esmefong

Gtm Skills — Testing skill for Claude Code

Testing community

Reusable AI skills for go-to-market engineering: ICP research, brand voice, account sourcing and scoring, signal-gated outbound, positioning, market prioritization, content QA.

How to install Gtm Skills

This entry records only its repository, not the path inside it, so there is no exact command to give. Open esmefong/gtm-skills and copy the folder into ~/.claude/skills/, or the file into ~/.claude/agents/.

What Gtm Skills does

Reusable AI skills for go-to-market engineering: ICP research, brand voice, account sourcing and scoring, signal-gated outbound, positioning, market prioritization, content QA. Claude Code format, runnable Python.

Alternatives in Testing

Validate Hypothesis — Business hypothesis validation skill 50 ★
Marketing Skill — Marketing skill for Claude Code — campaigns, funnels, ad copy (Google/Meta/LinkedIn Ads), personas, positionin 30 ★
Ultimate Product Discovery Skill — Ultimate Claude Skill for Product Discovery 13 ★

README

GTM Skills

[![verify](https://github.com/esmefong/gtm-skills/actions/workflows/verify.yml/badge.svg)](https://github.com/esmefong/gtm-skills/actions/workflows/verify.yml)

Built by [Esme Fong](https://esmefong.com). I spent 14 years building data and growth systems, most recently an AI-enabled marketing operating system across three brands as a solo marketer, which cut content production cycle time by 50%. This repo is the reusable machinery from those systems, rebuilt in public: the research, sourcing, outreach, positioning, and quality layers of a go-to-market engine, each one runnable and checked. I build things like this for teams that want their GTM to run as a system; I'm at [linkedin.com/in/esmefong](https://linkedin.com/in/esmefong) and the case studies behind these skills are at [esmefong.com](https://esmefong.com).

Each skill is a folder in the Claude Code skill format: a `SKILL.md` spec an AI agent can execute (a disciplined SOP with judgment rules baked in), reference docs, worked examples, and plain Python scripts. If you use Claude Code, copy a folder into `~/.claude/skills/` and it runs. If you don't, every script is standard-library Python with no API keys and runs anywhere, and every example reads as a document.

One company, end to end

All worked examples run on one real company, **Cohere**, from public data only. This is a standing demonstration, not client work, and I have no affiliation with Cohere; every claim in the examples carries a cited public source and date. The thread: [icp-research](icp-research/) defines who buys and why. [account-sourcing](account-sourcing/) turns that ICP into a ranked, evidenced account list. [outbound-engine](outbound-engine/) turns a live signal into a sequence, or refuses to send. [brand-voice-guide](brand-voice-guide/) extracts the voice that [positioning-message-house](positioning-message-house/) writes in and [content-quality-check](content-quality-check/) enforces. [market-prioritization](market-prio