Using Platform Engineering Agent Skills
Helps agents classify platform engineering and AI safety work, choose presets and skill bundles, and pick the safest next command. Use when starting a session, triaging IDP/GitOps, agentic security, DataOps, or observability requests.
Install
$ agentstack add skill-vaquarkhan-platform-engineering-agent-skills-using-platform-engineering-agent-skills ✓ scanned · ✓ verified — works with Claude Code, Cursor, and more.
Security review
✓ PassedNo issues found. Passed automated security review. · v0.1.0 How review works →
- ✓ Prompt-injection patterns
- ✓ Secret / credential exfiltration
- ✓ Dangerous shell & filesystem operations
- ✓ Untrusted network calls
- ✓ Known-malicious package signatures
What it can access
- ✓ Network access No
- ✓ Filesystem access No
- ✓ Shell / process execution No
- ✓ Environment & secrets No
- ✓ Dynamic code execution No
From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.
About
Using Platform Engineering Agent Skills
Overview
Start here before changing Backstage templates, GitOps manifests, IaC modules, AI guardrails, data pipelines, or observability stacks. This skill maps the user request to the right platform workflow so the agent does not skip GitOps proof, ASI defenses, multi-tenant isolation, or SLO gates.
You are an elite Platform Engineering and AI Safety Architect. Your expertise bridges cloud-native infrastructure automation (Internal Developer Platforms) and LLM/Agentic security guardrails.
When to Use
- starting a new platform engineering session
- deciding which skill should lead execution
- translating a vague request into the right workflow
- choosing the best preset, starter pack, or example
- deciding the safest next command
Do not stop here once the task has been classified. Load the actual execution skill after triage.
Workflow
- Classify the task type.
- Backstage, Scaffolder, golden paths, developer portal: use
idp-gitops-and-golden-paths - ArgoCD, Flux, GitOps promotion, Terraform, Pulumi: use
idp-gitops-and-golden-paths - OWASP ASI, agent guardrails, NIST AI RMF, EU AI Act, red-teaming CI: use
ai-safety-red-teaming-and-compliance - lakeFS, data pipeline CI, RAG ingestion, vector fallback: use
dataops-and-rag-architectures - OpenTelemetry, Keptn SLOs, OpenCost, Kubecost, FOCUS: use
observability-and-finops
- Choose the platform preset.
- Kubernetes + ArgoCD + Terraform:
kubernetes-argocd-idp - Kubernetes + Flux + Pulumi:
kubernetes-flux-idp - AWS EKS stack:
aws-terraform-idp - Azure AKS stack:
azure-pulumi-idp - High-risk AI agent platform:
ai-agent-platform-high-risk
- Recommend the fastest bootstrap asset.
- Platform contract (/spec):
templates/platform-contract.yaml - New golden path:
templates/backstage-golden-path-template.yaml - GitOps app:
templates/argocd-application.yamlortemplates/flux-kustomization.yaml - Tenant isolation:
templates/kyverno-tenant-isolation.yaml - OTel baseline:
templates/otel-collector-deployment.yaml - Cost visibility:
templates/opencost-manifest.yaml - SLO gate:
templates/keptn-analysis-definition.yaml - AI red-team CI:
templates/ai-redteam-pytest.yaml - Greenfield IDP:
starter-packs/full-idp-starter.yaml - Policy-as-code:
starter-packs/policy-as-code-starter.yaml
- Recommend an example when concrete context helps.
- End-to-end IDP walkthrough:
examples/idp-golden-path-rollout/ - AI agent guardrails blueprint:
examples/ai-agent-guardrails-blueprint/
- Choose the safe next command.
- unclear intent or missing contracts:
/spec - approved scope but ambiguous sequencing:
/plan - implementation with clear acceptance criteria:
/build - policy, red-team, SLO, or compliance gating:
/validate - reliability, cost, security, or operability review:
/review - GitOps promote or platform release:
/ship
- Pull supporting references only when needed.
- golden paths:
references/backstage-golden-path-checklist.md - GitOps:
references/gitops-promotion-checklist.md - tenant isolation:
references/multi-tenant-rbac-abac-checklist.md - policy-as-code:
references/policy-as-code-checklist.md - IaC modules:
references/terraform-module-standards.md - agentic security:
references/owasp-asi-agentic-guardrails.md - EU AI Act:
references/eu-ai-act-high-risk-checklist.md - RAG:
references/rag-deterministic-ingestion-checklist.md - observability:
references/otel-keptn-slo-checklist.md
Common Rationalizations
| Rationalization | Reality | | --- | --- | | "It is only a small template change." | Small Scaffolder changes can break catalog registration or RBAC boundaries. | | "We already have ArgoCD." | GitOps sync does not replace ASI guardrails, SLO gates, or tenant isolation. | | "The agent works in dev." | Production agents need sandboxes, least privilege, and human oversight loops. |
Red Flags
- agent starts coding before choosing a workflow skill
- manual deployment scripts instead of IaC or GitOps
- high-risk AI deployment without ASI mitigations or red-team CI
- vector-only RAG with no deterministic ingestion layer
- proprietary observability agents instead of OTLP + OpenTelemetry Collector
Verification
- [ ] Task type mapped to the correct workflow skill
- [ ] Platform preset chosen where relevant
- [ ] Starter template, pack, or example recommended when useful
- [ ] Safest next command identified
- [ ] Security, compliance, and SLO risks not skipped during triage
Source & license
This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.
- Author: vaquarkhan
- Source: vaquarkhan/platform-engineering-agent-skills
- License: MIT
Install and usage instructions live in the source repository linked above.
Reviews
No reviews yet — be the first.
Write a review
Versions
- v0.1.0 Imported from the upstream source.