Platform Engineer
Internal developer platform design: golden paths, self-service infrastructure, Backstage/Port catalogs, developer portal, and platform-as-a-product principles.
A platform engineering skill focused on treating infrastructure as a product for internal developers. It covers golden path templates, self-service provisioning, developer portal setup with Backstage or Port, platform metrics (DORA, developer NPS), and the team topologies required to operate an effective platform without becoming a bottleneck.
Added Mar 20, 2026
$npx skills add johnefemer/skillfish --skill platform-engineer What This Skill Can Do
Concrete capabilities you get when you install this skill.
Design golden path templates for new services with pre-wired CI/CD, observability, and security
Set up self-service infrastructure provisioning with Backstage, Port, or Crossplane
Define platform team operating model: what to own, what to delegate, how to avoid bottlenecks
Measure platform adoption and developer experience with DORA metrics and NPS surveys
Implement service catalog with component metadata, dependencies, and ownership
Build paved roads for database provisioning, secret management, and environment creation
Real Conversations
See how the Platform Engineer skill responds to actual requests. These are representative examples of what your AI agent will produce.
We want developers to be able to spin up a new microservice in under 10 minutes with all our standards pre-applied. What should a golden path template include?
A production golden path template for a new microservice should include everything a team needs to ship safely from day one:
Minimum viable golden path — what to include:
service-template/
├── .github/workflows/
│ ├── ci.yml # lint, test, build, push image
│ └── deploy.yml # deploy to dev on merge, prod on tag
├── k8s/
│ ├── deployment.yaml # resource limits pre-set, liveness/readiness probes
│ ├── service.yaml
│ └── hpa.yaml # horizontal pod autoscaler
├── src/
│ └── main.ts # health endpoint at /healthz already wired
├── Dockerfile # multi-stage, non-root user, .dockerignore
├── opentelemetry.ts # tracing + metrics pre-configured for your stack
├── CODEOWNERS # ownership pattern
└── backstage-template.yaml # Backstage Software Template descriptor
The non-negotiables:
- Health endpoint (
/healthz) responding before template is accepted - Resource
requestsandlimitsset (prevents node starvation) - Default network policy: deny-all ingress except from service mesh
- Secret injection via Vault/AWS Secrets Manager (no env vars with plaintext secrets)
- Observability bootstrapped: metrics endpoint, structured logging, trace context propagation
Backstage scaffolder action to wire it up:
steps:
- id: fetch-template
action: fetch:template
input:
url: ./skeleton
values:
serviceName: ${{ parameters.serviceName }}
owner: ${{ parameters.owner }}
- id: create-repo
action: publish:github
input:
repoUrl: github.com?owner=myorg&repo=${{ parameters.serviceName }}
Target: developer runs backstage scaffold, fills 5 fields, gets a repo with first CI passing in <5 minutes.
Get Started
How to install and use this skill in your preferred environment.
Skills are designed for AI coding agents (Claude Code, Cursor, Windsurf) and IDE-based workflows where the agent can read files, run scripts, and act on your codebase.
Models & Context
Which AI models and context windows work best with this skill.
Recommended Models
Works well with Claude Sonnet or GPT-4o. YAML template generation and platform design benefit from models with strong code and architecture reasoning.
Context Window
SKILL.md is ~10KB. Include current team topology, tool inventory, and sample templates in context for tailored recommendations.
Pro tips for best results
Be specific
Include numbers — users, budget, RPS — so the skill can size the architecture.
Share constraints
Compliance needs, team size, and existing stack all improve the output.
Iterate
Start with a high-level design, then ask follow-ups for IaC, cost analysis, or security review.
Combine skills
Pair with companion skills below for end-to-end coverage.
Works Great With
These skills complement Platform Engineer for end-to-end coverage. Install them together for better results.
Ready to try Platform Engineer?
Install the skill and start getting expert-level guidance in your workflow — any agent, any IDE.
$npx skills add johnefemer/skillfish --skill platform-engineer