Marketing Mindset - Setup Guide¶

Source: skills.sh (28.8K installs) GitHub: axelfreeman/marketing-mindset (52⭐, MIT) Category: Marketing / Growth Strategy First Seen: Sep 2026 (repo created Sep 1, 2026) Quality Tier: 🟢 Production (Gen Agent Trust Hub Pass / Socket Pass / Snyk Pass)

Marketing Mindset is "the marketing OS for AI agents" by Axel Freeman - a 15-year hands-on B2B internet marketer. It is deliberately not another bag of CRO/SEO/copywriting tactics; it encodes how a working marketer decides, so an agent gives a real opinion instead of a template. Six principles (no stale sources, three-month horizons, user's right to make the first move, fast-testable hypotheses, marketing runs ahead of the product, marketing never works for free) plus client-stage playbooks, competitor-as-source-of-truth rules, and memorable heuristics. The README's adapter table lists Hermes (Nous) as a tested platform using SKILL.md, alongside Claude Code, Codex, Cursor, ChatGPT, Grok, and others.

Installation¶

npx skills add axelfreeman/marketing-mindset

For weaker or low-context models the publisher ships a compressed variant (SKILL.lite.md) with the same mindset condensed to essentials. A first-client gate script (scripts/first-client-gate.py) is included for evaluation.

Prerequisites¶

Requirement	Details
No API keys	Fully self-contained skill - SKILL.md, principles, and playbooks
Hermes Agent	Reads SKILL.md natively; listed in the publisher's tested-models table
Optional: examples	`examples/demo-transcript.md` shows a worked example (SaaS for finance, 0 customers → sharp 3-month plan)

What It Provides¶

Capability	How It Works
Decision framework	Evaluates "should I do X to get Y" requests against the six principles instead of dumping tactics
Three-month horizon	Kills 2-year marketing cycles; every idea evaluated on a 90-day usefulness window
Client-stage playbook	First client by hand and free → 2-10 by copying competitors → scale
Competitor source of truth	Positioning and angles reverse-engineered from what competitors actually do
First-move doctrine	The user has the right to make the first move - bold or hacky first steps are not blocked
Exchange accounting	Every marketing action must trade for something; zero-exchange work is refused with the math spelled out
Heuristics	The McDonald's Burger (photograph the product better than it is), think like a cancer cell, the stop-list, the Despair Dividend

Quick Start¶

Install the skill (above)
Trigger it with a decision question: "should we do X to get Y", "where do we get first customers", or "evaluate this positioning"
The skill answers as a director-level marketer - honest feedback, three-month framing, concrete deliverables (competitor analysis, three outbound angles, sharper positioning)
For a worked example, read examples/demo-transcript.md
Pair with execution skills (cold-email, content-strategy) once the direction is decided

CorpusIQ Use Cases¶

Positioning and launch work - sharpen CorpusIQ messaging, launch plans, and competitive angles with an operator's framework rather than generic LLM doctrine
Outreach decision support - judge cold-outreach campaigns and affiliate programs against the exchange principle before burning sends
Content angle selection - belief-bridge and help-first content angles stress-tested against the three-month usefulness window
Growth experiment design - every hypothesis must be fast-testable by any teammate; the skill enforces the test design discipline
Investor and partner decks - marketing claims and go-to-market narratives get the same honest due-diligence read the skill gives founders

Integration with CorpusIQ¶

CorpusIQ is the connector layer that lets AI assistants read the systems a business already runs on - accounting, payments, CRM, email, files and analytics - and answer with the source named. It is read-only by design: it retrieves data and never writes to, or acts in, a connected system, which is what makes it safe to point at live accounts.

This page is part of the CorpusIQ directory, the reference we maintain for the MCP server and agent-skill ecosystem. Run alongside CorpusIQ, a third-party server like this one can be combined with a company's own business data in a single answer.

See what CorpusIQ connects

Limitations / Verification¶

Opinionated framework - the skill is one marketer's method, not a strategy library; it deliberately ships judgment, not templates
Freshness dependency - principle 1 forbids learning from stale sources; the skill expects recent data to work from, which agents must fetch themselves
B2B bias - strongest for services and B2B SaaS; consumer and marketplace dynamics get partial coverage

Security¶

All three skills.sh security audits pass (verified Sep 7, 2026):

Audit	Verdict
Gen Agent Trust Hub	Pass
Socket	Pass
Snyk	Pass

← Skills Catalog | Discovery Page → Powered by CorpusIQ