About startup-growth-review
A skill for turning a product, page, funnel, or go-to-market idea into a prioritized growth plan. It supports audits, experiment planning, and copy rewrites, with recommendations tied to measurable behavior.
Key features
- 01Audit conversion and funnel gaps
- 02Plan experiments and launch sequences
- 03Rewrite messaging and onboarding copy
- 04Prioritize by impact and confidence
- 05Identify measurement gaps
Use cases
- 01Reviewing signup or retention performance
- 02Planning a product launch experiment
- 03Revising calls to action or onboarding
- 04Separating evidence from assumptions
Install
Copy the command and run it in a terminal.
$
npx skills add https:/ / github.com/ sylphxai/ skills --skill startup-growth-reviewAfter installing, type/startup-growth-reviewin Claude Code to confirm it loaded.
Bundled filesagents/openai.yamlreferences/growth-levers.md
Details
From the project READMEStartup Growth Review
Use this skill to turn a product, page, funnel, or go-to-market idea into a prioritized growth plan.
Operating modes
- Audit: identify conversion, activation, retention, pricing, or positioning gaps.
- Plan: propose experiments, channel strategy, launch sequence, or instrumentation.
- Rewrite: improve messaging, offers, CTAs, or onboarding copy.
Workflow
- Clarify the growth unit: visitor, signup, activated user, paid customer, retained team, referral, or expansion.
- Read
references/growth-levers.mdfor the relevant funnel stage. - Separate evidence from assumptions. If data is missing, state the minimum measurement needed.
- Prioritize by expected impact, confidence, speed, and strategic fit.
- Produce a short action plan with owners, success metric, and fastest validation.
When not to use
- Do not use for generic advice the base model already handles without this skill's specific artifact contract.
Guardrails
- Prefer one sharp wedge over many generic channels.
- Do not recommend spam, dark patterns, fake scarcity, deceptive testimonials, or manipulative UX.
- Preserve product truth: do not invent claims, logos, metrics, integrations, or customer proof.
- Tie every suggestion to a measurable behavior.
- Call out when the real blocker is positioning, not traffic.
Output format
Diagnosis: <one paragraph>
Top moves:
1. <move> — metric, why now, validation
2. <move> — metric, why now, validation
3. <move> — metric, why now, validation
Instrumentation gaps:
- <event or property needed>
Do next in 48 hours:
- <small concrete action>
Comments
Markdown supported