AI Growth Skills Pack

Your growth work ships
before the question is answered.
This pack asks it first.

Drop one folder into your project. Claude Code now must check your positioning says something specific, verify your copy answers 'who is this for?' and 'why does it matter?', plan how a test ends before it starts, and confirm any AI performance claims have a real source — or it cannot mark the work done.

1
Hard Gate
6
Skills
3
Agents
0
Unsourced AI Claims

What gets skipped without this pack

Four ways growth work fails before it ships

Not edge cases. The default outcome when no growth quality enforcement exists.

No Positioning Test

"We help teams move faster"

The positioning passed every internal review. It failed the "so what?" test with every prospect. Competitors said exactly the same thing. Three steps into the chain of consequences, nobody could explain why it mattered for their specific problem.

Blocked by positioning-audit — category, narrow audience, differentiated claim that doesn't apply to main competitor
No Copy Gate

"Powerful AI for modern teams"

The hero copy tested well with the team. "Who cares?" — there was no audience identifiable from the copy alone. "Why now?" — there was no urgency. It could have been written for any product, any year, any market.

Blocked by copy-quality-gate — "so what?", "who cares?", and "why now?" must all pass before copy ships
No Stopping Rule

"We'll stop the test when it's significant"

The experiment ran for two weeks. It looked good at week two, so they stopped. It looked good because of random variation — the sample wasn't there yet. The feature shipped. The lift didn't hold in production.

Blocked by experiment-design — fixed stopping rule, calculated sample size, and all four decision outcomes written before the test starts
No Messaging Review

"Up to 10× faster with AI"

The claim shipped without a source. It wasn't wrong — but it wasn't sourced, the population wasn't scoped, and the baseline wasn't stated. A journalist asked for the methodology. There wasn't one.

Blocked by ai-messaging-review — every quantified claim needs source, population, and baseline; every AI label replaced with specifics

The 6 Rules

One rule per growth quality failure

One rule blocks messaging from shipping without sourced claims. Five rules enforce how every piece of growth work is done.

ai-messaging-review Every performance claim needs a source, says exactly who it applies to, and states what it's compared against. Every vague 'AI' label replaced with what the AI actually does. Verdict: ship, fix these things, or hold. Hard Gate
positioning-audit Your category is something people search for. Your audience is specific enough to name 10 real people. Your claim doesn't apply to your main competitor. And you can explain why it matters in two sentences or fewer. Workflow
copy-quality-gate Who this is for is clear from the copy alone. Why they should care has a real answer within two 'so what?' questions. The urgency is genuine — not 'now more than ever.' All three must pass. Workflow
funnel-analysis The biggest leak is calculated (users × exit rate), not guessed. The root cause is confirmed two independent ways. 'Users drop off at step 3' is not a root cause — you need to know why. Workflow
experiment-design What you're testing, how you'll measure it, how many users you need (calculated), when it ends (fixed), when you can call it early (with a threshold), and what you'll do in each of the four possible outcomes — all written before the test starts Workflow
retention-design The moment a user first gets real value, the pattern that brings them back (what triggers it, what they do, what they get), and how to re-engage users who've gone quiet — all three designed before any retention feature ships Workflow

How it works

Claude must produce real proof — not just say it's done

Each rule requires a specific output with real values. Claude cannot say copy passed without showing the actual 'so what?', 'who cares?', and 'why now?' test results.

copy-quality-gate — completion statement
Copy review complete.
Copy: growth/copy-reviews/homepage-hero-2026-06-07.md

Copy reviewed:
"Stop manually synthesising 400 support tickets every Friday.
Prism delivers the week's product decisions — with evidence chains — on Monday morning."

So what? chain (≤ 2 steps):
Step 1: 15–20h/week of manual synthesis eliminated
Step 2: PM makes defensible product decisions with evidence — chain ends ✓

Who cares? Product Managers at 20–200 person B2B SaaS with unstructured feedback — identifiable from copy alone ✓
Why now? Feedback volume doubles every quarter; manual synthesis doesn't scale — real urgency ✓

Verdict: SHIP ✓

An agent that skips the positioning audit cannot identify the audience from copy alone or verify the "so what?" chain terminates in ≤ 2 steps.
That is the enforcement.


The 3 Specialists

Three AI roles Claude can take on

Each specialist has a narrow focus. All produce output at specific file paths — no chat-only results.

Sonnet

growth-critic

Reviews positioning, copy, and experiment designs against all 6 rules. Verdict: pass, fix these things, or blocked. Writes the report to a real file.

Sonnet

experiment-designer

Designs experiments before they start — calculated sample size, exactly when to stop, and what decision you'll make in each of the four possible outcomes. Writes it to a real file.

Opus

messaging-reviewer

Reviews any copy that makes claims about AI. Checks every number has a source, every claim has a defined scope, and every 'AI' label says what the AI actually does. Writes the report to a real file.


Before and after

What changes when enforcement exists

The difference is questions answered before work ships, not problems discovered after it does.

✗ Without builder-growth

Hero copy: "The AI-powered platform for modern teams." Same as six competitors.
Positioning: "We help teams move faster." Three "so what?" steps later: still no answer.
A/B test run until it looked significant. Sample wasn't there. Lift didn't hold.
"Up to 10× faster with AI" shipped without a source, scope, or baseline.
Funnel "optimised" at the highest-traffic step. The actual leak was two steps earlier.
Retention feature built without defining the activation moment or habit loop.

✓ With builder-growth

Copy names the audience, the problem, and the urgency. Survives "who cares?" cold.
Positioning passes search query test. Audience: 10 PMs named. Claim: doesn't fit competitor.
n=1,842 per arm. Minimum 2 weeks. Fixed p < 0.001 stopping rule. All four outcomes decided.
"15–20h/week eliminated — internal study, 12 PMs, Q1 2026, vs. manual synthesis baseline."
Primary leak: step 3 exit × volume = 2,400 users/week. Root cause: two independent methods.
Activation moment defined. Habit loop mapped. Reactivation path: lapse trigger + personalised landing.

Installation

30 seconds to enforce growth quality

Installs only skills/ and .claude/agents/. Non-destructive — won't overwrite existing files.

macOS / Linux
bash <(curl -fsSL https://raw.githubusercontent.com/
RBraga01/builder-growth/master/install.sh)
Windows (PowerShell)
irm https://raw.githubusercontent.com/
RBraga01/builder-growth/master/install.ps1 | iex
Start with positioning-audit before writing any copy. Run copy-quality-gate before any text ships. Run ai-messaging-review for any copy that describes AI capabilities. Use experiment-design before any A/B test starts — not after it's running.

The builder-* ecosystem

One pack per domain

All packs work standalone. All share the same enforcement model. Use one, some, or all.

Foundation

A Team

25 pre-configured engineering specialists, 18 workflow skills, a lead orchestrator, and a Pipeline Auditor. The base layer.

You are here

builder-growth

Growth quality gates. Positioning, copy, funnel analysis, experiments, retention design, AI messaging review.

AI Engineering

builder-ai

Eval, prompt versioning, fallback design, RAG pipelines, safety review. Three hard gates that block ship.

Design

builder-design

AI UI design enforcement. States, streaming UI, prompt UX, accessibility, and design tokens.

Product

builder-product

Product quality gates. PRD, feature scoping, metric definition, research synthesis, A/B test design, AI feature validation.