Cloud Deploy Skill

Manage and analyze Google Cloud Deploy pipelines using gcloud deploy commands.

Discovery-First Rule

ALWAYS discover before acting. Never assume pipeline names, release names, rollout names, or target names.

# Discover delivery pipelines
gcloud deploy delivery-pipelines list --format=json \
  | jq '[.[] | {name: .name | split("/") | last, uid: .uid, createTime: .createTime, stages: [.serialPipeline.stages[] | {target: .targetId, strategy: .strategy.standard.verify // .strategy.canary}]}]'

Parallel Execution Requirement

ALL independent operations MUST run in parallel using background jobs (&) and wait.

for pipeline in $(gcloud deploy delivery-pipelines list --format="value(name)" | xargs -I{} basename {}); do
  {
    gcloud deploy delivery-pipelines describe "$pipeline" --region="$REGION" --format=json
  } &
done
wait

Helper Functions

# Get pipeline details
get_pipeline() {
  local pipeline="$1" region="$2"
  gcloud deploy delivery-pipelines describe "$pipeline" --region="$region" --format=json \
    | jq '{name: .name | split("/") | last, stages: [.serialPipeline.stages[] | {target: .targetId, profiles: .profiles, strategy: .strategy}], condition: .condition}'
}

# List releases for a pipeline
list_releases() {
  local pipeline="$1" region="$2" limit="${3:-10}"
  gcloud deploy releases list --delivery-pipeline="$pipeline" --region="$region" --format=json --limit="$limit" \
    | jq '[.[] | {name: .name | split("/") | last, createTime: .createTime, renderState: .renderState, condition: .condition, targetSnapshots: [.targetSnapshots[]? | .name | split("/") | last]}]'
}

# List rollouts for a release
list_rollouts() {
  local release="$1" pipeline="$2" region="$3"
  gcloud deploy rollouts list --release="$release" --delivery-pipeline="$pipeline" --region="$region" --format=json \
    | jq '[.[] | {name: .name | split("/") | last, target: .targetId, state: .state, approvalState: .approvalState, deployStartTime: .deployStartTime, deployEndTime: .deployEndTime, failureReason: .failureReason}]'
}

# List targets
list_targets() {
  local region="$1"
  gcloud deploy targets list --region="$region" --format=json \
    | jq '[.[] | {name: .name | split("/") | last, gke: .gke, cloudRun: .run, executionConfigs: .executionConfigs}]'
}

Common Operations

1. Pipeline Overview

pipelines=$(gcloud deploy delivery-pipelines list --region="$REGION" --format="value(name)" | xargs -I{} basename {})
for pipeline in $pipelines; do
  {
    echo "=== Pipeline: $pipeline ==="
    get_pipeline "$pipeline" "$REGION"
    list_releases "$pipeline" "$REGION" 5
  } &
done
wait

2. Release Management

# Latest release details
gcloud deploy releases list --delivery-pipeline="$PIPELINE" --region="$REGION" --limit=1 --format=json \
  | jq '.[0] | {name: .name | split("/") | last, renderState: .renderState, buildArtifacts: .buildArtifacts, deliveryPipelineSnapshot: .deliveryPipelineSnapshot.serialPipeline.stages | length}'

# Release render status
gcloud deploy releases describe "$RELEASE" --delivery-pipeline="$PIPELINE" --region="$REGION" --format=json \
  | jq '{renderState: .renderState, renderStartTime: .renderStartTime, renderEndTime: .renderEndTime, targetRenders: .targetRenders}'

3. Rollout Status

# Current rollouts across all releases
releases=$(gcloud deploy releases list --delivery-pipeline="$PIPELINE" --region="$REGION" --limit=5 --format="value(name)" | xargs -I{} basename {})
for release in $releases; do
  {
    echo "Release: $release"
    list_rollouts "$release" "$PIPELINE" "$REGION"
  } &
done
wait

# Detailed rollout status
gcloud deploy rollouts describe "$ROLLOUT" --release="$RELEASE" --delivery-pipeline="$PIPELINE" --region="$REGION" --format=json \
  | jq '{state: .state, approvalState: .approvalState, targetId: .targetId, phases: .phases, metadata: .metadata}'

4. Approval Workflows

# Find rollouts pending approval
for pipeline in $(gcloud deploy delivery-pipelines list --region="$REGION" --format="value(name)" | xargs -I{} basename {}); do
  {
    releases=$(gcloud deploy releases list --delivery-pipeline="$pipeline" --region="$REGION" --limit=3 --format="value(name)" | xargs -I{} basename {})
    for release in $releases; do
      gcloud deploy rollouts list --release="$release" --delivery-pipeline="$pipeline" --region="$REGION" --format=json \
        | jq --arg p "$pipeline" --arg r "$release" '[.[] | select(.approvalState=="NEEDS_APPROVAL") | {pipeline: $p, release: $r, rollout: .name | split("/") | last, target: .targetId}]'
    done
  } &
done
wait

5. Target Configuration

# List all targets with their runtime config
list_targets "$REGION"

# Check target execution environment
gcloud deploy targets describe "$TARGET" --region="$REGION" --format=json \
  | jq '{name: .name | split("/") | last, gke: .gke, run: .run, requireApproval: .requireApproval, executionConfigs: [.executionConfigs[] | {usages: .usages, workerPool: .workerPool, serviceAccount: .serviceAccount, artifactStorage: .artifactStorage}]}'

Output Format

Present results as a structured report:

Gcp Cloud Deploy Report
═══════════════════════
Resources discovered: [count]

Resource       Status    Key Metric    Issues
──────────────────────────────────────────────
[name]         [ok/warn] [value]       [findings]

Summary: [total] resources | [ok] healthy | [warn] warnings | [crit] critical
Action Items: [list of prioritized findings]

Target ≤50 lines of output. Use tables for multi-resource comparisons.

Anti-Hallucination Rules

1. NEVER assume resource names — always discover via CLI/API in Phase 1 before referencing in Phase 2.
2. NEVER fabricate metric names or dimensions — verify against the service documentation or --help output.
3. NEVER mix CLI commands between service versions — confirm which version/API you are targeting.
4. ALWAYS use the discovery → verify → analyze chain — every resource referenced must have been discovered first.
5. ALWAYS handle empty results gracefully — an empty response is valid data, not an error to retry.

Counter-Rationalizations

Shortcut	Counter	Why
"I'll skip discovery and check known resources"	Always run Phase 1 discovery first	Resource names change, new resources appear — assumed names cause errors
"The user only asked for a quick check"	Follow the full discovery → analysis flow	Quick checks miss critical issues; structured analysis catches silent failures
"Default configuration is probably fine"	Audit configuration explicitly	Defaults often leave logging, security, and optimization features disabled
"Metrics aren't needed for this"	Always check relevant metrics when available	API/CLI responses show current state; metrics reveal trends and intermittent issues
"I don't have access to that"	Try the command and report the actual error	Assumed permission failures prevent useful investigation; actual errors are informative

Common Pitfalls

1. Render vs deploy: A release must render successfully before rollouts can begin. Check renderState before investigating rollout failures.
2. Approval timeout: Rollouts pending approval do not time out by default. Stale approvals can block the pipeline indefinitely.
3. Target ordering: Pipeline stages are sequential. A rollout to stage N requires successful completion of stage N-1. Cannot skip stages.
4. Skaffold version: Cloud Deploy uses Skaffold for rendering. Skaffold version mismatches between local development and Cloud Deploy can cause render failures.
5. Service account permissions: The execution service account needs permissions to deploy to GKE or Cloud Run targets. Check IAM bindings on the target.