google-agents-cli-observability

Observability Guide

Cloud Trace works out of the box — no infrastructure needed. Prompt-response logging and BigQuery Agent Analytics require Terraform-provisioned infrastructure (service account, GCS bucket, BigQuery dataset). Run agents-cli infra single-project --apply --project PROJECT_ID to provision these resources. Go projects get the BigQuery telemetry stack too; the GCS completion upload behind prompt-response logging and the BigQuery Agent Analytics plugin are Python only. See references/cloud-trace-and-logging.md for details, env vars, and verification commands. If your project isn't scaffolded yet, see /google-agents-cli-scaffold first.

Deployment order

If you already ran agents-cli deploy imperatively: Do not apply Terraform afterward. For Agent Runtime, Terraform owns the whole engine (service account, deployment spec, env vars), so a later apply can't be reconciled without taking over the resource. Either delete the deployment and start over, or keep the SDK-deployed instance — skip infra single-project and set the observability env vars by re-running agents-cli deploy --update-env-vars "KEY=VALUE,..."; deploy matches the existing Reasoning Engine by display name and updates it in place, preserving env vars set outside the deploy. You must also grant its service account the telemetry IAM roles the Terraform module would otherwise provision: roles/storage.admin (write completions to the logs bucket), roles/logging.logWriter, roles/cloudtrace.agent, plus roles/bigquery.dataOwner + roles/bigquery.jobUser when scaffolded with --bq-analytics. The full set lives in deployment/terraform/single-project/iam.tf (from app_sa_roles) and telemetry.tf. Terraform-managed env vars aren't available in this mode.

Reference Files

Observability Tiers

Choose the right level of observability based on your needs:

Ask the user which tier(s) they need — they can be combined. Cloud Trace is always on; the others are additive.

Cloud Trace

Scaffolded agents use OpenTelemetry to emit distributed traces. Every agent invocation produces spans that track the full execution flow.

Span Hierarchy

ADK projects. These are ADK's span names; other frameworks emit their own (generate_content comes from the shared google-genai instrumentor either way).

invoke_workflow (top-level run)
  └── invoke_agent (one per agent in the chain)
        ├── call_llm (model request)
        │     └── generate_content (underlying GenAI model call)
        └── execute_tool (tool execution)

Setup by Deployment Type

Wired at app startup — ADK Python: get_fast_api_app(otel_to_cloud=True) in app/fast_api_app.py; ADK Go: setupObservability() in observability.go; other templates call their own. references/cloud-trace-and-logging.md has the details.

View traces: Cloud Console → Trace → Trace explorer

ADK: for detailed setup instructions (Agent Runtime CLI/SDK, Cloud Run, custom deployments), fetch https://adk.dev/integrations/cloud-trace/index.md.

Prompt-Response Logging

Captures GenAI interactions and exports to GCS (JSONL) and BigQuery (via log sinks + external tables). Content is governed by two independent tiers; the net Terraform-deploy default is full content in GCS/BigQuery, none in traces:

The tiers are independent: GCS/BigQuery uploads capture full content whenever their upload vars are set and do not honor OTEL_INSTRUMENTATION_GENAI_CAPTURE_MESSAGE_CONTENT, which governs the traces/events tier only.

ADK Python reads it as the experimental-semconv enum:

• NO_CONTENT — no content in spans/events (scaffolded default)
• EVENT_ONLY — content in Cloud Logging events
• SPAN_ONLY / SPAN_AND_EVENT — content in trace spans
• true / false — invalid; fall back to NO_CONTENT

ADK Go reads the same variable as a boolean: "1" or "true" capture content, every other value — including the enum members above — elides it.

For the full mechanics (semconv opt-in, declarative Terraform config, env-var table, enabling/disabling, verification commands), see references/cloud-trace-and-logging.md. For ADK logging docs (log levels, configuration, debugging), fetch https://adk.dev/observability/logging/index.md.

BigQuery Agent Analytics Plugin

ADK projects. Optional ADK plugin that logs structured agent events to BigQuery. Enable with --bq-analytics at scaffold time. See references/bigquery-agent-analytics.md for details.

Third-Party Integrations

Many third-party observability platforms can ingest agent telemetry (via OpenTelemetry or custom instrumentation). The table below covers common ones; the full list is larger (see the pointer below it).

Ask the user which platform they prefer — present the trade-offs and let them choose. ADK: fetch a platform's setup page at https://adk.dev/integrations/<slug>/index.md (slugs for the table above: agentops, arize-ax, phoenix, mlflow-tracing, monocle, weave, freeplay); ADK has more observability integrations (Datadog, Galileo, LangWatch, Latitude, Future AGI, Respan, Zespan, …) — browse the complete, current list at https://adk.dev/integrations/ (observability topic). On other frameworks the OpenTelemetry-based platforms still work, but follow the platform's own setup docs.

Troubleshooting

Related Skills

• /google-agents-cli-deploy — Deployment targets, CI/CD pipelines, and production workflows
• /google-agents-cli-workflow — Development workflow, coding guidelines, and operational rules
• /google-agents-cli-adk-code — ADK API quick reference for writing agent code, Python and Go