Retention Analyzer

Yt retention

by Jakeschincariol · MIT license · ★ 1,136 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".

---
name: yt-retention
description: >-
  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".
---

# yt-retention

The retention graph is the only honest feedback YouTube gives you. Almost nobody exports it.

```bash
python3 retention.py retention.csv --duration 600
python3 retention.py retention.csv --transcript transcript.srt
```

Overview

The retention graph is the only honest feedback YouTube gives you. Almost nobody exports it.

Manual install (npx, ZIP)

Claude Codenpx skills add Jakeschincariol/youtube-agent-skill --skill yt-retention -g -a claude-code

Claude Code (degit)npx degit Jakeschincariol/youtube-agent-skill/skills/yt-retention#main ~/.claude/skills/yt-retention

Claude.ai / desktopZip the folder and upload it under Settings → Capabilities → Skills.

Codex, Cursor, other agentsPoint them at SKILL.md. The other files are plain text.

Connects with Agent Alley MCP · agentalley.io/mcp

Files included

SKILL.md1.7 KB

Read SKILL.md here

Jakeschincariol/main1 file shown

SKILL.md1.7 KB

Browse every file on GitHub↗

SKILL.md

Show the full text42 lines

yt-retention/SKILL.md42 lines · 1.7 KB

PreviewCode

Outline

RawView on GitHub

yt-retention

The retention graph is the only honest feedback YouTube gives you. Almost nobody exports it.

python3 retention.py retention.csv --duration 600
python3 retention.py retention.csv --transcript transcript.srt

Getting the file: Studio -> a video -> Analytics -> Engagement -> the audience-retention chart -> the download icon -> "Audience retention".

Three different problems

HOOK LEAK - what is lost in the first 30 seconds. Under 25% is healthy. This is always the first thing to fix and it is always the first fifteen seconds of script, never the edit.
CLIFFS - single steep drops. A cliff is a moment: a topic change with no signposting, a sponsor read, a long setup. With --transcript the tool prints what was being said there, which is what makes the report actionable instead of interesting.
SLIDE - the steady bleed across the middle. A flat slide is pacing. The fix is cutting, not rewriting.

What to hand back

Name the single biggest leak and one change for it. Not a list of five. Then, only if asked, the rest. And if the hook leak is healthy and the slide is flat, say the video is fine and the problem is packaging - send them to /yt-package.

The gate

Nothing here publishes. This skill writes and you publish. Every output ends in a block the user copies, and the last line of every run is the question: ship it, or change it?

1	---
2	name: yt-retention
3	description: >-
4	Read a YouTube Studio audience-retention export and find where viewers
5	actually leave, then say what to change. Use for "why do people stop
6	watching", "my retention is bad", a pasted retention chart or CSV, or "fix
7	my pacing".
8	---
9
10	# yt-retention
11
12	The retention graph is the only honest feedback YouTube gives you. Almost nobody exports it.
13
14	```bash
15	python3 retention.py retention.csv --duration 600
16	python3 retention.py retention.csv --transcript transcript.srt
17	```
18
19	Getting the file: Studio -> a video -> Analytics -> Engagement -> the audience-retention chart ->
20	the download icon -> "Audience retention".
21
22	## Three different problems
23
24	- HOOK LEAK - what is lost in the first 30 seconds. Under 25% is healthy. This is always the
25	first thing to fix and it is always the first fifteen seconds of script, never the edit.
26	- CLIFFS - single steep drops. A cliff is a moment: a topic change with no signposting, a
27	sponsor read, a long setup. With `--transcript` the tool prints what was being said there, which
28	is what makes the report actionable instead of interesting.
29	- SLIDE - the steady bleed across the middle. A flat slide is pacing. The fix is cutting, not
30	rewriting.
31
32	## What to hand back
33
34	Name the single biggest leak and one change for it. Not a list of five. Then, only if asked, the
35	rest. And if the hook leak is healthy and the slide is flat, say the video is fine and the problem
36	is packaging - send them to `/yt-package`.
37
38	## The gate
39
40	Nothing here publishes. This skill writes and you publish. Every output ends in a block the user
41	copies, and the last line of every run is the question: ship it, or change it?
42