Tone-Of-Voice From Email History Extractor

Point Cowork at the last month of your Gmail with substantive writing and it distills your tone of voice into memory.md so drafts sound like you.

Content Creationcontent-creationp3
by AgentVoltv1.0.0Published 1y ago1 install
Free to sign up · every skill included with AgentVolt Pro

Before & after

Same prompt, same agent — the only difference is the skill.

PromptMake my AI drafts sound more like how I actually write emails
Beforegeneric AI agent
I could analyze your writing style by looking at samples and extracting patterns. You'd need to share some examples of your emails with me. I could then identify things like word choices, sentence length, formality level, and common phrases you use. This might take some time to analyze manually. Would you be able to provide some recent emails that represent your typical style?
Afterwith Tone-Of-Voice From Email History Extractor
TONE-OF-VOICE PRINCIPLES (from 47 substantive emails, past 30 days)

OPENERS
• Start direct, no "Hi" or "hope you're well" preamble
• Lead with the ask or outcome: "We need to pause the feature launch"
• Use imperative mood when time-sensitive

VOCAB & PHRASING
• Contractions everywhere ("don't" not "do not"; "we've" not "we have")
• Short punchy sentences (avg 12 words, rarely exceed 20)
• Jargon preference: "unblock", "lift", "de-risk" over "enable", "improve", "reduce"

STRUCTURE
• Always end with a single clear next step, not multiple options
• Use numbered lists for complexity; bullets for 3 items or fewer
• No:
  - Long paragraphs (max 3 lines)
  - CC'd email threads (use Slack instead)
  - Apologies or over-hedging

TONE
• Confident but not dismissive
• Speed > polish (typos okay if urgent)
• Assume competence; no hand-holding

Saved to memory.md; auto-apply on future Claude drafts

About this skill


name: tone-of-voice-email-extractor description: Point Cowork at the last month of your Gmail with substantive writing and it distills your tone of voice into memory.md so drafts sound like you.

Tone-Of-Voice From Email History Extractor

What This Does

The creator tells Cowork to read the past month of substantive Gmail messages, extract the patterns that define his tone, and save them as writing style principles to memory.md. Subsequent drafts inherit that voice automatically.

What You Get

  • Gmail connector read
  • substantive-thread filter
  • per-pattern tone distillation
  • persistent save to memory.md
  • auto-apply on future drafts

Customize

  • lookback_window: how many days back to read
  • message_type_filter: only long threads or long replies
  • save_destination: memory.md or a custom skill
  • refresh_cadence: monthly or quarterly
  • keep_examples: preserve raw samples

Sample Output

Cowork reads a month of Gmail and appends 12 tone-of-voice principles to memory.md; the next email draft uses colloquial openers and short punchy closes matching the writers style.

Target Audience

Anyone who wants Claude to write in their voice without pasting samples every time

SKILL.md preview

SKILL.md
---
name: tone-of-voice-email-extractor
description: Use this skill when a writer’s tone of voice should be extracted from their own past email writing and saved as reusable style principles.
version: 1.0.0
category: Content Creation
author: AgentVolt
license: proprietary
tags:
  - content-creation
  - p3
---

# Tone-Of-Voice From Email History Extractor

Reads a window of a user's own substantive email writing and distills recurring tone-of-voice patterns into a saved style reference, so future drafts sound like the person without needing samples pasted in each time.

## When to use
- Wanting Claude-drafted emails or content to sound like the user's own natural voice

… (sign up to view the full skill)
Sign up to view, copy, and install the full skill

More content & writing skills

View all Content & Writing skills →