signal-builder
Scans a prospect website + enrichment data and produces a ranked signal analysis. Each signal scored 1-10 with a recommended campaign approach.
- "Score the signals on stripe.com for our offer"
- "What's the angle for this prospect?"
Open Source
Nine drop-in skills that encode four years of GTM delivery. Score signals, design personalization, write campaigns, and pull every data layer underneath - without re-explaining the framework every session.
npx skills add Zevenue/gtm-skills The loop
Nine skills, but only three workflows. Each skill does one job and hands its output to the next. The data layer feeds the brain. The brain feeds the writers.
Inputs · Data layer
Brain
Outputs · Methodology
The catalog
Methodology skills run on Claude alone. Data layer skills wrap external APIs - a one-time .env setup unlocks all of them.
Scans a prospect website + enrichment data and produces a ranked signal analysis. Each signal scored 1-10 with a recommended campaign approach.
Designs the personalization variables for a campaign. Encodes four archetypes (verbatim-pain, manual-task, strategic-alternative, failure-mode) with grammar, sources, Claygent prompts, fallbacks, examples.
Writes 3-email cold campaigns using the Situation -> Insight -> Inquisition methodology. Enforces deliverability rules, word limits, and a QA checklist.
Semantic web search via Exa neural. Take a GTM thesis and return ranked URLs from the open web. The "find" step before any scraping.
Web-to-markdown layer. Hand it URLs, get clean LLM-ready markdown. Sits between finding URLs and reasoning about them.
Narrow Apollo wrapper for person + company enrichment. Title, employer, headcount, industry, funding, tech stack. No search, no sequencing.
Email resolution + verification via Prospeo. Take a person and return a verified email + confidence score. The "find an inbox" step before send.
Queries TheirStack for job postings at a set of companies. Hiring patterns are one of the strongest timing signals for outbound.
Scrapes recent LinkedIn posts from prospect profiles via Apify and scans for a theme. Returns quoted excerpts and post links per profile.
Install
Drops the skills into .claude/skills/ in the current project. Works in any repo.
# install all skills
npx skills add Zevenue/gtm-skills
# install specific skills
npx skills add Zevenue/gtm-skills \
--skill signal-builder email-writer
# list available
npx skills add Zevenue/gtm-skills --list Use Claude Code's native marketplace and plugin system.
# add the marketplace
/plugin marketplace add Zevenue/gtm-skills
# install the bundle
/plugin install gtm-skills Old-school. Useful if you want to fork and edit before installing.
git clone https://github.com/Zevenue/gtm-skills.git
cp -r gtm-skills/.claude/skills/* \
~/.claude/skills/ Methodology skills run on Claude alone. The 6 data layer skills wrap external APIs - drop the keys you have and skip the rest.
cd gtm-skills
cp .env.example .env
pip install -r requirements.txt
# then fill in whichever keys you have:
EXA_API_KEY=...
FIRECRAWL_API_KEY=...
APOLLO_API_KEY=...
PROSPEO_API_KEY=...
THEIRSTACK_API_KEY=...
APIFY_API_TOKEN=... Opinions, encoded
They argue for a specific way of running outbound. Five judgments we've sharpened across four years of delivery - baked into every skill.
If three well-targeted emails don't land, more won't help. The signal or the angle was wrong.
If you need heavy personalization to make copy work, your targeting is wrong.
Cold email CTAs should be "Is this you?" not "Can we schedule 30 minutes?"
A 4/10 is fine. It tells the email writer to calibrate tone, not skip the prospect.
A company posting a Head of RevOps role is in pain you can speak to.
The skills above are everything we use ourselves. If you want help wiring them into your stack - or building custom skills for your specific offer, ICP, and tools - that's what our AI Enablement program does.