Cold Email Campaign Strategy Claude Skill

The Cold Email Campaign Strategy Claude skill turns a client website into at least 15 cold outbound campaign ideas, each with a targeting approach and value proposition. It is free and open source under the MIT license, from the Cold Outbound Skills repo by GrowthEngineX. It is built for agencies and operators planning experiments before a client launch.

Outbound Intermediate MIT license 799 GitHub stars
Install
Official (clone repo)
git clone https://github.com/growthenginenowoslawski/coldoutboundskills.git ~/cold-email-ai-skills
Skills CLI
npx skills add growthenginenowoslawski/coldoutboundskills --skill campaign-strategy
Manual
cp -r ~/cold-email-ai-skills/skills/campaign-strategy ~/.claude/skills/campaign-strategy/
View on GitHub Checked

What the Cold Email Campaign Strategy skill does

Every campaign has two levers: the list and the message. The skill ranges ideas from broad to niche. Broad lists lean on AI personalization, while niche lists can state the filter directly in the copy. The deeper and narrower the list, the more the message can speak to that exact group.

It starts with a research protocol. Claude follows real navigation links, collects every customer and case study, and looks for patterns in industry, size and buyer role. It may then challenge the targeting you were given. It also splits ideas into AI-personalized and no-AI campaigns, so you can run simple tests first.

Each idea passes a simple test: what would a rep look for in ten minutes of research? That keeps AI strategies tied to public data and away from sources that are not reliably accessible. It adds front-end offer ideas that let a prospect see value inside a free trial.

  • Crawls a client site systematically, extracting every customer and case study mentioned.
  • Generates at least 15 to 20 campaign ideas from broad to niche targeting.
  • Always includes a creative ideas campaign, a new hire campaign and a lookalike campaign.
  • Maps each idea to a value proposition: more money, saved time, saved money or lower risk.
  • Describes an AI personalization strategy using only publicly available data for each campaign.
  • Adds no-AI campaigns and front-end offer suggestions that show value inside a trial.
  • Can load an ICP profile and lead magnet file from earlier steps in the repo.
  • Challenges the given targeting if case studies suggest a wider or different ICP.

How it works

  1. Gather context

    Takes a website URL at minimum, and ideally an ICP profile file, lead magnet ideas, audience limits and call notes.

  2. Research the site

    Starts at the homepage, follows navigation, and reads customers, features, pricing, about and blog pages without guessing URLs.

  3. Mine case studies

    Extracts company, industry, size, results and problem for every customer, then spots patterns for lookalike ideas.

  4. Generate ideas

    Builds campaign ideas across broad, focused and niche levels, applying the ten-minute manual research test to each.

  5. Format the output

    Delivers an ideas table, no-AI campaigns and front-end offer suggestions with a summary of audience and differentiators.

Example prompts

Once the Cold Email Campaign Strategy skill is installed, ask Claude in plain language. These prompts trigger it:

  • Here is my client's website, give me 20 cold email campaign ideas
  • Plan outbound experiments for a SaaS that sells to sales teams, include a new hire campaign
  • What lookalike campaign could we run from this company's case studies?
  • Give me campaign ideas that need no AI personalization
  • Suggest a front-end offer for a 14-day trial of this product

What you get

  • A table of 15 to 25 or more campaign ideas with targeting level and value proposition
  • An AI personalization strategy for each idea based on public data
  • A section of no-AI campaigns
  • Front-end offer suggestions tied to the product's trial

Who it is for

  • Agencies onboarding a new outbound client
  • Founders planning their first cold email experiments
  • Teams that want a campaign backlog grounded in real customer patterns

How to install the Cold Email Campaign Strategy skill

What you need: Claude Code or Claude with web access to read the client's site. Optional inputs from other skills in the repo, such as an ICP profile, improve results.

  1. Official (clone repo)
    git clone https://github.com/growthenginenowoslawski/coldoutboundskills.git ~/cold-email-ai-skills
  2. Skills CLI
    npx skills add growthenginenowoslawski/coldoutboundskills --skill campaign-strategy
  3. Manual
    cp -r ~/cold-email-ai-skills/skills/campaign-strategy ~/.claude/skills/campaign-strategy/

After installing, restart Claude Code or start a new session. Claude loads the Cold Email Campaign Strategy skill automatically when your request matches it, or you can ask for it by name.

Tips for better results

  • Provide an ICP profile and lead magnet ideas if you have them, since the skill loads them first.
  • Let Claude challenge your target audience when case studies point somewhere wider.
  • Skip ideas that need G2 or Capterra data. The skill flags those sources as unreliable.

Limitations

  • It plans campaigns but does not write the email copy or build the lists.
  • Some enrichment ideas assume tools such as Clay, so execution may need extra software.

Cold Email Campaign Strategy Claude skill FAQ

Yes. It is free and open source under the MIT license, from the Cold Outbound Skills repo. Executing the campaigns may need paid tools, but the planning skill itself costs nothing.

Clone the repo, or run npx skills add growthenginenowoslawski/coldoutboundskills --skill campaign-strategy to install just this skill folder. Restart Claude Code afterward so it picks up the new skill, then ask for the task in plain language.

It is instructions only, so it can be zipped and uploaded in Claude.ai under Settings, Capabilities, Skills on paid plans. It needs web access to research the site, and some repo files it can load are optional.

Yes. It crawls the site, collects customers and case studies, and proposes broad to niche campaigns with value propositions and personalization ideas, all based on public data.

Always a creative ideas campaign, a new hire campaign and a lookalike campaign, plus stretch ideas using stacked signals, unusual titles, quantified pain and team structure.