yt-voice
Skill1 starsChanged 38 days agoMITView on GitHub
Build or refresh an optional creator voice profile from their own transcripts and stated audience, vocabulary, and format preferences. Use for "set up my voice", "write my voice.md", "make the scripts sound like me", "my scripts don't sound like me", or first-time setup of the YouTube pack.
1 other project uses a version of this.
Adopted by
Most recently taken up by
- gladosAagman/youtube-agent-skilltheir own version, first seen Nov 15, 2026
Other versions1 project uses a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 1
- gladosAagman/youtube-agent-skill★ 1same 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-seoSkill
- yt-seriesSkill
- yt-shortsSkill
- ytSkill
- yt-sponsorSkill
- yt-thumbnailSkill
- yt-trendSkill
- yt-viralSkill
The file
We split this skill into the rules it gives an agent
1 rule. About 0 are common, 1 is distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈1
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
- Distinctive: line 12
Read the shared operating guide for profile lookup, helper paths, capability limits, and cross-skill routing before using this workflow.
≈ “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-voice
Read the shared operating guide for profile lookup, helper paths, capability limits, and cross-skill routing before using this workflow.
Build a reusable voice profile from how the creator actually talks. Other skills can use it when
available, but do not require it before doing useful work. Save to the user's chosen location, or
~/.claude/youtube/voice.md. The canonical template is assets/voice.md,
which ships with this skill. Use this template for new profiles rather than maintaining a second copy.
A tool lives here and it runs. It measures the things you cannot hear about yourself.
python3 "$YT_SKILL_DIR/voiceprint.py" best-video.txt second.srt third.txtThe move
- Get three transcripts of the creator's own best-performing videos - the words they actually
said, not their scripts. Pasted,
.srt,.vtt, or ask them to export captions from Studio. If they have none (new channel), fall back to the interview below. - Run
voiceprint.pyacross all of them. It returns mean sentence length, filler rate, the signature words they lean on, and the hype words they already use. These are measurements, not grades - a long sentence is not worse than a short one, it is just a thing the scripts must respect. - Fill the six sections of the voice template from the measurements plus a short interview:
- Who I am talking to - one named person, not "creators".
- How I actually talk - from the mean sentence length and signature words.
- Words I never use - the hype words to ban (the creator chooses from the ones voiceprint found).
- Words I do use - the signature words. Keep them; they are what makes a script sound like them.
- What I will not claim - numbers, results, tools they have not got. The skills refuse these.
- My format - length, on-camera or not, intro or not, whether they swear.
- Write the profile to the requested path, or
~/.claude/youtube/voice.md. Preserve an existing profile's useful details and incorporate the requested updates. Mark inferred audience/tone details so the creator can correct them.
New channel, no transcripts
Interview instead: who is the one viewer, three phrases that are theirs, three words that are not, their format. Thinner than a voiceprint, but enough to start - and refresh it from real transcripts after three uploads.
What to hand back
- the saved or attachable voice profile, with its actual location if written
- the voiceprint measurements that informed it
- one line: the one section to tighten after their next three videos
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.