youtube
Skill35 starsChanged 6 months agoMITView on GitHub
Search YouTube intelligently for the best video on any topic. Use this skill whenever the user wants to find YouTube videos, search for tutorials, watch something, or discover content on a topic — even if they just say "find me a video about X" or "look up X on YouTube." Runs multiple searches, interprets the user's intent, and returns one clear recommendation plus a few alternatives.
36 other projects use a version of this, 1 of them edited.
Adopted by
Most recently taken up by
- DmacMcgreg/psibottheir own version, first seen Nov 8, 2026
- pietz/skillstheir own version, first seen Nov 8, 2026
- tubeagentkit/youtube-mcptheir own version, first seen Oct 27, 2026
- dcbruck/claude-skillstheir own version, first seen Oct 20, 2026
- ZeroPointRepo/youtube-mcptheir own version, first seen Oct 20, 2026
- manifoldmcp/marketing-skillstheir own version, first seen Oct 17, 2026
Used with
Projects with it also use tiktok, linkedin and 1 more.
Other files people use alongside this one.
Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
G-HunterAiSkilltiktok
TikTok platform integration. Manage videos, view analytics, and track engagement.
Also in 5 projects that use this
kkoppenhaverSkilllinkedin
Write LinkedIn posts in Keanan's voice. Use when drafting LinkedIn content, creating professional posts, or sharing builder-focused insights. Triggers include requests to "write a LinkedIn post", "draft something for LinkedIn", "help me post on LinkedIn", or any LinkedIn content creation. Generates upbeat, builder-centric content mixing practical how-tos with candid humor.
Also in 3 projects that use this
SentrySkillskill-creator
Create new agent skills following the Agent Skills specification. Use when asked to "create a skill", "add a new skill", "write a skill", "make a skill", "build a skill", or scaffold a new skill with SKILL.md. Guides through requirements, writing, registration, and verification.
Also in 4 projects that use this
Other versions36 projects use a version of this.
Word for word · 0
None found.
Edited · 1
- blueincite/claude-skills★ 051% same
Wrote their own version · 35
- eat-pray-ai/yutu★ 716same name
- sigcli/sigcli★ 293same name
- refly-ai/refly-skills★ 204same name
- hyperfx-ai/marketing-skills★ 112same name
- adityaarsharma/youtube-marketing-skills★ 58same name
- oakreader/oakreader★ 46same name
- ericmjl/skills★ 41same name
- tuan3w/obsidian-vault-agent★ 39same name
- s-choung/Research-Skills★ 36same name
- tubeagentkit/youtube-mcp★ 24same name
25 more not listed
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository11 other files it gives its agents.
- ask-hrSkill
- brainstormSkill
- contentSkill
- emailSkill
- factcheckSkill
- find-skillSkill
- hatchetSkill
- meeting-prepSkill
- postreadySkill
- postSkill
- promptSkill
The file
We split this skill into the rules it gives an agent
7 rules. 0 are common, 7 are distinctive, and 0 are very specific to this project.
0
Common
Most of the 28 edited versions we read keep these.
7
Distinctive
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 18
Before searching, think about what the user actually wants:
- Distinctive: line 20
Do they want a quick overview or a deep dive?
- Distinctive: line 24
Use this to guide your search strategy and final pick.
- Distinctive: line 58
Use this format exactly:
- Distinctive: line 74
Keep the "also worth watching" descriptions short — a phrase, not a sentence.
- Distinctive: line 81
Don't narrate the search process.
- Distinctive: line 82
If the topic is ambiguous, interpret and go — don't ask for clarification first.
“Common” is measured: kept by at least 6 in 10 of the 28 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
YouTube Search Skill
Find the single best YouTube video for what the user needs, plus 3–4 solid alternatives. No filters, no menus — just good judgment.
Workflow
Step 1 — Interpret the intent
Before searching, think about what the user actually wants:
- Are they a beginner or do they seem technical?
- Do they want a quick overview or a deep dive?
- Is this a how-to, an explainer, a review, or something else?
- What would the ideal video look like for this person?
Use this to guide your search strategy and final pick.
Step 2 — Run multiple searches
Generate 3–4 distinct search angles and run them via Zapier:youtube_find_video. Each search should approach the topic differently — vary the framing, vocabulary, and specificity. Don't just repeat the same query with minor tweaks.
Example for "what is MCP":
"Model Context Protocol explained"— direct explainer angle"MCP AI agents tutorial beginner"— hands-on/beginner angle"why MCP matters AI tools 2025"— strategic/why-it-matters angle
For each search call use:
max_results: 10
order: relevance
published_after: <12 months ago in ISO 8601>
output_hint: title, video URL, duration, published date, channel nameStep 3 — Pool and evaluate candidates
Combine all results across searches (deduplicating by video ID). For each candidate assess:
- Relevance: Does it directly address what the user needs?
- Quality signals: Channel credibility, title clarity, not clickbait
- Duration fit: Does the length match the likely intent? (Quick explainer vs. deep dive)
- Recency: More recent is generally better
- Avoid: Shorts under 2 min, tangentially related content, duplicate channels (max 1 per channel in final output)
Step 4 — Pick the best and select alternatives
Choose the single video you'd genuinely recommend if a friend asked. Then pick 3–4 alternatives that offer something meaningfully different — different depth, angle, channel style, or length. Don't just list the next highest-ranked; make sure each alternative has a distinct reason to exist.
Step 5 — Present results
Use this format exactly:
Recommended Title · Channel · Duration One sentence on why this is the best pick for what they asked.
Also worth watching
- Title · Channel · Duration — one phrase on what makes this different
- Title · Channel · Duration — one phrase on what makes this different
- Title · Channel · Duration — one phrase on what makes this different
- Title · Channel · Duration — one phrase on what makes this different (optional)
Keep the "also worth watching" descriptions short — a phrase, not a sentence. The recommendation gets the full sentence; the alternatives just need a differentiating hook.
Tone
- Confident. You picked one. Don't hedge with "this might be good" — say why it's the best fit.
- Don't narrate the search process.
- If the topic is ambiguous, interpret and go — don't ask for clarification first.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

