Cloud Build Skill

Manage and analyze Google Cloud Build using gcloud builds commands.

Discovery-First Rule

ALWAYS discover before acting. Never assume trigger names, build IDs, or worker pool names.

# Discover build triggers
gcloud builds triggers list --format=json \
  | jq '[.[] | {name: .name, id: .id, description: .description, disabled: .disabled, eventType: (if .github then "github" elif .triggerTemplate then "csr" elif .pubsubConfig then "pubsub" else "manual" end), createTime: .createTime}]'

Parallel Execution Requirement

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

for trigger_id in $(gcloud builds triggers list --format="value(id)"); do
  {
    gcloud builds triggers describe "$trigger_id" --format=json
  } &
done
wait

Helper Functions

# Get recent builds
list_builds() {
  local limit="${1:-25}"
  gcloud builds list --limit="$limit" --format=json \
    | jq '[.[] | {id: .id, status: .status, startTime: .startTime, duration: .duration, source: .substitutions.REPO_NAME // .source.repoSource.repoName, branch: .substitutions.BRANCH_NAME, trigger: .buildTriggerId}]'
}

# Get build details
get_build_details() {
  local build_id="$1"
  gcloud builds describe "$build_id" --format=json \
    | jq '{id: .id, status: .status, startTime: .startTime, finishTime: .finishTime, duration: .duration, steps: [.steps[] | {name: .name, status: .status, timing: .timing}], artifacts: .artifacts, logUrl: .logUrl}'
}

# List worker pools
list_worker_pools() {
  local region="$1"
  gcloud builds worker-pools list --region="$region" --format=json \
    | jq '[.[] | {name: .name, state: .state, machineType: .privatePoolV1Config.workerConfig.machineType, diskSizeGb: .privatePoolV1Config.workerConfig.diskSizeGb, network: .privatePoolV1Config.networkConfig}]'
}

# Get build logs
get_build_logs() {
  local build_id="$1"
  gcloud builds log "$build_id" 2>&1 | tail -100
}

Common Operations

1. Build Health Overview

# Recent builds with status summary
gcloud builds list --limit=50 --format=json \
  | jq '{total: length, success: [.[] | select(.status=="SUCCESS")] | length, failure: [.[] | select(.status=="FAILURE")] | length, timeout: [.[] | select(.status=="TIMEOUT")] | length, cancelled: [.[] | select(.status=="CANCELLED")] | length, builds: [.[:10][] | {id: .id[:8], status: .status, duration: .duration, trigger: .buildTriggerId}]}'

2. Trigger Management

# List all triggers with config
gcloud builds triggers list --format=json \
  | jq '[.[] | {name: .name, id: .id, disabled: .disabled, github: .github, triggerTemplate: .triggerTemplate, filename: .filename, includedFiles: .includedFiles, ignoredFiles: .ignoredFiles, substitutions: .substitutions}]'

# Check disabled triggers
gcloud builds triggers list --format=json \
  | jq '[.[] | select(.disabled == true) | {name: .name, id: .id}]'

3. Build History and Failure Analysis

# Recent failures
gcloud builds list --filter="status=FAILURE" --limit=10 --format=json \
  | jq '[.[] | {id: .id, startTime: .startTime, duration: .duration, failedStep: (.steps[] | select(.status=="FAILURE") | .name), trigger: .buildTriggerId}]'

# Get failure details for a specific build
get_build_details "$BUILD_ID"
get_build_logs "$BUILD_ID"

4. Worker Pool Configuration

# List worker pools across regions
for region in us-central1 us-east1 europe-west1; do
  {
    echo "Region: $region"
    list_worker_pools "$region"
  } &
done
wait

5. Artifact Management

# Check artifact output from recent builds
gcloud builds list --limit=10 --format=json \
  | jq '[.[] | select(.artifacts) | {id: .id[:8], images: .images, artifacts: .artifacts.objects.location}]'

# List images built
gcloud builds list --limit=20 --format=json \
  | jq '[.[] | select(.images) | {id: .id[:8], images: .images, status: .status}]'

Output Format

Present results as a structured report:

Gcp Cloud Build 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. Build timeout: Default timeout is 10 minutes (600s). Long builds silently fail with TIMEOUT status. Check and set --timeout per trigger.
2. Machine type: Default machine type is e2-standard-2. CPU-intensive builds benefit from e2-highcpu-8 or e2-highcpu-32. Check step durations.
3. Source access: Private GitHub repos require Cloud Build GitHub app installation. CSR repos need IAM permissions. Check source configuration on failures.
4. Step ordering: Steps run sequentially by default. Use waitFor with step IDs for parallel step execution in cloudbuild.yaml.
5. Service account: Cloud Build uses a default service account with broad permissions. For least-privilege, assign a custom service account per trigger.