Claude adds 12 XML-tagged templates for YouTube roadmaps, hooks, Shorts, and retention

The pack is less a single prompt than a reusable scaffold for Claude. Each template starts by forcing the model to ask for inputs — niche, audience, available hours, top videos, or current subscriber count — before generating a plan growth roadmap. That structure shows up across strategy, packaging, and operations prompts, which is the clearest practical takeaway for creators using Claude as a planner rather than a copy generator.

The templates span channel planning, acquisition, packaging, production, and monetization. The roadmap prompt maps quarterly growth levers and income unlock points; the upload-system prompt turns available weekly hours into a production map with shortcuts and hard-stop checklists; and the email-conversion prompt extends the channel into lead magnets, CTAs, and a three-email welcome sequence upload system email funnel.

The strongest prompts are the ones that convert vague goals into constrained outputs. The Shorts template asks for a niche and main topic, then produces 30 concepts, first-three-second hooks, an awareness/conversion/retention label, and a map from Shorts to long-form videos. That makes it a funnel planner, not just an idea list.

The editing-side prompts are equally specific. The hook template generates eight openings across structures like question, bold claim, story, data, and social proof, then scores “open loop strength” and rewrites the winner for Shorts hook lab. The retention prompt goes deeper by marking likely drop-off moments, rewriting weak sections with pattern interrupts, and scripting the final 60 seconds around one next action.

What makes these prompts interesting is the discipline in the rule blocks. Many explicitly ban fuzzy outputs: title variations must stay under 65 characters and use different triggers; growth tactics must be ranked by effort versus subscriber return; competitor gaps need proof points and a clear number-one opportunity title testing first 1000. For creative teams, that is the reusable pattern: make Claude collect context first, then force comparative scoring, explicit rankings, and fixed output sections.

The limitation is that this is still one creator’s prompt pack, not an official Claude feature release. The thread demonstrates a solid prompt design pattern for YouTube work, but it does not include examples of completed outputs, measured performance gains, or native Claude product changes prompt pack.