  • Idempotent batch handling: designed to be run repeatedly without re-indexing already processed videos, enabling scheduled batch ingests.

Quick Start

Trigger the skill with the phrase youtube bulk ingest to begin processing the newest transcripts in your workspace.

Frequently Asked Questions about youtube-bulk-ingest

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I bulk ingest YouTube transcripts to extract timestamped insights?▼

You can bulk ingest YouTube transcripts by triggering the system with the phrase "youtube bulk ingest". It scans your local workspace directory, skips already-indexed videos, and extracts timestamped insights from the transcripts.

What directory structure is required to process YouTube transcripts in batches?▼

Processing YouTube transcripts requires a local workspace organized with a youtube/ directory. This directory must contain transcript.json files and a metadata/ folder with JSON files to attach source information and context during the ingest process.

Can I run the YouTube transcript ingest process repeatedly without creating duplicate insights?▼

Yes, the YouTube transcript ingest process is idempotent. It builds a manifest and filters out already-indexed transcripts, allowing you to safely run scheduled batch ingests repeatedly without duplicating insights in the system.

How does the system handle metadata when ingesting local video batches?▼

When ingesting local video batches, the system uses metadata files within the youtube/ directory structure. It reads the metadata/*.json files to attach source information and context to the extracted timestamped insights.

What operations are supported after ingesting YouTube transcripts into the insight system?▼

After ingesting YouTube transcripts, the insight system supports search_insights, add_insight, and refresh_memory_index operations. These allow you to query, update, and maintain your extracted timestamped insights.

Related Skills

Explore More Data & Analytics Skills→
akaiholaakaihola
2

YouTube Transcription

Transcribe YouTube videos and local audio files into Markdown transcripts.

Community
Intermediate
spoonbobospoonbobo

youtube-digest

Fetch recent YouTube videos, extract transcripts, and generate key insight digests.

Community
Intermediate
jamie8johnsonjamie8johnson
13

cqs-ref

An error occurred during the request.

Community
Intermediate
simoneschiavosimoneschiavo

youtube

Extract plain or timestamped transcripts and insights from YouTube videos.

Community
Intermediate