Voice calibration

Generate text in <your name>'s voice. Use when asked to "write as me", "in my voice", or when drafting anything sent or published under <your name>'s name.From its SKILL.md

Install

npx -y skills add runsagents/voice-calibration

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

SKILL.md

1.5 KB, 344 tokens by cl100k_base, as published. Nobody here has run it

Write as <you>

Calibrated <date> against <corpus size> of <your name>'s own messages (<years>). Voice anchor: <era> only, earlier eras are history, not the voice.

Contract

Only generate when explicitly asked. Never auto-send anything.
Require register (<professional | casual>) and language (<...>) before writing, ask if unspecified.
Output is always a draft for <your name>'s review.

Keep these (identity, not errors)

<quirk, e.g. "space before ? and !">
<quirk, e.g. "no trailing period on short lines">
<quirk, e.g. "mid-sentence language blends: ...">
<sentence rhythm, e.g. "median message is 5–6 words, no filler">
<openers/closers, e.g. "greeting + name + !; ends with a concrete next step">

Fix these silently (noise, not character)

<typos, autocorrect damage, dropped articles...>

Anti-patterns (instant voice-breakers)

<things you never say, corporate filler, wrong emoji, over-formality...>

Calibrated exemplars (approved by <you>)

<register, context>:

<full example message, approved verbatim>