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

yt-retention

Skill1 starsChanged 38 days agoMITView on GitHub

Read a YouTube Studio audience-retention export and find where viewers actually leave, then say what to change. Use for "why do people stop watching", "my retention is bad", a pasted retention chart or CSV, or "fix my pacing".

6 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

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

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈3

    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:

    Read an audience-retention export alongside the video length and, when available, a timestamped transcript.

    line 15
  3. Distinctive:

    Gradual decline: look for repeated material or delayed payoff.

    line 42

≈ “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-retention

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

Read an audience-retention export alongside the video length and, when available, a timestamped transcript. A drop suggests a place to inspect; it does not establish why viewers left.

python3 "$YT_SKILL_DIR/retention.py" retention.csv --duration 600 --axis percent
python3 "$YT_SKILL_DIR/retention.py" retention.csv --axis seconds --transcript transcript.srt
python3 "$YT_SKILL_DIR/retention.py" shorts-retention.csv --axis seconds --duration 30 --hook-seconds 3

--axis auto uses column labels to identify seconds, percent, or elapsed-time ratios. For an ambiguous unit-less axis, specify --axis seconds, --axis percent, or --axis fraction; percent and fraction axes need the actual runtime. Do not guess units from whether the largest value happens to be below 100. The helper also recognizes an audienceWatchRatio column for retention fractions; inspect its diagnostics and make other source units explicit before analysis.

Get the audience-retention data from Studio's video Analytics/Engagement export when available; if the UI does not offer that export, work from supplied chart data or a clearly labeled manual reading. Never fabricate samples from a chart you cannot read.

Patterns to inspect

  • Opening loss: compare the opening window with videos of similar length and format. The helper uses the first 10% of runtime, capped at 30 seconds, when duration is supplied. Set --hook-seconds for a different window; for a Short, inspect the first seconds alongside stayed-to-watch data and the opening footage.
  • Cliffs: inspect the exact transcript/footage at a sharp drop for topic changes, sponsor reads, slow setup, or an unmet promise. An edit or transition may matter as much as the script.
  • Gradual decline: look for repeated material or delayed payoff. Compare against the channel's own curves before calling normal attrition a failure.
  • Peaks and loops: a replay or confusing segment can both make a peak. Shorts can exceed 100% viewed; distinguish replay behavior from a causal claim about satisfaction.

What to hand back

Name the most consequential supported observation, its timecode, and one proposed change. Separate measured changes from interpretations. If the curve is healthy, use /youtube-agent-pro:yt-analytics for other evidence instead of declaring packaging at fault from retention alone. Respect requests for a full audit rather than withholding other findings.

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