Skills DirectorySkills Directory
SkillsLearnSecurityCategoriesDocsCommunityBlog
Sign InSubmit Skill
Skills Directory

Security-tested agent skills for Claude, coding agents, and AI workflows.

Directory

  • Browse Skills
  • All Skills A–Z
  • Claude Skills
  • Claude Code Skills
  • Agent Skills
  • Categories
  • Submit a Skill

Learn

  • Learn Hub
  • Install Claude Skills
  • Write SKILL.md
  • Skills vs MCP
  • Directories Compared

Security

  • Security
  • Methodology
  • Secure Claude Skills
  • Security Badges

Company

  • About
  • Community
  • Blog
  • API Docs
  • Advertise

2026 Skills Directory. All rights reserved.

Back to skills

Cold Email 20 Videos

ASecurity

Write and review cold emails using what the 20 most-watched cold email videos of 2026 actually agree on, weighted by how many of them repeat each rule, and flag the choices they disagree on so you A/B test those instead of trusting one guru. Trigger on "write a cold email", "review my cold email", "cold email sequence", or /cold-email.

6 stars
0 votes
0 copies
0 views
Added 9/19/2026
ai-agentsrustgoaws

Security Analysis

A100/100

Scanned 9/19/2026

Install to Claude Code

$npx -y skills add ToolMonsters/gtm-skills-from-youtube --skill cold-email-20-videos --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Cold Email 20 Videos?

Add the live security badge to your README — it updates automatically with every re-scan.

Security grade badge for Cold Email 20 Videos
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/toolmonsters-cold-email-20-videos/badge)](https://www.skillsdirectory.com/skills/toolmonsters-cold-email-20-videos)

More formats (shields.io, HTML) on the badges page.

Download Zip
Files
SKILL.md
---
name: cold-email-20-videos
description: Write and review cold emails using what the 20 most-watched cold email videos of 2026 actually agree on, weighted by how many of them repeat each rule, and flag the choices they disagree on so you A/B test those instead of trusting one guru. Trigger on "write a cold email", "review my cold email", "cold email sequence", or /cold-email.
author: Yonathan Cohen (Tool Monsters)
version: 1.0.0
updated: 2026-09-14
license: MIT
---

# /cold-email

Built from the transcripts of the 20 most-watched cold email videos published between March and September 2026 (203,575 views, 97,000 words). 422 pieces of advice were extracted, 392 were kept because their exact quote was found in the transcript, and they were grouped into 156 distinct rules. The number next to each rule is how many of the 20 videos say it. Sources are in `sources.md`.

No rule was repeated by more than 9 of the 20. Treat the weights as odds, not laws.

## Inputs

1. What you sell, in one sentence.
2. Who you are writing to: role, company, and one real detail about them (something they posted, launched, hired for, or complained about).
3. What you can give for free that is genuinely useful to them.

If input 2 has no real detail, stop and ask for one. A fake detail is worse than none.

## The rules, by weight

### Strong (6 to 9 of 20)

- **Lead with something free and useful, not a call ask.** (9/20) A tailored idea, a short list, a first piece of the work done. Not a PDF, not "hop on a call".
- **Personalize with one real, researched detail.** (8/20) If it could be sent to 1,000 people, rewrite it.
- **Keep it short.** (7/20) Around 50 to 80 words for the first email. Each sentence earns the next.
- **Test copy and offers continuously.** (7/20) Never judge the channel on one variant.
- **Follow up several times, a few days apart.** (6/20) Most say 2 to 5 follow-ups.

### Solid (4 to 5 of 20)

- **Verify every email before it goes out.** (5/20)
- **End with a soft question**, like "worth a look?" (5/20)
- **Segment by role and company type**, one message per segment. (5/20)
- **Subject lines short, casual, lowercase feel.** "quick question, {first name}". (5/20)
- **Recontact non-responders later with a new angle**, not the same email. (5/20)
- **First email in plain text, no links.** (4/20)
- **Show one proof point**: a result, a known customer. (4/20)
- **Target a real signal**: hiring, funding, new leader, expansion. (4/20)
- **Do not pitch or ask to close in email one.** (4/20)
- **Reply to interested people within minutes.** (4/20)

### Mentioned (3 of 20)

Simple words, no jargon · build copy from phrases your buyers used on calls · small specific lists beat huge generic ones · never send from your main domain.

## Where the experts disagree: do not pick, test

When a draft touches one of these, write **two versions** and label them A and B.

| Choice | Side A | Side B |
|---|---|---|
| Calendar link | Include a link or time slots to cut friction | Never put a calendar link in a cold email |
| Follow-ups | 2 to 5 short bumps | One follow-up only, framed as a new offer |
| Personalization depth | Deep, researched, company-specific | Light: over-personalizing email one backfires |
| CTA | Soft question | Specific, easy call ask |
| Volume | Start small and precise | Send far more than you think |

## Steps

1. Read the three inputs. Refuse to invent the personal detail.
2. Draft email one applying every Strong rule, then the Solid ones that fit.
3. Count the words. Over 90, cut.
4. Check it against the list: free value first, one real detail, no link, soft question, no pitch.
5. If a choice from the disagreement table applies, output version A and version B.
6. Draft follow-ups only if asked, each with a new angle.
7. Under the email, list which rules you applied, with their weight, and which choice is being A/B tested.

## Never

- Never claim a reply rate. None of these numbers are yours until you send.
- Never invent a mutual connection, a compliment, or a detail about the prospect.
- Never write "I hope this finds you well".

Attribution

ToolMonstersToolMonsters
View sourceMore from ToolMonsters →
SSkills DirectorySkills Directory

Ship a skill? Prove it's safe.

Free 120-pattern security scan, letter grade, and an embeddable README badge.

Submit a skill

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments (0)

No comments yet. Be the first to comment!

SSkills DirectorySkills Directory

Ship a skill? Prove it's safe.

Free 120-pattern security scan, letter grade, and an embeddable README badge.

Submit a skill

Related Skills

Caveman

Ultra-compressed communication mode. Cuts token usage ~75% by speaking like caveman while keeping full technical accuracy. Supports intensity levels: lite, full (default), ultra, wenyan-lite, wenyan-full, wenyan-ultra. Use when user says "caveman mode", "talk like caveman", "use caveman", "less tokens", "be brief", or invokes /caveman. Also auto-triggers when token efficiency is requested.

1023331 votes

Hyperplan

Adversarial multi-agent planning skill. Self-orchestrates 5 hostile category members (unspecified-low, unspecified-high, deep, ultrabrain, artistry) via team-mode for ruthless cross-critique debate, distills only the defensible insights, then MANDATORILY hands the distilled insight bundle to the `plan` agent for executable plan formalization. Use when planning needs maximum rigor and surfacing of weak assumptions, blind spots, and over-engineering. Triggers: 'hyperplan', 'hpp', '/hyperplan', ...

686011 votes

Mcp Code Execution

Routes multi-tool workflows through MCP servers for large datasets and pipelines. Use when Bash tool overhead is limiting throughput on data-heavy tasks.

3331 votes

catchup

Recovers prior coding-agent session context by running `catchup <agent> --since-compact`, which extracts a clean summary of a previous Codex, Claude Code, Antigravity, OpenCode, or Pi Agent session. Use when the user says "catch up", "what did the last session do", "get me up to speed", "I switched agents", or asks to recover/summarize a previous session before continuing. Do NOT use for the current conversation, git history, or any non-agent log.

611 votes

math-skill

A comprehensive mathematical reasoning skill for AI assistants — handles arithmetic to research-level problems with rigorous step-by-step reasoning, systematic verification, and transparent uncertainty handling

381 votes
View all in ai-agents →