agentleFS
Sign inStart free
ahmedsakriahmedsakri/youtube-agent-pro/skills/yt-analytics/SKILL.md

yt-analytics

Skill1 starsChanged 36 days agoMITView on GitHub

Analyze YouTube Studio data for Shorts or long-form videos, distinguish traffic surfaces, and identify evidence-backed performance hypotheses. Use for "read my analytics", "which video worked", "my views dropped", "what's my CTR telling me", or a pasted Studio export.

1 other project uses a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

7 rules. About 0 are common, 7 are distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈7

    Distinctive · estimate

    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:

    Read the shared operating guide for profile lookup, helper paths, capability limits, and cross-skill routing before using this workflow.

    line 12
  2. Distinctive:

    Start by identifying the format, traffic sources, publication age, and comparable channel baseline.

    line 15
  3. Distinctive:

    Do not apply long-form CTR thresholds to Shorts-feed views.

    line 17
  4. Distinctive:

    Use impressions, impression click-through rate, average view duration, and actual video length.

    line 26
  5. Distinctive:

    Use the Shorts mode to keep the long-form funnel heuristic out of the report.

    line 44
  6. Distinctive:

    If feed exposure is limited, gather more evidence before blaming SEO or posting frequency.

    line 50
  7. Distinctive:

    Use a channel-specific baseline when available.

    line 63

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 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

yt-analytics

Read the shared operating guide for profile lookup, helper paths, capability limits, and cross-skill routing before using this workflow.

Start by identifying the format, traffic sources, publication age, and comparable channel baseline. A low view count alone cannot tell you whether the cause was topic, packaging, retention, or timing. Do not apply long-form CTR thresholds to Shorts-feed views.

Long-form Browse or Search analysis

python3 "$YT_SKILL_DIR/funnel.py" --impressions 42000 --ctr 3.1 --avd 2:10 --length 9:40
python3 "$YT_SKILL_DIR/funnel.py" stats.csv

Use impressions, impression click-through rate, average view duration, and actual video length. An Overview export may lack runtime or traffic-source detail; obtain the missing fields rather than substituting watch time, views, or AVD for them. Inspect the output for incomplete inputs.

The helper's bands suggest what to inspect first, not a proven single cause:

  • Limited impressions: compare video age, topic demand, traffic source, and the channel's prior comparable uploads. /youtube-agent-pro:yt-idea can help test a topic hypothesis.
  • Low CTR for the same surface: examine the title/thumbnail promise together through /youtube-agent-pro:yt-package and /youtube-agent-pro:yt-thumbnail.
  • Weak retention for comparable videos: examine the curve and footage using /youtube-agent-pro:yt-retention, rather than assuming a title rewrite fixes it.

Shorts-feed analysis

python3 "$YT_SKILL_DIR/funnel.py" shorts.csv --format shorts

Use the Shorts mode to keep the long-form funnel heuristic out of the report. Assess available shown in feed, stayed to watch / viewed versus swiped away, engaged views, average view duration, average percentage viewed, retention curve, and repeat viewing. Note which metrics are missing and each denominator. Distinguish raw Shorts starts/replays from engaged views; do not compare raw view totals across a metric-definition change without accounting for it.

If feed exposure is limited, gather more evidence before blaming SEO or posting frequency. If viewers swipe away early, inspect the first frame and opening promise. If they stay but leave before the payoff, inspect pacing and payoff timing. Loops can produce percentage viewed above 100%; this is not automatically an error. Search/Browse packaging may matter for those surfaces, but is not an explanation for Shorts-feed performance by itself.

What to hand back

  • A compact metric table with format, source, date range, sample size, and missing inputs.
  • The best-supported hypothesis, evidence for it, and important alternative explanations.
  • One practical experiment and what metric would support or reject it.
  • For a batch, compare like-for-like cohorts rather than ranking by lifetime views alone.

Use a channel-specific baseline when available. Built-in CTR/APV thresholds are illustrative; never promise that the suggested change will increase distribution or guarantee views.

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.

Check your setup, free