Traction scoreGitHub stars can be faked, so popularity alone can be misleading. Traction Score looks for broader signs of recent attention, adoption, and active maintenance.
GrowthEngineX’s outbound playbooks start with the customer profile and offer before moving into infrastructure, lists, and copy. A guided kickoff connects those decisions into a campaign plan.
Compatible AgentsThe repository documents support for these agents. The skills may also work with other agents that can load SKILL.md files, but they may need some setup or small changes.
Follow the documented setup, then try a first task.
Load the signal playbooks directory in Claude Code
Installs: The repository's complete collection of 19 signal playbooks. The listed skill path is a supported selection, not the complete inventory. · Claude Code
Before you start
Claude Code installed.
An existing local checkout of growthenginenowoslawski/coldoutboundskills.
Repository setup completed, including tsx and the cold-email-starter-kit npm dependencies.
Your own API keys configured for the skills you intend to use.
In Claude Code, point Claude at the skills/playbooks directory within your local repository checkout. The author documents this directory-loading method because the playbooks live one level deeper than the other skills.
A lead CSV with at least 100 rows and columns email, first_name, last_name, job_title or title, and company_name or company.
Optional company_domain, company_industry, and company_headcount columns.
Optional client-profile.yaml containing declared ICP filters; this is a sample filename to replace with your actual profile file.
Available email-verification information for assessing verification coverage.
Make sure list-quality-scorecard is installed in your agent before trying this task.
Use list-quality-scorecard to assess my supplied lead CSV before upload. Return the letter grade, eight dimension scores, top five issues, and pre-send checklist. Use my client-profile.yaml if supplied to assess ICP fit. Report recommendations without modifying the list or uploading it.
A Smartlead account with campaign activity, preferably across multiple inbox types.
Optional sub-client ID if the comparison should cover a sub-client.
Make sure deliverability-test-public is installed in your agent before trying this task.
Use deliverability-test-public to compare my Smartlead account's inbox types over the default seven-day lookback. Show sent counts, raw reply rates, and bounce rates by inbox type, and explain any differences while flagging small samples. Keep this to analysis.
A competitor's or existing customer's LinkedIn source-account URL.
The client's domain and explicit ICP criteria, including relevant countries, headcount bands, and job titles.
Existing client DNC, TAM, and live-campaign information needed to plan suppression checks.
Use playbook-linkedin-engagement to prepare a one-post pilot from my supplied competitor or customer LinkedIn account and client ICP. Identify missing setup, explain how the source account must be resolved to a bare domain, and outline the source-company exclusions, ICP filters, deduplication, and suppression checks. Keep personalization lines off. Return a pilot checklist for review before any paid calls or DNC updates.