Write Workplace Messages That Sound Like You

Where this fits

You've just finished a tense negotiation over Slack and need to send a follow-up that clarifies next steps without sounding like a robot. Or you've drafted an email to a difficult stakeholder but can't tell if it's too passive or too aggressive. The workplace-message-writer skill from the OpenAgentSkill ecosystem steps in at that exact moment—when you have the raw material and intent but need a human-sounding, send-ready version.

Why agents benefit

• Detects intent without guesswork: Triggers on explicit phrases like "polish this" or "how do I say this," and asks a single clarifying question only when the user's intent is ambiguous.
• Preserves the user's voice: Retains common words, tone, formality, and even the user's characteristic bluntness—no cookie-cutter corporate speak.
• Prioritizes truth over polish: Never invents facts, alters commitments, or shifts responsibility, even if that means leaving a slightly awkward sentence in place.
• Knows when to stay quiet: Correctly refuses to act on background context, strategy discussions, or long-form documents, preventing unwanted rewrites.
• Asks only what matters: When missing info would change tone, action, or timeline, it asks one precise question instead of interrogating the user.

Practical scenarios

Following up after a heated meeting

You paste a rough note about action items and ownership. The skill turns it into a neutral, actionable message that keeps your tone firm but not confrontational, and highlights who does what by when.

Replying to a passive-aggressive email

You have a draft reply that's too emotional. The skill rewrites it to be direct and professional, keeping your core points but removing the sting—without making you sound like a customer-service bot.

Sending a status update to a busy manager

You provide bullet points of what's done, blocked, and next. The skill composes a concise message with clear asks, so your manager can act immediately without digging for context.

Add it to your agent workflow

Install the skill into your agent's skill directory:

npx skills add jeffy-Peng/jeffy-skills --skill workplace-message-writer

Then, in your agent, you can simply say:

"Here's what I want to tell my teammate: 'The report is late again. Need it by Friday.' Can you make this sound less harsh but still clear?"

The skill will produce something like: "Hey, just checking in on the report—could we target Friday for completion? Let me know if you need anything from me." It keeps your directness while softening the edge.

Compare before adopting

Before committing, compare this skill against your existing message-generation approach. Look at: quality signals—does the output actually sound like you, or does it default to generic corporate language? Maintenance freshness—is the repo actively updated to handle new workplace channels and tone variations? Alternatives—generic prompt templates may work for simple cases but often miss the nuance of preserving personal voice. Workflow fit—does the skill's trigger logic align with how your users actually ask for help, or does it require too much setup?

Why it is worth tracking

With 20 stars and a focused scope, this skill is a solid example of a niche, high-utility agent capability. Its clear trigger rules and refusal behavior make it easy to test and integrate. Track it if you build agents that handle communication tasks—evaluate it when you need a reliable way to draft messages that respect user intent and voice, and watch for updates that expand its handling of different cultural or organizational contexts. The skill's design philosophy—prioritizing authenticity over AI polish—is a signal worth noting for any agent that interacts with humans on behalf of users.