feat: add voice calibration from writing samples#64
Open
mvanhorn wants to merge 1 commit intoblader:mainfrom
Open
feat: add voice calibration from writing samples#64mvanhorn wants to merge 1 commit intoblader:mainfrom
mvanhorn wants to merge 1 commit intoblader:mainfrom
Conversation
Add optional Voice Calibration section to SKILL.md. When users provide a sample of their own writing, the skill analyzes their sentence rhythm, word choices, and quirks before rewriting - producing output that sounds like them instead of generic clean human. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
Adds an optional Voice Calibration section to SKILL.md. When users provide a sample of their own writing, the skill analyzes their sentence rhythm, word choices, and quirks before rewriting, so the output sounds like them instead of generic clean text.
Why this matters
The humanizer removes AI patterns well, but the result is one-size-fits-all. Everyone gets the same "clean human" voice. Users on TikTok and Reddit are excited about removing AI tells, but the natural follow-up is wanting output that sounds like their writing, not just "not AI."
Changes
SKILL.md: Added## Voice Calibration (Optional)section between "Your Task" and "PERSONALITY AND SOUL". Instructs Claude to analyze a user's writing sample for sentence length, word choice, paragraph structure, punctuation habits, and verbal tics before rewriting. Falls back to the existing personality section when no sample is provided. Version bumped to 2.4.0.README.md: Added usage example for voice calibration under the Usage section. Updated version history.39 lines added, 1 changed. No structural changes to the repo.
Testing
Tested by invoking
/humanizerwith a writing sample and verifying Claude analyzed the sample's characteristics before rewriting AI-heavy text. The output matched the sample's casual tone and short sentence style instead of producing the default clean-but-generic voice.This contribution was developed with AI assistance (Claude Code).