yt-endscreen
Skill1 starsChanged 37 days agoMITView on GitHub
Place end screens and cards where they keep the session going instead of ending it - the right next video, at the right second, read off the retention curve. Use for "end screen", "cards", "what plays next", "where do I put cards", "keep people on my channel", or routing viewers between videos.
In this repository25 other files it gives its agents.
- openai/AGENTS.mdAGENTS.md
- llms.txtllms.txt
- yt-analyticsSkill
- yt-auditSkill
- yt-briefSkill
- yt-chaptersSkill
- yt-collabSkill
- yt-commentSkill
- yt-editSkill
- yt-funnelSkill
- yt-ideaSkill
- yt-packageSkill
- yt-planSkill
- yt-repurposeSkill
- yt-retentionSkill
- yt-scriptSkill
- yt-seoSkill
- yt-seriesSkill
- yt-shortsSkill
- ytSkill
- yt-sponsorSkill
- yt-thumbnailSkill
- yt-trendSkill
- yt-viralSkill
- yt-voiceSkill
The file
We split this skill into the rules it gives an agent
5 rules. About 0 are common, 3 are distinctive, and 2 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.
2
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 13
Read the shared operating guide for profile lookup, helper paths, capability limits, and cross-skill routing before using this workflow.
- Very specific: line 26
Use the available voice profile or creator context (see the operating guide) for the close - the end screen has a spoken hand-off, and "one …
- Very specific: line 28
If there is a retention export, read it first (/youtube-agent-pro:yt-retention).
- Distinctive: line 51
Write the spoken hand-off (one ask) and the end-screen layout.
- Distinctive: line 52
Place 0-2 cards, each at a leak, each pointing somewhere relevant.
≈ “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.
Files
yt-endscreen
Read the shared operating guide for profile lookup, helper paths, capability limits, and cross-skill routing before using this workflow.
An end screen can guide viewers to a relevant next video. Choose by continuity and observed viewer behavior rather than assuming the newest upload is best. Verify current feature eligibility before giving implementation steps. Shorts need a Shorts-appropriate related-video or spoken CTA plan; do not assume long-form end screens/cards are available in the Shorts feed.
Pair it with /youtube-agent-pro:yt-retention - the retention curve says where the viewer still is at the end, and whether a card mid-video would catch a leak or cause one.
Before you place
- Use the available voice profile or creator context (see the operating guide) for the close - the end screen has a spoken hand-off, and "one ask, not three" is the rule from /youtube-agent-pro:yt-script applied here too.
- If there is a retention export, read it first (/youtube-agent-pro:yt-retention). The end-screen timing is only as good as knowing how many viewers are left to see it.
The rules
- End screen: the last 5-20 seconds. Point it at the video that continues the thought, not the newest upload. The best next video is the one a viewer who just finished this one would choose - a part 2, the thing this one referenced, the obvious next question.
- Lead with watch, not subscribe. One video element and one subscribe element is the ceiling. If the video did its job, the next-video click is worth more than the subscribe ask, and it earns the subscribe anyway.
- Hold something back for it. The last 20 seconds need a reason to stay - do not fully resolve before the end screen or there is nothing left to hold the viewer through it. A flat outro is where the final cliff is.
- Cards: at a leak, not at a peak. A card mid-video catches viewers who are about to leave anyway - place it just before a known drop-off from the retention curve, never over a peak where it only distracts a viewer who was staying.
The move
- From the retention curve, find the end-screen window (how many viewers remain) and any mid-video leak a card could catch.
- Pick the next video by continuity, not recency.
- Write the spoken hand-off (one ask) and the end-screen layout.
- Place 0-2 cards, each at a leak, each pointing somewhere relevant.
What to hand back
- the end-screen pick (which video, why), with the spoken hand-off line
- the end-screen layout (video + subscribe), and the seconds to place it
- the card placements (timecode + target), or "none - no leak worth a card"
- one line: the single routing change most likely to lift session time
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
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.