agentleFS
fahdAghribifahdAghribi/youtube-channel-skills/skills/youtube-channel-audit/SKILL.md

youtube-channel-audit

Skill2 starsChanged 13 days agoMITView on GitHub

Audit a YouTube channel end to end (positioning, audience, pillars, branding, home tab, playlists, packaging consistency, upload history) and write the shared channel profile every other skill reads. Use when starting with a new or existing channel, or when growth stalls. Not for diagnosing one video's numbers (use youtube-analytics-reviewer).

2 other projects use a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

3 rules. About 0 are common, 2 are distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈2

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 1

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Before a rebrand, niche change, or language switch

    line 15
  2. Very specific:

    Load the profile: If ../../references/channel-profile.md has filled: yes, start from it and only ask about gaps.

    line 25
  3. Distinctive:

    Never recommend deleting old videos to "reset" the channel: there is no such mechanism.

    line 61

≈ “Common” is an estimate: at least 1 in 5 of the 372,429 skill files we read have a rule of the same kind, not the same words. “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

YouTube Channel Audit

Score the channel against what makes YouTube recommend it, return a prioritised fix list, and fill ../../references/channel-profile.md so every later task starts with context.

When to use

  • "Audit my channel", "why is my channel not growing", "set up my channel"
  • First run of the bundle (profile says filled: no)
  • Before a rebrand, niche change, or language switch

Input

  • Channel URL or @handle (or a description if the channel does not exist yet)
  • Goal: subscribers, watch time for monetisation (YPP), leads/sales, authority
  • Optional: Studio exports (Overview, Content tab CSV), the creator's time budget per week

Steps

  1. Load the profile. If ../../references/channel-profile.md has filled: yes, start from it and only ask about gaps.
  2. Collect data (tier-dependent):
    • Tier 1+: python3 scripts/research_outliers.py @handle --limit 50 --min 0 gives the last 50 uploads with views, duration and the outlier ratio vs the channel median.
    • Tier 2 (own channel): python3 scripts/analytics_report.py top --days 90 adds watch time, % viewed and subscribers per video.
    • Tier 0: ask the user to paste the Content tab or a list of the last 20 titles with views.
  3. Interview the 4 starred profile questions plus: who is the viewer, what do they search, which 3-5 channels do they also watch.
  4. Score the 8 areas below (0-3 each) using references/audit-checklist.md.
  5. Find the pattern in the outliers: which topics/formats/titles beat the median by 2x+? Which are flops? This is the strongest evidence of what the audience wants from this channel.
  6. Prioritise: return at most 7 fixes, ordered by impact ÷ effort. Each fix has a concrete next action and which skill executes it.
  7. Write the profile. Show the filled channel-profile.md as a diff, ask for approval, then save it with filled: yes and today's date in updated:.

The 8 areas

#AreaWhat "3" looks like
1PositioningOne-line promise a stranger understands in 5 s; videos all serve it
2Audience fitClear primary viewer; titles use their words
3Pillars2-4 pillars, each with at least one proven outlier
4PackagingRecognisable thumbnail style, titles complement thumbnails, no repetition
5Retention craftHooks start in the first line, avg % viewed at or above channel norm
6CadenceSustainable schedule kept for 8+ weeks
7Channel pageBanner within safe area, handle, About with keywords + links, trailer for non-subs, sections by pillar, playlists
8ConversionEnd screens, pinned comment, playlists chaining videos; links that match the goal

Output

  1. Scorecard table (area, score, evidence, fix).
  2. What the outliers say (3-5 bullets, each with a video and its ratio).
  3. Top fixes (≤ 7), each: action, why, effort (S/M/L), skill to run next.
  4. Profile diff awaiting approval.

Hard rules

Global voice rules: see root SKILL.md §Voice rules.

  • Outlier ratios and numbers come from script output or the user. Never estimate views.
  • Never recommend deleting old videos to "reset" the channel: there is no such mechanism.
  • Recommend a niche change only with evidence (outliers in the new topic, or the user's explicit goal).

Files

  • references/audit-checklist.md — scoring criteria per area
  • ../../references/channel-profile.md — the profile this skill writes
  • ../../references/algorithm-heuristics.md — why each area matters
  • ../../references/metadata-limits.md — banner, avatar and trailer specs

Related skills

  • youtube-idea-research — turn the outlier patterns into a backlog
  • youtube-content-planner — turn pillars + cadence into a calendar
  • youtube-title-thumbnail — fix packaging issues found here
  • youtube-analytics-reviewer — deep dive on individual videos

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.