How do I install the Voice Fingerprint skill?

Run npx skillmds@latest add tchr-dev/voice-fingerprint in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.

Is Voice Fingerprint safe to use?

This skill has not completed SkillMD's automated safety review yet. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with Voice Fingerprint?

This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.

Is Voice Fingerprint free to use?

Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.

Who published Voice Fingerprint?

tchr-dev (@tchr-dev) published this skill. Their other Agent Skills are listed on their SkillMD profile.

Voice Fingerprint by tchr-dev — Agent Skill

Name: Voice Fingerprint
Author: tchr-dev

Voice Fingerprint

Run the mechanical voice-fingerprint scanner across all chapters to identify outliers in sentence rhythm, vocabulary domain balance, dialogue ratio, abstract noun density, and similar metrics. No LLM call — just runs the Python script. Use during revision to find chapters that drift from the novel's voice.

tchr-dev e0cd4f2 1.7 KB Updated Sep 17, 2026

File contents

Voice fingerprint

This skill runs scripts/voice_fingerprint.py and presents the result.

Procedure

Set the env var so the script targets the right novel:

AUTONOVEL_NOVEL_DIR=<novel-dir> python scripts/voice_fingerprint.py

Output is written to <novel-dir>/edit_logs/voice_fingerprint.json and a summary table is printed.

Important caveat

The vocabulary wells in scripts/voice_fingerprint.py (WELL_MUSICAL, WELL_TRADE, WELL_BODY) are defaults from "The Second Son of the House of Bells." For a different novel, these need to be replaced with vocabulary domains relevant to the new world.

If the vocab wells haven't been customised for this novel, edit the script first. Otherwise the well percentages are noise. The other metrics (sentence length CV, paragraph length, dialogue ratio, em-dash density, "the way" count, simile density, "He"-start %) are universal and worth running unchanged.

Reading the output

sentence_length_cv ≥ 0.4 is healthy. Below 0.3 is uniform-prose territory.
he_start_pct > 25% suggests sentence-start monotony.
the_way_count > 5 in a single chapter is leaning on a simile crutch.
Outliers (>1.5σ from mean) are the chapters where voice has drifted. Investigate them first.

After running, summarise outliers and recommend chapters to inspect with evaluate-chapter or hand to gen-brief for a voice-deviation revision.

Frequently asked questions