keywords-seo
Skill2 starsChanged 49 days agoMITView on GitHub
Keywords, tags, description and the search gap. Triggered by "what keywords should I use", "optimise the SEO", "give me the tags", "how do people find me". In Spanish, "qué keywords uso", "optimiza el SEO", "dame las etiquetas", "cómo me encuentran".
In this repository23 other files it gives its agents.
- CLAUDE.mdCLAUDE.md
- llms.txtllms.txt
- analyse-channel-packagingSkill
- analyse-thumbnailsSkill
- channel-analyticsSkill
- channel-identitySkill
- channel-kickoffSkill
- competitor-boardSkill
- competitor-ideationSkill
- competitor-mappingSkill
- competitor-scanSkill
- image-generation-coreSkill
- image-refinementSkill
- likeness-preservationSkill
- memory-reviewSkill
- monetisationSkill
- packagingSkill
- script-writingSkill
- thumbnail-best-practicesSkill
- thumbnail-inspirationSkill
- transparencySkill
- youtube-banner-specSkill
- youtube-profile-specSkill
The file
Files
Keywords and SEO
WHEN
Choosing keywords and tags, writing or reviewing a description, or finding the search gap: what people search for that the channel does not answer yet.
EXECUTION ORDER
- Field truth first
These are the real terms people arrive through, and the videos suggesting yours. Not an estimate. Starting here rather than with the proxy is what separates this skill from a generic keyword generator.python3 tools/yt_search_terms.py --days 90 --type YT_SEARCH python3 tools/yt_search_terms.py --days 90 --type RELATED_VIDEO - Expand with the proxy
python3 tools/kw_research.py --kw "seed 1; seed 2" --deep--deepexpands with the alphabet: 26 more autocomplete calls, no API quota cost.heuristic. - Competitive gap
python3 tools/yt_search.py --query "candidate keyword"Costs 100 quota units per query. Cached for 24 h. Look at the median age of the results: an old top is a refresh opportunity. - Tags and description, crossing the real data from step 1 with the expansion from step 2.
SOURCES
Step 1 youtube_api — the only hard data in this skill. Steps 2 and 3
heuristic and derived. When giving a score, say it is your own.
OUTPUT
- Main keyword, with the evidence for it (preferably from step 1).
- 8-12 tags.
- A 120-180 word description with the keywords woven in naturally.
- Any keyword that comes only from the proxy and not from real traffic is flagged as such.
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.

