Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add hyperfx-ai/marketing-skills --skill youtubegit clone --depth 1 https://github.com/hyperfx-ai/marketing-skillsWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/hyperfx-ai/marketing-skills/youtube)<a href="https://agentmods.dev/skills/hyperfx-ai/marketing-skills/youtube"><img src="https://agentmods.dev/badge/skills/hyperfx-ai/marketing-skills/youtube/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/hyperfx-ai/marketing-skills/youtube"><img src="https://agentmods.dev/badge/skills/hyperfx-ai/marketing-skills/youtube.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00098 | $0.01919 |
| Opus 5 | $0.00049 | $0.00959 |
| Sonnet 5 | $0.00020 | $0.00384 |
| Haiku 4.5 | $0.00010 | $0.00192 |
Grade A, and why
youtube scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 11d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 163 lines — stays where its author put it; the contents beside it link to each section on GitHub.
YouTube
Fetch the full transcript of any YouTube video and turn it into whatever the user needs — summaries, blog posts, social content, quotes, show notes, or raw text. Then package videos for upload: high-CTR thumbnails, SEO titles, and descriptions.
Routing
| User intent | Where to go |
|---|---|
| Transcript, summary, repurposing, quotes, chapters | This guide (below) |
| Thumbnails, style cloning, SEO titles/descriptions | references/thumbnails.md |
Requirements
- Hyper MCP installed. https://app.hyperfx.ai/mcp
- YouTube toolkit enabled at https://app.hyperfx.ai/apps — provides
youtube_video_transcripts_fetchandyoutube_videos_read. - Thumbnail workflows additionally need the image generation and sandbox toolkits.
If youtube_video_transcripts_fetch is not in the tool list, stop and tell the user to enable the YouTube toolkit in Hyper.
Two tools — pick the right one
| Tool | When to use | Returns |
|---|---|---|
youtube_video_transcripts_fetch |
You need the raw transcript text or timestamped segments. Fast, reliable, always get this first. | Full text string + segments with start/duration timestamps |
youtube_videos_read |
You need AI-powered extraction from the video — summaries, Q&A, topic segmentation, translation, visual descriptions. | Free-form answer to your instruction |
Default: start with youtube_video_transcripts_fetch. Use youtube_videos_read when you need something the raw text can't give you (e.g. visual descriptions, translation, or a structured extraction from a very long video).
Critical rules
youtube_video_transcripts_fetchtakes 15–30 seconds. It spins up an isolated sandbox. Tell the user it's running and to expect a short wait — don't make them think it's stuck.- Both video IDs and full URLs are accepted.
"NZLAdOL9fP8"and"https://www.youtube.com/watch?v=NZLAdOL9fP8"both work. - Don't fabricate transcript content. Always fetch before summarizing. Never rely on training knowledge about what a specific video says.
- Very long videos (>2 hours):
youtube_video_transcripts_fetchhandles these fine. Only useyoutube_videos_readon long videos if you specifically need AI-powered extraction — it can hit token limits on very long content. - No transcript available: Some videos have transcripts disabled. If
youtube_video_transcripts_fetchfails, tryyoutube_videos_readas a fallback — it uses a different extraction method.
What ships with it
9 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
- references/composition-tips.md 3.1 KB
- references/thumbnails.md 4.2 KB
- references/youtube-thumbnail-best-practices.md 4.3 KB
- scripts/analyze_thumbnail_concepts.py 4.8 KB runs code
- scripts/clone_top_thumbnail_style.py 5.6 KB runs code
- scripts/generate_seo_description.py 4.9 KB runs code
- scripts/generate_seo_titles.py 3.9 KB runs code
- scripts/generate_thumbnail.py 4.8 KB runs code
- scripts/research_top_thumbnails.py 2.1 KB runs code
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 11d ago First seen · 163 lines · 98 tokens per session scan A 5dfc71adaa49
youtube is a skill published in the GitHub repository hyperfx-ai/marketing-skills (86 stars, last pushed 16d ago), licensed MIT. It adds 98 tokens to every session and 1,919 once invoked, about $0.0005 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
creative-generate
A workflow for creating advertising images and banners from a strategy brief. It separates the main visual artwork from the accurately typeset text and produces variants for display or social formats.
meta-ads
When the user wants to set up, optimize, or manage Meta (Facebook/Instagram) Ads. Also use when the user mentions "Meta Ads," "Facebook Ads," "Instagram Ads," "Meta Pixel," "Conversions API," "Advantage+," "lookalike audience," or "Meta retargeting." For landing pages, use landing-page-generator.
creative-refresh
Refresh ad copy and creative assets based on performance signals and brand voice. Use when the user asks to refresh creative, propose new ad copy, A/B test creatives, update RSA assets, swap Performance Max asset-group headlines, descriptions, images or logos, rotate underperformers, or visually evaluate / compare…
visual-content
When the user wants to plan, create, or repurpose visual content (images, infographics, social post images) across channels. Also use when the user mentions "content images," "social media images," "infographic," "visual content," "post image," "image specs," "visual repurposing," "content visuals," or "image for…
creator-program
When the user wants to plan, implement, or optimize creator program strategy. Also use when the user mentions "creator program," "creator partnership," "content co-creation," "creator ambassador," "creator economy," "creator collaboration," "UGC program," "creator incentives," or "creator community." For influencers…
podcast-marketing
When the user wants to plan, create, or market a podcast. Also use when the user mentions "podcast," "podcast strategy," "podcast SEO," "show notes," "podcast distribution," "Spotify podcast," "Apple Podcasts," "podcast discoverability," "PodcastEpisode schema," or "podcast repurposing." For show notes, use…