SKILL.md
gcp-iam
Google Cloud Integration
This skill delegates all GCP provisioning and operations to the official Google Cloud Python client libraries.
# Core GCP client library
pip install google-cloud-python
# Vertex AI + Agent Engine (AI/ML workloads)
pip install google-cloud-aiplatform
# Specific service clients (install only what you need)
pip install google-cloud-bigquery # BigQuery
pip install google-cloud-storage # Cloud Storage
pip install google-cloud-pubsub # Pub/Sub
pip install google-cloud-run # Cloud RunSDK Docs: https://github.com/googleapis/google-cloud-python
Vertex AI SDK: https://cloud.google.com/vertex-ai/docs/python-sdk/use-vertex-ai-python-sdk
Use the Google Cloud Python SDK for all GCP provisioning and operational actions. This skill provides architecture guidance, cost modeling, and pre-flight requirements — the SDK handles execution.
Architecture Guidance
Consult this skill for:
- GCP service selection and trade-off analysis
- Cost estimation and optimization (committed use discounts, sustained use)
- Pre-flight IAM / Workload Identity Federation requirements
- IaC approach (Terraform AzureRM vs Deployment Manager vs Config Connector)
- Integration patterns with Google Workspace and other GCP services
- Vertex AI Agent Engine for multi-agent workflow design
Agent & AI Capabilities
| Capability | Tool |
|---|---|
| LLM agents | Vertex AI Agent Engine |
| Model serving | Vertex AI Model Garden |
| RAG | Vertex AI Search + Embeddings API |
| Multi-agent | Agent Development Kit (google/adk-python) |
| MCP | Vertex AI Extensions (MCP-compatible) |