AI Prompts Toolkit
GTM AI prompt library and prompt-loop patterns — Claygent research, LLM scoring, cold email drafts, reply classification, account briefs, and iterate-until-quality workflows for sales and marketing. Use when writing Clay AI prompts, designing prompt chains, or building research→draft→score loops in Clay, n8n, or Cursor. Triggers on: "GTM prompts", "Claygent prompt", "AI prompt loop", "cold email prompt", "reply classification prompt", "LLM column Clay", "prompt chain GTM", "research prompt sales".
Install
/plugin marketplace add LeadMagic/gtm-skillsThe repo is a Claude Code plugin marketplace. Run /plugin afterwards to pick the plugin that includes this skill.
npx skills add LeadMagic/gtm-skills --skill ai-prompts-toolkitgit clone --depth 1 https://github.com/LeadMagic/gtm-skills.git /tmp/gtm-skills
mkdir -p ~/.claude/skills
cp -r /tmp/gtm-skills/skills/tools/ai-prompts-toolkit ~/.claude/skills/ai-prompts-toolkitPersonal skills live in ~/.claude/skills. Use .claude/skills in a project folder to share a skill with your team.
From the SKILL.md
# AI Prompts Toolkit ## Overview Generic AI prompts produce generic GTM output — invented personalization, pattern-guessed emails, and hallucinated metrics. GTM prompts need **constraints**: source URLs, word limits, banned claims, ICP context, and explicit failure behavior when data is missing. This skill is the GTM prompt library: copy-paste prompts for Claygent, Clay AI columns, n8n LLM nodes, and Cursor agents — plus **prompt loops** that iterate research → draft → score → revise until quality gates pass. ## When to Use - "Write a Claygent prompt for [task]" - "GTM prompt for cold email personalization" - "Reply classification prompt" - "Prompt loop for account research" - "LLM column in Clay for ICP scoring" - "AI prompt chain for outbound" Load `gtm-context` first if ICP/positioning is undefined — prompts without context hallucinate. ## Authoritative Foundations - **Anthropic — Prompt Engineering.** Separate instructions from data; specify output format; define what to do when information is missing (return empty, not guess). - **Clay Claygent.** Web research agent — must require `source_url` on every factual claim. Credit-heavy — use only after structured enrich fails. - **SPICED (WbD).** Discovery and research prompts map to Situation, Pain, Impact, Critical Event, Decision — not free-form summaries. - **MEDDICC (Whyte).** Research prompts for enterprise deals extract evidence for Metrics, Champion, Economic Buyer — score confidence 0/1/2.
Details
- Skill name
ai-prompts-toolkit- Repository
- LeadMagic/gtm-skills
- Path
skills/tools/ai-prompts-toolkit/SKILL.md- License
- MIT
- Repo updated
- 14 October 2026
- Listed in
- heilcheng/awesome-agent-skills
FAQ
What does the AI Prompts Toolkit skill do?
GTM AI prompt library and prompt-loop patterns — Claygent research, LLM scoring, cold email drafts, reply classification, account briefs, and iterate-until-quality workflows for sales and marketing. Use when writing Clay AI prompts, designing prompt chains, or building research→draft→score loops in Clay, n8n, or Cursor. Triggers on: "GTM prompts", "Claygent prompt", "AI prompt loop", "cold email prompt", "reply classification prompt", "LLM column Clay", "prompt chain GTM", "research prompt sales".
How do I install the AI Prompts Toolkit skill?
Claude Code plugin: /plugin marketplace add LeadMagic/gtm-skills. Skills go in ~/.claude/skills (personal) or .claude/skills (project), and Claude loads them automatically when a task matches the description.
Is AI Prompts Toolkit popular?
It lives in LeadMagic/gtm-skills, which has 79 GitHub stars and 18 forks, and is included in 1 curated skill list.
Related skills
1p Tagging Pixels
79LeadMagic/gtm-skills
Set up first-party tracking and analytics — website pixels, conversion tracking, UTM architecture, cookie consent, server-side tagging vs client-side, 1P vs 3P data strategy, and identity resolution. Use when setting up tracking, implementing pixels, building attribution, or preparing for privacy-first measurement. Triggers on: "pixel", "tracking", "conversion tracking", "first-party data", "server-side tagging", "UTM", "cookie consent", "1P vs 3P", "identity resolution", or any tracking implementation request.
A B Testing
79LeadMagic/gtm-skills
Design statistically sound GTM experiments for copy, channels, landing pages, sequencing, pricing, and funnel conversion. Produces hypothesis, sample-size logic, success metrics, test plan, analysis method, and scale/stop/kill recommendation. Use when planning A/B tests, split tests, or experiment roadmaps.
Abm 1 to 1
79LeadMagic/gtm-skills
Runs strategic 1-to-1 ABM for 5-15 high-value accounts: account plans, custom microsites, executive engagement, direct mail, and board-level connections. Use when a named enterprise account justifies bespoke content and executive time, or when the user needs a strategic account plan. Triggers on: "1-to-1 ABM", "strategic ABM", "custom ABM", "key account marketing", "named account".
AI Sdr Setup
79LeadMagic/gtm-skills
Deploy AI SDR agents safely and effectively — vendor selection, pilot scope, guardrails, signal routing, human handoff design, QA review, and performance metrics. Use when setting up 11x, Artisan, AiSDR, Jason AI, or any automated SDR workflow for outbound prospecting.
Account Selection
79LeadMagic/gtm-skills
Builds and tiers a target account list: fit and intent scoring models, tier assignment, and TAM segmentation. Use when the user needs to decide which accounts to target, score or tier an account list, or prioritize a TAM for ABM or outbound. Triggers on: "account selection", "target account list", "tier accounts", "prioritize accounts", "TAM segmentation".
Churn Prediction
79LeadMagic/gtm-skills
Builds churn prediction: leading indicators, account risk scores, early-warning alerts, and intervention playbooks tied to each risk level. Use when the user wants to predict which accounts will churn, score churn risk, or set up a retention early-warning system. Triggers on: "churn prediction", "predict churn", "churn model", "early warning", "risk scoring".
Browse the most popular Claude skills, all ai, llms & agents skills, or write your own with the skills generator.