{"id":"508a6ccf-dbdb-4efd-af83-d25bb2e4ae5f","shortId":"5HwDE9","kind":"skill","title":"youtube-transcript","tagline":"Fetch and work with YouTube video transcripts and content. Use when the user pastes a YouTube URL and wants a transcript, summary, blog post, social post, quote extraction, show notes, or any other text derived from a video. Also use when the user wants to repurpose video content","description":"# YouTube Transcript\n\nFetch 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.\n\n## Requirements\n\n- **Hyper MCP installed.** [https://app.hyperfx.ai/mcp](https://app.hyperfx.ai/mcp)\n- **YouTube toolkit enabled** at [https://app.hyperfx.ai/integrations](https://app.hyperfx.ai/integrations) — provides `youtube_transcript` and `youtube_reader`.\n\nIf `youtube_transcript` is not in the tool list, stop and tell the user to enable the YouTube toolkit in Hyper.\n\n## Two tools — pick the right one\n\n| Tool | When to use | Returns |\n| --- | --- | --- |\n| `youtube_transcript` | You need the raw transcript text or timestamped segments. Fast, reliable, always get this first. | Full text string + segments with start/duration timestamps |\n| `youtube_reader` | You need AI-powered extraction from the video — summaries, Q&A, topic segmentation, translation, visual descriptions. | Free-form answer to your instruction |\n\n**Default: start with `youtube_transcript`.** Use `youtube_reader` 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).\n\n## Critical rules\n\n1. **`youtube_transcript` takes 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.\n2. **Both video IDs and full URLs are accepted.** `\"NZLAdOL9fP8\"` and `\"https://www.youtube.com/watch?v=NZLAdOL9fP8\"` both work.\n3. **Don't fabricate transcript content.** Always fetch before summarizing. Never rely on training knowledge about what a specific video says.\n4. **Very long videos (>2 hours):** `youtube_transcript` handles these fine. Only use `youtube_reader` on long videos if you specifically need AI-powered extraction — it can hit token limits on very long content.\n5. **No transcript available:** Some videos have transcripts disabled. If `youtube_transcript` fails, try `youtube_reader` as a fallback — it uses a different extraction method.\n\n## Fetching the transcript\n\n```python\nyoutube_transcript(\n    video_id_or_url=\"https://www.youtube.com/watch?v=NZLAdOL9fP8\",\n    language=\"en\"   # optional — omit to auto-detect\n)\n```\n\n**Response structure:**\n\n```json\n{\n  \"success\": true,\n  \"video_id\": \"NZLAdOL9fP8\",\n  \"language\": \"English (auto-generated)\",\n  \"text\": \"Full transcript as one string...\",\n  \"segments\": [\n    { \"text\": \"This week we launched Hyper MCP.\", \"start\": 0.0, \"duration\": 3.2 },\n    { \"text\": \"It brings Hyper's built-in tools...\", \"start\": 3.2, \"duration\": 4.1 }\n  ],\n  \"total_duration\": 342.0\n}\n```\n\nUse `text` for most tasks. Use `segments` when you need timestamps (e.g. chapters, clip references, karaoke captions).\n\n## Using youtube_reader for AI-powered extraction\n\n```python\nyoutube_reader(\n    url=\"https://www.youtube.com/watch?v=NZLAdOL9fP8\",\n    instruction=\"Summarize the key points. Then list the main features demonstrated, with timestamps.\"\n)\n```\n\nGood `instruction` examples:\n- `\"Extract every claim made about pricing or cost.\"`\n- `\"List the action items mentioned, in order.\"`\n- `\"Translate this to Spanish.\"`\n- `\"What tools or products does the speaker mention by name?\"`\n- `\"Identify the main sections of this video and give me a timestamp for each.\"`\n\n## What to do with the transcript\n\nOnce you have the text, ask the user what they need — or infer it from context:\n\n| What the user wants | What to produce |\n| --- | --- |\n| Blog post | Restructure the transcript into intro → sections → CTA. Clean up filler words. Add subheadings. |\n| LinkedIn / Twitter post | Extract the 1–2 sharpest insights. Rewrite in first person if it's the user's own video. |\n| Summary | 3–5 bullet points of key takeaways. |\n| Show notes / description | Title, 2-sentence summary, timestamped chapters, links mentioned. |\n| Quote extraction | Pull verbatim quotes with `start` timestamps from the segments array. |\n| Repurpose for email | Rewrite as a narrative email — opening hook, key insight, CTA. |\n| Research / competitive analysis | Summarize what the speaker claims, what products they recommend, and what pain points they describe. |\n\n## Example outputs\n\n**Input:** `\"Get the transcript of https://www.youtube.com/watch?v=NZLAdOL9fP8 and write a LinkedIn post from it\"`\n\n**Flow:**\n1. Call `youtube_transcript(video_id_or_url=\"https://www.youtube.com/watch?v=NZLAdOL9fP8\")`\n2. Read the returned `text`\n3. Identify the 1–2 sharpest moments — what's surprising, useful, or quotable\n4. Draft a LinkedIn post in the speaker's voice (first person) with a hook and a clear point\n\n**Input:** `\"Summarize this video for me: [URL]\"`\n\n**Flow:**\n1. Call `youtube_transcript(video_id_or_url=\"[URL]\")`\n2. Return 4–6 bullet points of key takeaways, without padding or filler\n\n**Input:** `\"Pull the timestamps for each section of this video\"`\n\n**Flow:**\n1. Call `youtube_transcript(video_id_or_url=\"[URL]\")`\n2. Use the `segments` array to identify topic shifts\n3. Return a chapter list: `00:00 — Intro`, `01:23 — Feature demo`, etc.\n\n## Related skills\n\n| When to hand off | Skill |\n| --- | --- |\n| Mining comments from YouTube videos for customer research | [`customer-research`](../customer-research) |\n| Finding top YouTube videos by topic | Use `youtube_top_videos` directly |\n| Generating video content | [`video-generation`](../video-generation) |","tags":["youtube","transcript","marketing","skills","hyperfx-ai","agent-skills","ai-agent","claude","claude-code","claude-skills","codex","cursor"],"capabilities":["skill","source-hyperfx-ai","skill-youtube-transcript","topic-agent-skills","topic-ai-agent","topic-claude","topic-claude-code","topic-claude-skills","topic-codex","topic-cursor","topic-google-ads","topic-hermes","topic-marketing","topic-mcp","topic-meta-ads"],"categories":["marketing-skills"],"synonyms":[],"warnings":[],"endpointUrl":"https://skills.sh/hyperfx-ai/marketing-skills/youtube-transcript","protocol":"skill","transport":"skills-sh","auth":{"type":"none","details":{"cli":"npx skills add hyperfx-ai/marketing-skills","source_repo":"https://github.com/hyperfx-ai/marketing-skills","install_from":"skills.sh"}},"qualityScore":"0.462","qualityRationale":"deterministic score 0.46 from registry signals: · indexed on github topic:agent-skills · 24 github stars · SKILL.md body (5,406 chars)","verified":false,"liveness":"unknown","lastLivenessCheck":null,"agentReviews":{"count":0,"score_avg":null,"cost_usd_avg":null,"success_rate":null,"latency_p50_ms":null,"narrative_summary":null,"summary_updated_at":null},"enrichmentModel":"deterministic:skill-github:v1","enrichmentVersion":1,"enrichedAt":"2026-05-18T19:05:17.757Z","embedding":null,"createdAt":"2026-05-10T01:04:28.128Z","updatedAt":"2026-05-18T19:05:17.757Z","lastSeenAt":"2026-05-18T19:05:17.757Z","tsv":"'/customer-research':789 '/integrations](https://app.hyperfx.ai/integrations)':94 '/mcp](https://app.hyperfx.ai/mcp)':87 '/video-generation':807 '/watch?v=nzladol9fp8':263,359,446,642,661 '0.0':396 '00':763,764 '01':766 '1':217,555,651,670,707,740 '15':221 '2':250,291,556,583,662,671,716,749 '23':767 '3':266,572,667,758 '3.2':398,409 '30':222 '342.0':414 '4':287,680,718 '4.1':411 '5':322,573 '6':719 'accept':258 'action':473 'add':548 'ai':162,310,437 'ai-pow':161,309,436 'also':42 'alway':146,272 'analysi':617 'answer':179 'app.hyperfx.ai':86,93 'app.hyperfx.ai/integrations](https://app.hyperfx.ai/integrations)':92 'app.hyperfx.ai/mcp](https://app.hyperfx.ai/mcp)':85 'array':601,753 'ask':517 'auto':366,379 'auto-detect':365 'auto-gener':378 'avail':325 'blog':26,71,535 'bring':401 'built':405 'built-in':404 'bullet':574,720 'call':652,708,741 'caption':431 'chapter':427,587,761 'claim':465,622 'clean':544 'clear':697 'clip':428 'comment':779 'competit':616 'content':12,51,74,271,321,803 'context':527 'cost':470 'critic':215 'cta':543,614 'custom':784,787 'customer-research':786 'default':183 'demo':769 'demonstr':457 'deriv':38 'describ':632 'descript':175,204,581 'detect':367 'differ':344 'direct':800 'disabl':330 'draft':681 'durat':397,410,413 'e.g':202,426 'email':604,609 'en':361 'enabl':90,116 'english':377 'etc':770 'everi':464 'exampl':462,633 'expect':238 'extract':31,164,209,312,345,439,463,553,591 'fabric':269 'fail':334 'fallback':340 'fast':144 'featur':456,768 'fetch':4,54,273,347 'filler':546,728 'find':790 'fine':297 'first':149,561,690 'flow':650,706,739 'form':178 'free':177 'free-form':176 'full':56,150,255,382 'generat':380,801,806 'get':147,636 'give':200,500 'good':460 'hand':775 'handl':295 'hit':315 'hook':611,694 'hour':292 'hyper':82,121,393,402 'id':253,354,374,656,712,745 'identifi':492,668,755 'infer':524 'input':635,699,729 'insight':558,613 'instal':84 'instruct':182,447,461 'intro':541,765 'isol':228 'item':474 'json':370 'karaok':430 'key':450,577,612,723 'knowledg':280 'languag':360,376 'launch':392 'limit':317 'link':588 'linkedin':550,646,683 'list':109,453,471,762 'long':213,289,303,320 'made':466 'main':455,494 'make':244 'mcp':83,394 'mention':475,489,589 'method':346 'mine':778 'moment':673 'name':491 'narrat':608 'need':69,136,160,193,308,424,522 'never':276 'note':33,77,580 'nzladol9fp8':259,375 'omit':363 'one':127,385 'open':610 'option':362 'order':477 'output':634 'pad':726 'pain':629 'past':17 'person':562,691 'pick':124 'point':451,575,630,698,721 'post':27,29,72,536,552,647,684 'power':163,311,438 'price':468 'produc':534 'product':485,624 'provid':95 'pull':592,730 'python':350,440 'q':169 'quot':30,75,590,594 'quotabl':679 'raw':79,138,196 'read':663 'reader':100,158,190,301,337,434,442 'recommend':626 'refer':429 'relat':771 'reli':277 'reliabl':145 'repurpos':49,602 'requir':81 'research':615,785,788 'respons':368 'restructur':537 'return':132,665,717,759 'rewrit':559,605 'right':126 'rule':216 'run':235 'sandbox':229 'say':286 'second':223 'section':495,542,735 'segment':143,153,172,387,421,600,752 'sentenc':584 'sharpest':557,672 'shift':757 'short':240 'show':32,76,579 'skill':772,777 'skill-youtube-transcript' 'social':28,73 'someth':194 'source-hyperfx-ai' 'spanish':481 'speaker':488,621,687 'specif':284,307 'spin':225 'start':184,395,408,596 'start/duration':155 'stop':110 'string':152,386 'structur':208,369 'stuck':249 'subhead':549 'success':371 'summar':275,448,618,700 'summari':25,70,168,571,585 'surpris':676 'take':220 'takeaway':578,724 'task':419 'tell':112,230 'text':37,80,140,151,197,381,388,399,416,516,666 'think':246 'timestamp':142,156,425,459,503,586,597,732 'titl':582 'token':316 'tool':108,123,128,407,483 'toolkit':89,119 'top':791,798 'topic':171,756,795 'topic-agent-skills' 'topic-ai-agent' 'topic-claude' 'topic-claude-code' 'topic-claude-skills' 'topic-codex' 'topic-cursor' 'topic-google-ads' 'topic-hermes' 'topic-marketing' 'topic-mcp' 'topic-meta-ads' 'total':412 'train':279 'transcript':3,10,24,53,57,97,103,134,139,187,219,270,294,324,329,333,349,352,383,511,539,638,654,710,743 'translat':173,205,478 'tri':335 'true':372 'turn':63 'twitter':551 'two':122 'url':20,256,356,443,658,705,714,715,747,748 'use':13,43,131,188,299,342,415,420,432,677,750,796 'user':16,46,68,114,232,519,530,567 'verbatim':593 'video':9,41,50,61,167,214,252,285,290,304,327,353,373,498,570,655,702,711,738,744,782,793,799,802,805 'video-gener':804 'visual':174,203 'voic':689 'wait':241 'want':22,47,531 'week':390 'whatev':66 'without':725 'word':547 'work':6,265 'write':644 'www.youtube.com':262,358,445,641,660 'www.youtube.com/watch?v=nzladol9fp8':261,357,444,640,659 'youtub':2,8,19,52,60,88,96,99,102,118,133,157,186,189,218,293,300,332,336,351,433,441,653,709,742,781,792,797 'youtube-transcript':1","prices":[{"id":"3e290658-c569-4c27-8216-bd5623047fd8","listingId":"508a6ccf-dbdb-4efd-af83-d25bb2e4ae5f","amountUsd":"0","unit":"free","nativeCurrency":null,"nativeAmount":null,"chain":null,"payTo":null,"paymentMethod":"skill-free","isPrimary":true,"details":{"org":"hyperfx-ai","category":"marketing-skills","install_from":"skills.sh"},"createdAt":"2026-05-10T01:04:28.128Z"}],"sources":[{"listingId":"508a6ccf-dbdb-4efd-af83-d25bb2e4ae5f","source":"github","sourceId":"hyperfx-ai/marketing-skills/youtube-transcript","sourceUrl":"https://github.com/hyperfx-ai/marketing-skills/tree/main/skills/youtube-transcript","isPrimary":false,"firstSeenAt":"2026-05-10T01:04:28.128Z","lastSeenAt":"2026-05-18T19:05:17.757Z"}],"details":{"listingId":"508a6ccf-dbdb-4efd-af83-d25bb2e4ae5f","quickStartSnippet":null,"exampleRequest":null,"exampleResponse":null,"schema":null,"openapiUrl":null,"agentsTxtUrl":null,"citations":[],"useCases":[],"bestFor":[],"notFor":[],"kindDetails":{"org":"hyperfx-ai","slug":"youtube-transcript","github":{"repo":"hyperfx-ai/marketing-skills","stars":24,"topics":["agent-skills","ai-agent","claude","claude-code","claude-skills","codex","cursor","google-ads","hermes","marketing","mcp","meta-ads","openclaw","seo"],"license":"mit","html_url":"https://github.com/hyperfx-ai/marketing-skills","pushed_at":"2026-05-09T22:58:46Z","description":"Marketing skills for AI agents — paid ads, social media, SEO, competitor research, creative generation, email, analytics, and more. Powered by Hyper MCP.","skill_md_sha":"b8266e99dece3cbb7c3ecceeaf11345f82cc4a41","skill_md_path":"skills/youtube-transcript/SKILL.md","default_branch":"main","skill_tree_url":"https://github.com/hyperfx-ai/marketing-skills/tree/main/skills/youtube-transcript"},"layout":"multi","source":"github","category":"marketing-skills","frontmatter":{"name":"youtube-transcript","description":"Fetch and work with YouTube video transcripts and content. Use when the user pastes a YouTube URL and wants a transcript, summary, blog post, social post, quote extraction, show notes, or any other text derived from a video. Also use when the user wants to repurpose video content, research competitor videos, extract key points without watching, pull timestamps, translate a video, or analyze what someone said in a YouTube video."},"skills_sh_url":"https://skills.sh/hyperfx-ai/marketing-skills/youtube-transcript"},"updatedAt":"2026-05-18T19:05:17.757Z"}}