voice-editor

Edit AI-generated drafts to match a personal voice profile through six structured passes.

3|Updated Aug 8, 2026
One-click install
npx skills add https://github.com/Jose-Polanco-Oxte/Echos-Live-Music-Visualizer --skill voice-editor-jose-polanco-oxte
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: voice-editor
Source: https://github.com/Jose-Polanco-Oxte/Echos-Live-Music-Visualizer/tree/main/.agents/skills/writing-style/sub-skills/voice-editor
Command: npx skills add https://github.com/Jose-Polanco-Oxte/Echos-Live-Music-Visualizer --skill voice-editor-jose-polanco-oxte

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? AI-generated content often sounds generic, formal, and indistinguishable from anyone else's writing. This Skill transforms that generic output into authentic, voice-matched content by applying a systematic editing process against your personal VOICE.md profile. ## Core Features & Use Cases - Six-Pass Editing Workflow: Structure check, voice alignment, slop removal, specificity injection, rhythm and flow, and a final authenticity check. - Multiple Editing Modes: Choose light, standard, heavy, or rescue mode depending on how far the draft is from your voice, with expected edit percentages and time estimates for each. - Slop and AI-Tell Removal: Detects and eliminates forbidden phrases, staccato fragments, comparator sentences, and uniform sentence rhythms. - Use Case: You generated a blog post with AI and it reads like a corporate memo. Provide the draft and your VOICE.md profile, and receive an edited version with annotated changes, an edit summary, and suggestions for improving your voice profile. ## Quick Start Edit this draft to match my voice profile at resources/voice/VOICE.md and show me a summary of the changes you made.

Frequently Asked Questions about voice-editor

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I make AI-generated content sound like my own writing?

Provide the draft content along with a VOICE.md voice profile, and the skill runs a six-pass editing process covering structure, voice alignment, slop removal, specificity, rhythm, and authenticity. It outputs edited content with annotated changes and an edit summary.

What is a VOICE.md file and do I need one for voice editing?

VOICE.md is a voice profile document describing your sentence patterns, vocabulary, tone markers, and structural habits. The skill expects it at resources/voice/VOICE.md or a project-level path, and it is used as the reference for all voice alignment decisions.

How much of AI output should I expect to edit?

Expect to edit 30-40% of AI output even with a good style guide. Editing more than 50% suggests your style guide or prompts need refinement, while editing less than 20% may mean you are accepting overly generic content.

What editing modes are available for different draft quality levels?

Four modes exist: light (10-20% edits for near-voice content), standard (30-40% for typical AI output), heavy (50-70% for generic output), and rescue (extract core ideas and rewrite from scratch when content is fundamentally off but has good bones).

When should I rewrite instead of editing AI-generated content?

Use rescue mode when the content is fundamentally off but contains good ideas. Extract the core structure and specific facts, rewrite from scratch using the voice profile, then run standard editing passes on the new draft.