mohitagw15856/pm-claude-skills / plugins/pm-cowork/skills/reply-in-their-tone
reply-in-their-tone
VerifiedDraft email replies that match the sender's register β formality, length, directness, and emoji-tolerance read from their message, so the reply lands as native instead of off-key. Use when asked reply to this email, draft a response that doesn't sound stiff, match their tone, or answer this without sounding like a robot. Produces the tone read of the incoming message, the reply drafted in that register, and the adjustment knobs.
| File | Kind | Size |
|---|---|---|
| SKILL.md | skill md | 4.2 KB |
SKILL.md
View on GitHub---
name: reply-in-their-tone
description: "Draft email replies that match the sender's register β formality, length, directness, and emoji-tolerance read from their message, so the reply lands as native instead of off-key. Use when asked reply to this email, draft a response that doesn't sound stiff, match their tone, or answer this without sounding like a robot. Produces the tone read of the incoming message, the reply drafted in that register, and the adjustment knobs."
---
# Reply In Their Tone Skill
Replies fail on register more than content: the three-line casual ask answered with four formal paragraphs, the careful formal request answered with "hey! sure thing π". People read tone mismatch as social information β too formal reads as cold or annoyed, too casual reads as unserious. This skill reads the incoming message's register first (formality, length, directness, warmth markers), then drafts the reply *in that register* β content correct, tone native.
## What This Skill Produces
- **The tone read** β the sender's register across four dials: formality, length, directness, warmth/emoji
- **The reply** β content as instructed, register as read
- **The knobs** β one-notch-warmer and one-notch-more-formal variants when the call is close
- **The mismatch flag** β when the *right* move is deliberately not matching (escalations, boundary-setting), said explicitly
## Required Inputs
Ask for these if not provided:
- **The incoming email** β verbatim; the tone read works on the actual words
- **What the reply must accomplish** β the yes/no/ask/push-back content, decided by the user
- **The relationship** β first contact, colleague, boss, customer? Matching runs within the floor the relationship sets (never below professional-warm with strangers)
## Framework: The Four Dials
1. **Formality:** greeting style ("Dear Dr." vs "hey"), sign-off, contractions, titles. Match it β one notch *warmer* is safe when unsure; one notch stiffer never is.
2. **Length:** a three-line email wants a three-to-five-line reply. Answering short with long reads as lecture; long with short reads as brush-off. Match the sender's investment, Β±50%.
3. **Directness:** some write asks in line one; others cushion. Direct senders get the answer in sentence one; cushioners get one line of cushion first β not three.
4. **Warmth markers:** exclamation points, emoji, small talk. Reflect at *most* their level, never above; zero-emoji senders get zero emoji, always.
5. **The deliberate mismatch:** boundary-setting, escalation, and bad-news replies sometimes *should* run more formal than the incoming β a register shift is itself a message. When content calls for it, the skill says so and shifts on purpose rather than by accident.
## Output Format
# Reply: [to whom, accomplishing what]
## The Tone Read
Formality: [level, evidence] Β· Length: [theirs] Β· Directness: [direct/cushioned] Β· Warmth: [markers present]
## The Draft
[The reply, register-matched, ready to send]
## Knobs
[One notch warmer: the changed lines Β· one notch more formal: the changed lines]
[If mismatch was chosen: why, stated]
## Quality Checks
- [ ] Every dial cites evidence from the incoming message
- [ ] Reply length is within Β±50% of the incoming
- [ ] Warmth markers never exceed the sender's
- [ ] The content the user specified survives intact β tone-matching never softens the actual answer
- [ ] Deliberate mismatches are flagged as choices, not accidents
## Anti-Patterns
- [ ] Do not default to formal β stiffness reads as coldness, and coldness reads as a message
- [ ] Do not out-emoji the sender β one notch above reads as performative
- [ ] Do not cushion a direct sender's answer β they told you their preference in their own email
- [ ] Do not let tone-matching dilute a "no" β register is the wrapper, never the content
- [ ] Do not mimic idiosyncrasies (their typos, their catchphrases) β matching register isn't impersonation
## Example Trigger Phrases
- "Reply to this email."
- "Draft a response that doesn't sound stiff."
- "Answer this without sounding like a robot."
Run this skill in your browser: https://mohitagw15856.github.io/pm-claude-skills/skill/reply-in-their-tone.html
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| slack-gif-creatoranthropics/skillsKnowledge and utilities for creating animated GIFs optimized for Slack. Provides constraints, validation tools, and animation concepts. Use when users request animated GIFs for Slack like "make me a GIF of X doing Y for Slack." | Stars: 198k | Updated | Claude CodeCodex | Verified | Install |
| ponytailDietrichGebert/ponytailLazy senior dev mode for any coding task: the smallest change that fully solves it, and a reply a busy human understands in one read. | Stars: 196k | Updated | Claude CodeCodex | Verified | Install |
| ponytail-debtDietrichGebert/ponytailHarvest every shortcut: comment into one debt ledger, so deferrals get tracked instead of forgotten. One-shot report. | Stars: 196k | Updated | Claude CodeCodex | Verified | Install |
| ponytail-debtDietrichGebert/ponytailList every `shortcut:` comment (and older `ponytail:` ones) as a debt ledger. One-shot report, changes nothing. Use for "ponytail debt", "what did ponytail defer", "list the shortcuts", /ponytail-debt. | Stars: 196k | Updated | Claude CodeCodex | Verified | Install |