Voice memory
A forkable content system for AI agents that scouts signals, drafts in your voice, and edits out generic AI writing.
npx -y skills add AlexHoudz/content-os-starter-kit --skill voice-memoryAssembled 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.
What its author says it does
Copied from the file, not written here
Build and maintain a reusable writing voice from past posts, examples, bans, and audience notes. Use when setting up a content workflow, translating someone's style into a repeatable guide, updating a voice file after feedback, or checking whether enough inputs exist to stop generic AI writing.
SKILL.md
1.7 KB, 335 tokens by cl100k_base, as published. Nobody here has run it
Voice Memory
Overview
Turn scattered writing examples and preferences into a usable voice system.
This skill should help the user sound more like themselves and less like a generic AI writer.
Core workflow
1. Gather real examples
Use:
- good past posts
- bad drafts
- edits the user made
- phrases they use often
- phrases they hate
- audience notes
Prefer real writing over abstract brand language.
2. Build the voice files
Update or create:
templates/VOICE.mdtemplates/DO_NOT_SAY.mdtemplates/EXAMPLES.mdtemplates/POST_SCORECARD.md
3. Pull out patterns
Look for:
- sentence rhythm
- tone
- recurring phrasing
- typical hooks
- what sounds fake
- what gets cut out often
4. Keep it practical
The output should help future drafting. Do not write a beautiful brand manifesto that no one can use.
Good output
A good voice setup should tell a drafter:
- what this person sounds like
- what to avoid
- what examples matter most
- how to catch bad drafts early
Bad output
Reject these patterns:
- vague adjectives only
- empty brand language
- generic audience descriptions
- rules that are too abstract to apply
Update rule
When the user gives real feedback, update the templates. Treat strong edits as signal.
Resources
- Read
references/voice-patterns.mdto extract reusable voice traits. - Read
references/common-failures.mdto catch generic AI patterns.