All skills
google avatar

/agent-platform-alert-configuration

@efcb6f0
by googlegoogle/skills21k stars
1,698

Configures best-practice alerting policies for AI agents using OpenTelemetry (OTel) metrics, generating output as Terraform (.tf) configuration files. Use when analyzing, writing, or deploying alerting policies to monitor agent latency, error rates, token usage, and quality metrics. Don't use for standard infrastructure monitoring unrelated to AI agents, or when the agent is not instrumented with OpenTelemetry (for Reliability, Cost, Safety, Security alerts). NOTE: Reliability, Cost, Safety, and Security alerts use generic OTel metrics and work across runtimes (such as Cloud Run, Vertex AI). Quality alerts rely on Vertex AI Online Monitors and are strictly bound to Vertex AI deployments.

Use this Skill: https://skilld.dev/gh/google/skills/agent-platform-alert-configuration

This session only. Nothing lands on disk.

referencestelemetry_enablement.md

≈631 tokens on demand. Your agent reads this file only when SKILL.md points to it.

Telemetry Status & Enablement Check Reference

Required Telemetry Environment Variables

To successfully export evaluation traces, the agent's deployment spec MUST have the following environment variables:

Environment Variable Required Value Description
GOOGLE_CLOUD_AGENT_ENGINE_ENABLE_TELEMETRY "true" Enables tracing and Cloud Logging export.
OTEL_INSTRUMENTATION_GENAI_CAPTURE_MESSAGE_CONTENT "EVENT_ONLY" or "SPAN_AND_EVENT" or "true" Captures message payloads for evaluation.
OTEL_SEMCONV_STABILITY_OPT_IN "gen_ai_latest_experimental" Opts into Gen AI semantic conventions.

How to Enable Telemetry via Terraform

Update the google_vertex_ai_reasoning_engine resource's deployment_spec.env blocks:

resource "google_vertex_ai_reasoning_engine" "my_agent" {
  ...
  spec {
    deployment_spec {
      env {
        name  = "GOOGLE_CLOUD_AGENT_ENGINE_ENABLE_TELEMETRY"
        value = "true"
      }
      env {
        name  = "OTEL_INSTRUMENTATION_GENAI_CAPTURE_MESSAGE_CONTENT"
        value = "EVENT_ONLY"
      }
      env {
        name  = "OTEL_SEMCONV_STABILITY_OPT_IN"
        value = "gen_ai_latest_experimental"
      }
    }
  }
}

Telemetry and Data Generation Dependencies

Even if telemetry environment variables are configured on the agent, verify these additional dependencies. Note that quality metrics will NOT populate, and quality metrics alert policies and the Online Monitor MUST be skipped if the decision is to NOT enable the required APIs:

  1. API Enablement: Ensure the following APIs are enabled in the GCP project:

    • Cloud Trace API: cloudtrace.googleapis.com (needed for exporting spans)
    • Observability API: observability.googleapis.com (needed for trace storage and SQL queries by the evaluator)

    To verify if they are enabled:

    gcloud services list --enabled --project="{project_id}" \
      --filter="name:(cloudtrace.googleapis.com observability.googleapis.com)"
  2. IAM Permissions: The service account assigned to the agent (default or custom) MUST have the Cloud Trace Agent (roles/cloudtrace.agent) and Logs Writer (roles/logging.logWriter) roles.

Source: SKILL.md on GitHub

No alerts8d3 checks · Risk SAFE
  • Gen Agent Trust Hub8d

    This skill is a well-structured tool for configuring Google Cloud alerting policies for AI agents. It utilizes standard Google Cloud SDKs and CLI tools to discover environment metadata and generate Terraform configurations. The skill incorporates explicit safety tiers, requiring user confirmation for any actions that incur billing or create cloud resources.

  • Socket8d

    No alerts

  • Snyk8d

    Risk: LOW · No issues

Signed by skilld at efcb6f0. This ties the file your Agent reads to that commit on GitHub. It does not review the instructions.

Last checked against GitHub yesterday.

Activeupdated last week
metadata
{
  "version": "1.0.0",
  "category": "AiAndMachineLearning"
}
All 1 allowed tools
terraform gcloud python

README badge

README badge for google/skills/agent-platform-alert-configuration