Cold Email Skill: Write Outreach That Gets Replies

Where this fits

You're building an AI sales assistant that drafts prospecting emails for your team's SDRs. You type "write a cold email to a logistics VP about our API," and the assistant returns a generic template with placeholder brackets. That's the moment you realize: writing cold email that actually gets replies is a craft—subject lines, opening hooks, personalization, multi-touch sequences. The cold-email skill codifies that craft so your agent can produce outreach that sounds like a sharp human, not a mail-merge bot.

Why agents benefit

• Context-aware research first: The skill automatically reads .agents/product-marketing.md (or legacy contexts) before drafting, grounding emails in real product positioning instead of guessing.
• Structured email anatomy: It breaks down subject lines, opening lines, body copy, CTAs, and personalization—so your agent follows a proven persuasion structure rather than free-forming.
• Multi-touch follow-up logic: Generates complete sequences (initial + follow-ups) with spacing and angle changes, solving the "nobody's replying" problem.
• Anti-template guardrails: Explicit instructions to avoid salesy clichés, vague claims, and robotic phrasing—outputs pass the "human test."
• Trigger-aware routing: The description covers common phrasings ("cold outreach," "prospecting email," "follow-up sequence"), so your agent knows when to invoke this skill vs. a warm-email skill.

Practical scenarios

Scenario 1: Drafting a first-touch prospecting email

Your agent needs to email a list of CTOs about a new dev tool. The skill reads your product context, extracts a specific pain point from the lead's LinkedIn, and writes a 3-line opener that references their recent tech stack change—then a concise value proposition and a low-friction CTA like "Worth a 10-min chat?"

Scenario 2: Building a 5-touch follow-up sequence

A lead went silent after your first email. The skill generates a sequence: a value-add follow-up (sharing a relevant case study), a social-proof nudge, a break-up email, and a final "close the loop" message—each with distinct subject lines and timing suggestions.

Scenario 3: Rewriting a poor-performing outreach campaign

Your existing emails get 2% reply rates. The agent analyzes your current copy, identifies weak subject lines and generic openers, then rewrites them using the skill's principles—e.g., replacing "I hope this finds you well" with a specific observation about the prospect's company.

Add it to your agent workflow

Install the skill via npx:

npx skills add coreyhaines31/marketingskills --skill cold-email

Then invoke it naturally in your agent prompt: "Use the cold-email skill to write a 3-touch sequence for these 20 leads. Personalize each with their company's recent funding news." The skill will check for product-marketing context, ask clarifying questions if needed, and output ready-to-send drafts.

Compare before adopting

Before committing, compare this skill against alternatives on these axes:

• Quality signals: Does it produce genuinely human-sounding emails? Test with a few sample prompts and review reply-rate potential.
• Maintenance freshness: Check the GitHub repo for recent commits and issue responses—stale skills may rely on outdated sales tactics.
• Alternatives: Look at other outreach skills (e.g., sales-enablement for collateral, or generic email writers) and see which fits your agent's scope.
• Workflow fit: Does it integrate with your existing context files (product-marketing.md) and trigger phrases? A skill that requires heavy custom prompts may not be worth it.

Why it is worth tracking

With 46,642 stars and an active author, this skill shows strong community adoption and iterative improvements (v2.0.0 adds richer context handling). It's worth tracking because cold email is a high-value, high-friction task for AI agents—getting it right directly impacts pipeline generation. Evaluate it when you're building sales-focused agents or enhancing your existing outreach automation. For a deeper look at the skill's evolution and community usage, check out its OpenAgentSkill profile.