Agencies
Every client needs the same fifteen artefacts and the quality drifts with whoever's free.
A consistent floor across the roster, so the work doesn't depend on who picked it up.
See the setupFit
Who it's for
These skills assume you'd rather have an opinionated collaborator than a blank page. If you want templates, there are faster options. If you want judgement, start here.
By role
4Ecommerce
Email, SMS, ads, CRO, referrals — the channels multiply faster than the team does.
A playbook for every channel, run from the AI tools you already use.
See the setupFounders
You can ship the product but the marketing sits untouched because you don't know what good looks like.
Skills that carry the judgement, so you supply context and get work back you'd have paid for.
See the setupSaaS
Pricing, onboarding, activation, churn — every lever needs a specialist you don't have headcount for.
Playbooks for the whole funnel, installed into the tools you already build in.
See the setupBy agent
16ChatGPT
ChatGPT Skills read SKILL.md on business plans — and on any plan the playbooks drop into Projects or chats.
Free, MIT-licensed, readable markdown.
See the setupClaude Code
The full library running natively in Claude Code — copywriting, SEO, ads, email, pricing, and more.
Free, MIT-licensed, installed with one command.
See the setupCodex
Every skill is plain markdown Codex can load — the whole library works wherever your agent runs.
Free, MIT-licensed, installed with one command.
See the setupGitHub Copilot
Skills live in your repo as reviewable, versioned files — Copilot's agent works from the same playbooks as your team.
Free, MIT-licensed, installed with one command.
See the setupCursor
The same 49 skills inside Cursor, so marketing work ships from the editor your code already lives in.
Free, MIT-licensed, installed with one command.
See the setupGemini CLI
The open Agent Skills format Gemini CLI reads — one install, expert playbooks for every marketing discipline.
Free, MIT-licensed, installed with one command.
See the setupGrok Bot
Give xAI's always-on agents 49 expert playbooks — and Grok Build loads the Claude-compatible format natively.
Free, MIT-licensed, installed with one command.
See the setupHermes
Hermes reads SKILL.md natively — all 49 playbooks load without modification, from the CLI or /skills in chat.
Free, MIT-licensed, installed with one command.
See the setupInstinct
Point the personal assistant at a playbook and ask for the outcome — expert process instead of a generic answer.
Free, MIT-licensed, readable markdown.
See the setupOpenClaw
Readable markdown skills for the personal agent on your own hardware — auditable before your agent ever runs them.
Free, MIT-licensed, installed with one command.
See the setupOpenCode
A first-class target of the skills installer — open-source agent, open-source playbooks, inspectable end to end.
Free, MIT-licensed, installed with one command.
See the setupPerplexity Computer
Deliverable-first sub-agents work from the playbook you attach — audits, sequences, and campaign plans to a standard.
Free, MIT-licensed, readable markdown.
See the setupPi
Bring a playbook into the conversation — Pi's conversational strength plus a specialist's process for thinking work through.
Free, MIT-licensed, readable markdown.
See the setupVellum
Vellum's skills are SKILL.md directories — this library loads natively into the persistent personal assistant.
Free, MIT-licensed, installed with one command.
See the setupViktor
The AI employee in Slack and Teams executes across your ad platforms and CRM — these playbooks set the standard it works to.
Free, MIT-licensed, readable markdown.
See the setupWindsurf
Cascade works from the matching skill's playbook — focused marketing guidance instead of generic output.
Free, MIT-licensed, installed with one command.
See the setup