agentleFS
utillayerutillayer/youtube-agent-skill/skills/yt-viral/SKILL.md

yt-viral

Skill10 starsChanged 30 days agoMITView on GitHub

Find what is working in a niche on YouTube by ranking each video against its own channel's normal views, then name the formula behind the title. Use for "what's working right now", "find outliers in my niche", "why did this blow up", competitor research, or building a swipe file.

5 other projects use a version of this, 4 of them edited.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

1 rule. 0 are common, 1 is distinctive, and 0 are very specific to this project.

  • 0

    Common

    Most of the 3 edited versions we read keep these.

  • 1

    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
  1. Distinctive:

    Public listings only: Never a logged-in session, never anyone's credentials.

    line 39

“Common” is measured: kept by at least 6 in 10 of the 3 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.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

yt-viral

Raw view counts measure channel size, not ideas. A video worth studying is one that beat its own channel's median by a wide margin. That is what this ranks.

python3 ${CLAUDE_SKILL_DIR}/swipe.py collected.json --min 2.0
yt-dlp --flat-playlist -J "https://www.youtube.com/@channel/videos" > a.json
python3 ${CLAUDE_SKILL_DIR}/swipe.py a.json b.json c.json        # yt-dlp output works as is

The tools live in this skill's folder. ${CLAUDE_SKILL_DIR} is that folder in Claude Code; anywhere it does not expand (claude.ai, Codex), use the folder this SKILL.md was loaded from. It reads the formulas from ../yt-script/, so install the two together.

Collecting

At least four videos per channel, or a median means nothing; the tool skips thinner channels and says so. Ten to thirty per channel is better. yt-dlp --flat-playlist -J on a channel's /videos page is quickest and its output goes straight in; a hand-made list works too:

[{"channel":"...","title":"...","views":412000,"url":"...","duration":613}]

Shorts and long videos are compared separately when durations are known, and a video with no view count is left out rather than counted as zero. If the tool left some out, mention it.

Public listings only. Never a logged-in session, never anyone's credentials.

Reading the result

The multiple is the signal. The formula label is a judgement about the title's words, matched against the 21 formulas - not a claim about why it worked. Say so.

What to hand back: the top five with their multiples and formulas, and the one thing they share. Then the harder part: which of them the creator could make this week, in their voice, with what they have.

A swipe file is for the shape, never the video. Remaking someone's outlier with the names swapped is the "mass-produced, repetitive" content YouTube's inauthentic-content policy demonetises. If asked for that, say so and offer the structure to fill with the creator's own material.

Before you hand it over

This skill drafts; you publish. End with the copy-ready block and one question: post it as is, or another pass?

Download now (ZIP)

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
What did you use it for? e.g. Reviewing pull requests in a Rails app

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.