yt-seo
Skill1 starsChanged 38 days agoMITView on GitHub
Write accurate YouTube titles, descriptions, tags, and optional hashtags in the creator's language, grounded in the actual video and search intent. Use for video SEO, description drafts, metadata revisions, keyword choices, or search discovery for Shorts and long-form videos.
7 other projects use a version of this.
Adopted by
Most recently taken up by
- gladosAagman/youtube-agent-skilltheir own version, first seen Nov 15, 2026
- utillayer/youtube-agent-skilltheir own version, first seen Nov 8, 2026
- uplika/uplika-plugintheir own version, first seen Oct 25, 2026
- swayamdhawan/youtube-agenttheir own version, first seen Oct 15, 2026
- nessalazne/youtube-agent-skilltheir own version, first seen Sep 23, 2026
- Jakeschincariol/youtube-agent-skilltheir own version, first seen Sep 16, 2026
Other versions7 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 7
- Jakeschincariol/youtube-agent-skill★ 47same name
- utillayer/youtube-agent-skill★ 10same name
- nessalazne/youtube-agent-skill★ 5same name
- uplika/uplika-plugin★ 3same name
- gladosAagman/youtube-agent-skill★ 1same name
- swayamdhawan/youtube-agent★ 0same name
- tylerprogramming/claude-skills★ 0same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
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-endscreenSkill
- yt-funnelSkill
- yt-ideaSkill
- yt-packageSkill
- yt-planSkill
- yt-repurposeSkill
- yt-retentionSkill
- yt-scriptSkill
- 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
6 rules. 0 are common, 6 are distinctive, and 0 are very specific to this project.
0
Common
Most of the 7 edited versions we read keep these.
6
Distinctive
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
- Distinctive: line 12
Read the shared operating guide for profile lookup, helper paths, capability limits, and cross-skill routing before using this workflow.
- Distinctive: line 23
Read the supplied video, transcript, or reliable summary.
- Distinctive: line 25
Use the available creator profile or context.
- Distinctive: line 30
When adapting a batch, write a distinct description for each video from its own content.
- Distinctive: line 38
Opening description: write a clear, video-specific summary with the main phrase where it fits.
- Distinctive: line 40
Useful context: add only relevant details, verified credits, resources, or a concise call to action.
“Common” is measured: kept by at least 6 in 10 of the 7 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
yt-seo
Read the shared operating guide for profile lookup, helper paths, capability limits, and cross-skill routing before using this workflow. Read YouTube's verified metadata guidance for limits and surface-specific rules.
Help the right viewer recognize the actual video. Deliver the requested metadata without promising ranking, virality, or additional views. Metadata relevance matters in Search; Shorts-feed performance also depends on viewer response and personalization. Use the analytics skill when the request asks why reach is low, rather than diagnosing an SEO problem from views alone.
Ground the metadata
- Read the supplied video, transcript, or reliable summary. Identify the actual topic, names, episode/version, key moment, and payoff. A filename alone is not evidence of what happens.
- Use the available creator profile or context. Preserve the requested language and tone, including intentional code-switching. Do not require a profile to write useful metadata.
- Choose one or two central topics/search phrases from the content. Use actual channel search-term data or verified research if supplied; otherwise label keyword choices as relevance judgments, not measured search demand. Never invent search volumes, quotes, identities, dates, or results.
- When adapting a batch, write a distinct description for each video from its own content. Reuse verified channel links and credits only where appropriate; do not copy an unrelated episode's facts.
Title and description
- Title: name the topic or moment early, with an accurate reason to watch. Use relevant wording naturally. No unrelated celebrity, trend, or outcome claims. Respect a request to keep the current title; otherwise use the packaging skill for paired title/thumbnail options.
- Opening description: write a clear, video-specific summary with the main phrase where it fits. The first lines should explain the content before channel boilerplate; no repeated keyword lists.
- Useful context: add only relevant details, verified credits, resources, or a concise call to action. Do not pad the text to fill the field. Add chapters through the chapters skill only when meaningful and supported by actual timestamps.
- Length: validate the final title at no more than 100 characters and description at no more than 5,000 characters, including links and hashtags. These are platform limits; shorter preview targets are readability heuristics. Preview the opening lines when the relevant UI is available.
Tags and hashtags
Tags have a limited discovery role. Use a small, relevant set for common misspellings, alternate spellings, names, or abbreviations that need disambiguation; no fixed quota is required. Keep tags in the tag field rather than appending a keyword dump to the description.
Hashtags are optional, topic-specific discovery links. Do not claim they are ineffective or required for Shorts. If the user wants none, omit them. Otherwise prefer a few directly relevant hashtags; never add unrelated trending terms. More than 60 causes YouTube to ignore all hashtags on that content, but that limit is not a target. Distinguish hashtags from the separate tags field.
Format, language, and links
- Shorts: URLs in Shorts descriptions and comments are not clickable. Do not write "click the link in this description." Refer to a verified channel profile link or an available related-video link only if that destination is actually configured. A URL may remain as a reference without a false click instruction.
- Long-form: include verified useful links when requested; external clickable links depend on the channel's advanced-feature access. Do not assume access or manufacture destinations.
- Translations: provide separate localized title/description drafts when requested. Preserve names and meaning, and use natural search wording in each language. Do not concatenate several full translations into one title. YouTube supports localized title/description fields; English-oriented text heuristics are not a validated multilingual SEO score.
Deliver and check
Return the requested final title, description, tags, and optional hashtags as clearly separated fields, with a short explanation of the chosen topic terms if useful. Verify factual consistency, unique description, field lengths, language, links, and absence of stuffing. Flag missing source facts outside the publishable copy; do not slip placeholders into a supposedly finished description. This workflow drafts metadata. A Studio update requires the user's request and a separate available account tool under its own controls; verify the saved result before reporting a change.
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.

