Two Skills I Built This Week (And Why I Built Them)
/ 2 min read
Table of Contents
Part of the #shipped-with-ai series — one build per week, documented honestly.
Skills — reusable AI context files you can share and install — are becoming a core part of how I work. This week I published two. Both were built to solve a real problem I ran into, not as standalone projects.
laravel-prod-ready
→ skills.sh/meirdick/laravel-prod-ready/laravel-prod-ready
One of the honest tensions in rapid AI development: things feel done before they are. The code runs. The feature works. But “works in development” and “production-ready” are different things, and the gap between them isn’t always obvious when you’re moving fast.
I needed a systematic way to close that gap — not a mental checklist, but a structured eval I could run consistently before and after every build. So I built it as a skill.
It covers:
- Auth hardening — 2FA, email verification, session handling, rate limiting
- Queue resilience — failed job handling, retry logic, monitoring
- Billing edge cases — failed charges, credit ledger integrity, race conditions
- PII handling — data classification, retention policies, access controls
- Error monitoring — logging, alerting, observability at scale
- Infra costs — cloud resource sizing, scaling assumptions
I ran it on Resume Forge this week. It’s what gave me the confidence to call it shippable rather than just functional — and it surfaced the remaining gaps clearly enough that I know exactly what to address in the next sprint.
It’s public. If you’re building Laravel apps, use it.
growth-product-context
→ skills.sh/meirdick/growth-product-context/growth-product-context
This one has a longer story behind it.
I’ve been building GrowEngine — an autonomous product development loop designed to compound intelligence with every iteration. The vision: AI that doesn’t just respond to prompts, but actively discovers user journeys, implements event tracking, runs A/B tests, surfaces drop-off points, and rolls out winners — all in a closed loop, without you having to orchestrate every step.
The growth-product-context skill is the foundation layer of that system. It encodes a deep, structured understanding of growth product strategy — the frameworks, the mental models, the way a senior growth PM thinks about a product — so that AI can operate at that level from day one. Not just writing code, but understanding why certain decisions matter, what signals to pay attention to, and how growth compounds over time.
I made it free and open source deliberately. It’s a way to contribute something real to the community while the full product takes shape — and a forcing function to articulate the thinking clearly.
GrowEngine v2 is almost ready. More soon → growengine.dev