Systems Lab

Agent skill

subject-lines

Writes and optimizes subject lines for cold emails.

slowingReaches the webInstructions only422 words

Filed under Outbound email.

From Frontal-so/outbound-skills · 73 skills · 4 · pushed 2026-07-14

What it does when it runs

Writes and optimizes subject lines for cold emails. Use when the user asks to "write a subject line", "improve my subject line", "subject line ideas", "A/B test subject lines", "open rate optimization", "my open rates are low", or needs help with email subject lines specifically. Also triggers on "subject line best practices", "what subject line should I use". Do NOT use for writing full email body copy (use first-touch or follow-up), deliverability issues causing low opens, or newsletter subject lines.

Read from the skill and the 0 files bundled beside it. A skill’s own description is written to be selected by an agent, so it describes the job and not the dependencies.

Keys and connectors you must supply
None found.
Hosts it reaches
  • frontal.so
Tool permissions it declares
No allowed-tools in the frontmatter. It only issues instructions, so there is nothing to bound.
Actions present in the files
None. Instructions only.

Ask about subject-lines

Opens your assistant with this page's verified links already in the prompt.

Is this safe to install?ClaudeChatGPT
Adapt it to my stackClaudeChatGPT
What else do I need for it to workClaudeChatGPT
Rather ask a human? Talk to Cheetah
git clone --depth 1 --filter=blob:none --sparse https://github.com/Frontal-so/outbound-skills.git /tmp/outbound-skills
git -C /tmp/outbound-skills sparse-checkout set "master-skills/cold-email/.claude/skills/subject-lines"
mkdir -p ~/.claude/skills/subject-lines
cp -R "/tmp/outbound-skills/master-skills/cold-email/.claude/skills/subject-lines/." ~/.claude/skills/subject-lines/

Picked up without a restart. A project skill of the same name is shadowed by your personal one. For one repository only, swap ~/.claude/skills for .claude/skills. Claude Code docs ↗

The folder is the same in every client that implements the format — 46 of them — so if yours is not above, only the destination changes.

Reproduced in full from Frontal-so/outbound-skills/blob/27cd847be5ce7e9b749b1d672cbfb29935d911db/master-skills/cold-email/.claude/skills/subject-lines/SKILL.md, which is licensed MIT (repository). 422 words, 11 headings.

Subject Line Writing

You write high-performing subject lines for B2B cold emails. Subject lines should feel like an internal email from a colleague, not marketing.

Process

  1. Understand the email -- What framework and value prop is the email using?
  2. Draft 3-5 options -- Mix personalized, curiosity, and direct approaches
  3. Recommend A/B test -- Pick 2 contrasting styles to test against each other

Reference

Read {SKILL_BASE}/resources/frameworks/writing-frameworks.md for CTA rules and writing principles that apply to subject lines. Read {SKILL_BASE}/resources/prompts/personalization-prompts.md for the AI subject line prompt (2-word formula) and personalization hooks.

Rules

  • 2-4 words ideal -- Shorter subject lines outperform in cold email
  • All lowercase -- Feels casual and personal, not promotional
  • No spam trigger words -- Avoid "free", "guarantee", "limited time", "act now"
  • No punctuation overkill -- No exclamation marks, no ALL CAPS
  • Relevant to the email body -- Deceptive subject lines destroy trust and trigger spam
  • Same thread for follow-ups -- Email 2 uses RE: (same subject), Email 3 gets a new subject

Subject Line Formulas

1. Personalized (highest open rate)

{{trigger_topic}}
{{metric}} at {{company}}
{{years}} years

2. Curiosity Gap

quick question
one thing about {{topic}}
{{company}} -> {{outcome}}

3. Direct Value

3 hours back
{{number}} potential leads
Q1 revenue efficiency

4. Two-Word Formula (AI-generated)

Rules: Exactly 2 words, all lowercase, relevant to their business, no sales/buzz words.
Example: "marketing data", "hiring speed", "pipeline math"

A/B Testing Guidelines

  • Test 2 subject lines at a time (not more)
  • Minimum 100 sends per variant before judging
  • Measure open rate AND reply rate (high opens with low replies = clickbait)
  • Test contrasting styles: personalized vs. curiosity, short vs. slightly longer

Examples

Example 1: For a hiring platform email

Option A: "12 open roles" (personalized + specific)
Option B: "hiring speed" (two-word formula)
Option C: "quick question" (curiosity)
Recommendation: A/B test Option A vs Option C

Example 2: For a cybersecurity email

Option A: "{{company}} security" (personalized)
Option B: "quick math" (curiosity + direct)
Option C: "vulnerability scan" (two-word formula)
Recommendation: A/B test Option A vs Option B

Example 3: For a re-engagement email

Option A: "since we last talked" (acknowledges history)
Option B: "different angle" (curiosity)
Option C: "quick question" (universal opener)
Recommendation: A/B test Option A vs Option B

Part of Frontal — free, open GTM skills for your AI agent. Browse the library →

Other skills for the same job

Different authors, same problem. Matched on the words in the skill name, across every library in the catalogue except this one.

Need help setting it up?

This page tells you what subject-lines does and what it needs. Cheetah builds the agent setup it runs inside: data, CRM, sequencing and the guardrails.

Book a call →

The directory stays free. There is nothing gated behind this.