Project MonetRequest demo
Home/Blog/Marketing Skills for AI Agents: Claude Code, Codex & Cursor Guide

Marketing · Project Monet Briefing

Marketing Skills for AI Agents: What It Is, v2 Changes & How It Works

Marketing Skills packages repeatable SEO, CRO, copy, analytics, paid and GTM workflows into inspectable Agent Skills.

Published 2026-09-01 · Updated 2026-09-01 · By Project Monet Editorial Team

Marketing Skills v2 feeding SEO, CRO, content, paid media, growth and GTM workflows across AI coding agents.

01

What Marketing Skills is

Marketing Skills is an MIT-licensed open-source collection of Markdown-based Agent Skills created by Corey Haines. Its current README documents support for Claude Code, OpenAI Codex, Cursor, Windsurf and hosts that implement the Agent Skills specification.

  • Skills cover SEO, AI SEO, CRO, copywriting, analytics, paid ads, social, launch, pricing, retention and RevOps.
  • The product-marketing skill provides shared context used by downstream workflows.
  • The library is instructions and workflow structure, not a standalone AI model or SaaS dashboard.

02

What changed in v2

The current v2 migration section documents 17 renamed skills, consolidation of page-cro and form-cro into cro, and a move toward a cross-agent .agents/product-marketing.md context file.

  • Old v1 folders can remain beside new names after reinstalling.
  • The current README should be treated as the live rename map.
  • The shared .agents layout makes context less tied to one agent host.

03

How to install and use it

The recommended cross-agent route is npx skills add coreyhaines31/marketingskills. The project also documents selected-skill installs, a Claude Code plugin path, clone/submodule workflows and SkillKit.

  • Install only the workflows you need if a full library is unnecessary.
  • Populate accurate product-marketing context before expecting good downstream work.
  • Verify the agent actually discovers the installed skill rather than answering generically.

04

What a skill library does not guarantee

A skill can make a process more repeatable without making every output correct. The underlying model can still hallucinate facts, misread analytics or suggest outdated platform tactics.

  • Keep publishing, ad-budget and production changes behind appropriate review.
  • Treat GitHub popularity as an attention signal, not proof of marketing performance.
  • Model usage, APIs and connected tools can still carry costs even though the library itself is open source.

Sources

Primary and supporting sources

Facts were rechecked against the linked sources immediately before publication. Pricing, product availability and rollout status can change.

Project Monet

Useful signals. Clear decisions. Better digital work.

Project Monet turns relevant shifts in AI, creator tools and the web into practical context—and builds focused websites for businesses ready to grow.

Request a free homepage concept