Captures how a person or brand actually writes and turns it into reusable voice instructions every other content skill draws from. Use this before drafting any content for a new brand or client, when output keeps coming back sounding generic, when several writers need to sound like one, or when a voice needs defining without existing samples to learn from.
npx skills add https://github.com/cbrock84/headcount --skill brand-voice
The upstream dependency for every writing skill. Draft before this exists and it comes back sounding
like everyone else.
Collect ten to twenty real pieces the person considers representative — their best, not their most
recent. Then extract, with evidence:
voices are distinguished by rhythm before vocabulary.
Quote a real example for each. A voice guide with adjectives and no samples is unusable.
Build from a choice rather than a blank page. Offer archetypes — the practitioner showing their
work, the analyst, the contrarian, the teacher, the insider, the enthusiast — with a written
paragraph in each. Let them react. People cannot describe their voice but recognize it instantly.
Then interview around the choice: who are you writing for, what do you want them to feel, what would
you never say.
Produce two artifacts other skills read:
voice.md — the rules, each with a real example and a counter-example.about-me.md — background, expertise, recurring themes, and the stories available to draw on.Include a do-not list. Voice guides fail on what they permit, not what they prescribe.
Draft two paragraphs and read them to the person cold. If they say "close, but I wouldn't say it
that way," ask exactly what they would say — that correction is the most valuable data in the
process.
Take cbrock84/brand-voice from the repository into ~/.claude/skills for personal
use, or into .claude/skills inside a project.
The agent identifies a skill by the name field in its header. Two skills with the
same name cannot sit side by side — one of them will be ignored.