AI analysis grounded in the code graph — computed facts, not vibes · 2026-10-05T03:03:24Z
This is a collection of markdown "skills" files that give Claude Code, OpenAI Codex, Cursor, Windsurf, and other Agent Skills-spec agents structured knowledge for marketing tasks — CRO, copywriting, SEO, analytics, ads, retention, GTM, and more. Underpinning the skills is a tools/clis module (286 symbols) of JavaScript wrappers around marketing platform APIs — Resend, ActiveCampaign, Paddle, PartnerStack, Intercom, Brevo, Postmark, Klaviyo, Typeform, Plausible, Optimizely, DataForSEO, Lemlist, Beehiiv, Hunter, Buffer, and others — which agents can call when executing a skill's workflow. It's aimed at technical marketers and founders who want a coding agent to act as a marketing operator rather than just a chatbot.
The growth (38,806 stars, 299 gained in a single day, created only around 2026-01-15) tracks a very high release cadence: v2.6.0, v2.5.1, v2.5.0, v2.4.2, v2.4.1 all shipped within roughly two-and-a-half weeks, each adding or renaming a skill (marketing-loops, offers, ai-seo OKF coverage, public-relations). The commit history shows near-daily feature PRs across ad-creative (multiple sub-releases: 2.2.0 through 2.7.0), ai-seo, social, and prospecting, which is unusually fast iteration for a skills library and likely keeps it visible in feeds. The README also links the project to Corey Haines' existing audience (Swipe Files, Conversion Factory, AI Marketing Training, Magister), which plausibly explains distribution beyond organic GitHub discovery — though the graph facts alone don't quantify that effect.
What changed recently, how it's actually built (from the code graph), and whether you should care. Free account — no card, no spam.