Menu
The review-first prompt is selected by default. You can switch to a direct command or download a local copy.
Read SKILL.md and any companion files shown by SkillsMP before deciding whether to install.
Install with Codex or Claude Copy this prompt, paste it into Codex, Claude, or another assistant, and let it review the skill page and install it for you.
A direct command skips the review prompt. Inspect the source before running it.
npx skills add https://github.com/tomevault-io/skills-registry --skill workload-identity-federation-implementationThe command stays on one line. Scroll horizontally to inspect it before copying.
Prefer a local copy? Download the files currently available to SkillsMP.
| Use when this capability is needed.
> Use when this capability is needed.
Review architecture and API design for the vfs-s3 project. Use when the user mentions @architect, asks to review an issue's design, discuss module boundaries, API shape, or architectural decisions for vfs-s3. Also trigger when the user wants to create an ADR (Architecture Decision Record) or evaluate a technical approach for the project. Intended for dispatch from Codex automation or Claude routines; GitHub trigger phrase: @vfs-s3-bot please prepare design doc Use when this capability is needed.
Based on SOC occupation classification
Showing SKILL.md
| name | workload-identity-federation-implementation |
| description | >- Use when this capability is needed. |
from google.cloud import storage
# Credentials automatic
client = storage.Client(project='PROJECT_ID')
bucket = client.bucket('my-bucket')
blob = bucket.blob('data.txt')
blob.download_to_filename('data.txt')
from google.cloud import secretmanager
client = secretmanager.SecretManagerServiceClient()
secret_name = f"projects/PROJECT_ID/secrets/api-key/versions/latest"
response = client.access_secret_version(request={"name": secret_name})
api_key = response.payload.data.decode('UTF-8')
# SERVICE_ACCOUNT_A in PROJECT_A can impersonate SERVICE_ACCOUNT_B in PROJECT_B
gcloud iam service-accounts add-iam-policy-binding \
service-account-b@PROJECT_B.iam.gserviceaccount.com \
--role="roles/iam.serviceAccountUser" \
--member="serviceAccount:service-account-a@PROJECT_A.iam.gserviceaccount.com"
Containers need cloud access. But service account keys are static credentials that never rotate, frequently get stolen, and live forever.
Workload Identity Federation lets containers prove their identity to cloud providers without ever storing keys. The Kubernetes cluster itself becomes a trusted identity provider.
Production Hardening
Workload Identity eliminates the largest attack surface in containerized environments. This is foundational. Get it right.
Instead of storing a static key, your container presents a signed JWT token to prove it's running in your cluster.
| Approach | Token | Rotation | Revocation | Audit |
|---|---|---|---|---|
| Service Account Keys | Static, never changes | Manual | Manual | Weak |
| Workload Identity | Dynamic, short-lived | Automatic | Immediate | Full |
Service account keys are abandoned credentials. Workload Identity is ephemeral proof.
How It Works
- Pod requests token - Kubernetes API issues signed JWT
- Token presented to GCP - GCP validates signature
- GCP issues access token - Short-lived credential for GCP APIs
- Automatic rotation - Token refreshes before expiration
See examples.md for detailed code examples.
This guide is split into focused modules:
See examples.md for detailed code examples.
Verification
Test authentication from inside a pod:
kubectl run -it --image=google/cloud-sdk:slim test-wi \ --serviceaccount=app-sa \ -n production \ -- gcloud auth list
Common Mistakes
- Forgetting to annotate the Kubernetes ServiceAccount
- Using wrong format in IAM binding (
serviceAccount:PROJECT_ID.svc.id.goog[NAMESPACE/SA_NAME])- Not granting
roles/iam.workloadIdentityUserrole- Metadata server enabled on nodes (
workloadMetadataConfig.modemust beGKE_METADATA)
See examples.md for detailed code examples.
Problems:
# Kubernetes ServiceAccount with annotation
apiVersion: v1
kind: ServiceAccount
metadata:
name: app-sa
annotations:
iam.gke.io/gcp-service-account: app-gcp@PROJECT_ID.iam.gserviceaccount.com
Benefits:
See Migration Guide for detailed migration steps.
from google.cloud import storage
# Credentials automatic
client = storage.Client(project='PROJECT_ID')
bucket = client.bucket('my-bucket')
blob = bucket.blob('data.txt')
blob.download_to_filename('data.txt')
from google.cloud import secretmanager
client = secretmanager.SecretManagerServiceClient()
secret_name = f"projects/PROJECT_ID/secrets/api-key/versions/latest"
response = client.access_secret_version(request={"name": secret_name})
api_key = response.payload.data.decode('UTF-8')
# SERVICE_ACCOUNT_A in PROJECT_A can impersonate SERVICE_ACCOUNT_B in PROJECT_B
gcloud iam service-accounts add-iam-policy-binding \
service-account-b@PROJECT_B.iam.gserviceaccount.com \
--role="roles/iam.serviceAccountUser" \
--member="serviceAccount:service-account-a@PROJECT_A.iam.gserviceaccount.com"
Workload Identity eliminates static keys. Tokens rotate automatically. Access revokes immediately. Audit trail complete. Zero-trust credential model in place.
Instead of storing a static key, your container presents a signed JWT token to prove it's running in your cluster.
| Approach | Token | Rotation | Revocation | Audit |
|---|---|---|---|---|
| Service Account Keys | Static, never changes | Manual | Manual | Weak |
| Workload Identity | Dynamic, short-lived | Automatic | Immediate | Full |
Service account keys are abandoned credentials. Workload Identity is ephemeral proof.
How It Works
- Pod requests token - Kubernetes API issues signed JWT
- Token presented to GCP - GCP validates signature
- GCP issues access token - Short-lived credential for GCP APIs
- Automatic rotation - Token refreshes before expiration
See examples.md for detailed code examples.
This guide is split into focused modules:
See examples.md for detailed code examples.
Verification
Test authentication from inside a pod:
kubectl run -it --image=google/cloud-sdk:slim test-wi \ --serviceaccount=app-sa \ -n production \ -- gcloud auth list
Common Mistakes
- Forgetting to annotate the Kubernetes ServiceAccount
- Using wrong format in IAM binding (
serviceAccount:PROJECT_ID.svc.id.goog[NAMESPACE/SA_NAME])- Not granting
roles/iam.workloadIdentityUserrole- Metadata server enabled on nodes (
workloadMetadataConfig.modemust beGKE_METADATA)
See examples.md for detailed code examples.
Problems:
# Kubernetes ServiceAccount with annotation
apiVersion: v1
kind: ServiceAccount
metadata:
name: app-sa
annotations:
iam.gke.io/gcp-service-account: app-gcp@PROJECT_ID.iam.gserviceaccount.com
Benefits:
See Migration Guide for detailed migration steps.
from google.cloud import storage
# Credentials automatic
client = storage.Client(project='PROJECT_ID')
bucket = client.bucket('my-bucket')
blob = bucket.blob('data.txt')
blob.download_to_filename('data.txt')
from google.cloud import secretmanager
client = secretmanager.SecretManagerServiceClient()
secret_name = f"projects/PROJECT_ID/secrets/api-key/versions/latest"
response = client.access_secret_version(request={"name": secret_name})
api_key = response.payload.data.decode('UTF-8')
# SERVICE_ACCOUNT_A in PROJECT_A can impersonate SERVICE_ACCOUNT_B in PROJECT_B
gcloud iam service-accounts add-iam-policy-binding \
service-account-b@PROJECT_B.iam.gserviceaccount.com \
--role="roles/iam.serviceAccountUser" \
--member="serviceAccount:service-account-a@PROJECT_A.iam.gserviceaccount.com"
Workload Identity eliminates static keys. Tokens rotate automatically. Access revokes immediately. Audit trail complete. Zero-trust credential model in place.
See examples.md for code examples.
See reference.md for complete documentation.
Converted and distributed by TomeVault — claim your Tome and manage your conversions.