Adk Deploy Guide
Deploy Google ADK agents to Agent Engine, Cloud Run, or GKE with Terraform, secrets, CI/CD, and observability without guessing production steps.
Overview
adk-deploy-guide is an agent skill for the Operate phase that walks solo builders through deploying and troubleshooting ADK agents on Google Cloud (Agent Engine, Cloud Run, GKE, CI/CD, secrets, and observability).
Install
npx skills add https://github.com/google/adk-docs --skill adk-deploy-guideWhat is this skill?
- Must-read gate before any ADK deploy—pairs with scaffold `make` pipelines or ADK CLI quick deploy
- Five reference guides: Cloud Run, Agent Engine, GKE Autopilot, Terraform/IAM, and event-driven (Pub/Sub, Eventarc, BigQu
- Covers Agent Engine deploy.py, AdkApp, session types, networking, secrets, and CI/CD differences across targets
- Explicit exclusions: API cheatsheet, evaluation, and scaffolding point to adk-cheatsheet, adk-eval-guide, and adk-scaffo
- Observability cross-links to adk observability skills for production runbooks
- Five dedicated reference topics: Cloud Run, Agent Engine, GKE, Terraform patterns, and event-driven triggers
- Explicit three-skill exclusion list for cheatsheet, eval, and scaffold concerns
Adoption & trust: 2.6k installs on skills.sh; 1.4k GitHub stars; 2/3 security scanners passed (skills.sh audits).
What problem does it solve?
You have a working ADK agent locally but no trusted sequence for GCP targets, Terraform, secrets, or production debugging.
Who is it for?
Indie builders shipping ADK agents to Google Cloud with scaffolded Make/Terraform or ADK CLI who need production-grade deploy and troubleshoot steps in one place.
Skip if: Teams only needing API call patterns, offline evaluation design, or initial project scaffolding without a deploy plan—use adk-cheatsheet, adk-eval-guide, or adk-scaffold instead.
When should I use this skill?
MUST READ before deploying any ADK agent; use when deploying agents to Google Cloud or troubleshooting deployments.
What do I get? / Deliverables
After following the guide you pick a deployment target, apply the matching reference runbook, and wire observability—then use adk-cheatsheet or adk-eval-guide only for the problems this skill explicitly excludes.
- Chosen GCP deployment target runbook (Agent Engine, Cloud Run, or GKE)
- Configured secrets, networking, and session-type choices per reference
- CI/CD or deploy script alignment with ADK production workflows
Recommended Skills
Journey fit
Production deployment and cloud troubleshooting belong on the Operate shelf because they assume you are moving or running agents in GCP, not designing the agent API surface. Infra is the canonical subphase for Cloud Run, GKE, Terraform patterns, Workload Identity, and event-driven triggers on Google Cloud.
How it compares
Use instead of generic “deploy to Cloud Run” chat advice when the workload is ADK-specific Agent Engine or GKE patterns with session and IAM nuances.
Common Questions / FAQ
Who is adk-deploy-guide for?
Solo and small-team builders using Google ADK who are ready to push agents to Agent Engine, Cloud Run, or GKE and need Google’s deployment-oriented procedural knowledge inside their coding agent.
When should I use adk-deploy-guide?
Use it in Operate (infra) before any GCP deploy or when production behavior breaks—after Build (integrations) when the agent code exists, and during Ship when you are validating launch-ready hosting choices.
Is adk-deploy-guide safe to install?
It is Apache-2.0 Google documentation packaged as a skill; review the Security Audits panel on this Prism page and treat cloud deploy steps as operations that can change live infrastructure.
Workflow Chain
Requires first: adk scaffold
SKILL.md
READMESKILL.md - Adk Deploy Guide
# ADK Deployment Guide > **Scaffolded project?** Use the `make` commands throughout this guide — they wrap Terraform, Docker, and deployment into a tested pipeline. > > **No scaffold?** See [Quick Deploy](#quick-deploy-adk-cli) below, or the [ADK deployment docs](https://adk.dev/deploy/). > For production infrastructure, scaffold with `/adk-scaffold`. ### Reference Files For deeper details, consult these reference files in `references/`: - **`cloud-run.md`** — Scaling defaults, Dockerfile, session types, networking - **`agent-engine.md`** — deploy.py CLI, AdkApp pattern, Terraform resource, deployment metadata, CI/CD differences - **`gke.md`** — GKE Autopilot cluster, Terraform-managed Kubernetes resources, Workload Identity, session types, networking - **`terraform-patterns.md`** — Custom infrastructure, IAM, state management, importing resources - **`event-driven.md`** — Pub/Sub, Eventarc, BigQuery Remote Function triggers via custom `fast_api_app.py` endpoints > **Observability:** See the **adk-observability-guide** skill for Cloud Trace, prompt-response logging, BigQuery Analytics, and third-party integrations. --- ## Deployment Target Decision Matrix Choose the right deployment target based on your requirements: | Criteria | Agent Engine | Cloud Run | GKE | |----------|-------------|-----------|-----| | **Languages** | Python | Python | Python (+ others via custom containers) | | **Scaling** | Managed auto-scaling (configurable min/max, concurrency) | Fully configurable (min/max instances, concurrency, CPU allocation) | Full Kubernetes scaling (HPA, VPA, node auto-provisioning) | | **Networking** | VPC-SC and PSC supported | Full VPC support, direct VPC egress, IAP, ingress rules | Full Kubernetes networking| | **Session state** | Native `VertexAiSessionService` (persistent, managed) | In-memory (dev), Cloud SQL, or Agent Engine session backend | In-memory (dev), Cloud SQL, or Agent Engine session backend | | **Batch/event processing** | Not supported | `/invoke` endpoint for Pub/Sub, Eventarc, BigQuery | Custom (Kubernetes Jobs, Pub/Sub) | | **Cost model** | vCPU-hours + memory-hours (not billed when idle) | Per-instance-second + min instance costs | Node pool costs (always-on or auto-provisioned) | | **Setup complexity** | Lower (managed, purpose-built for agents) | Medium (Dockerfile, Terraform, networking) | Higher (Kubernetes expertise required) | | **Best for** | Managed infrastructure, minimal ops | Custom infra, event-driven workloads | Full Kubernetes control | **Ask the user** which deployment target fits their needs. Each is a valid production choice with different trade-offs. --- ## Quick Deploy (ADK CLI) For projects without Agent Starter Pack scaffolding. No Makefile, Terraform, or Dockerfile required. ```bash # Cloud Run adk deploy cloud_run --project=PROJECT --region=REGION path/to/agent/ # Agent Engine adk deploy agent_engine --project=PROJECT --region=REGION path/to/agent/ # GKE (requires existing cluster) adk deploy gke --project=PROJECT --cluster_name=CLUSTER --region=REGION path/to/agent/ ``` All commands support `--with_ui` to deploy the ADK dev UI. Cloud Run also accepts extra `gcloud` flags after `--` (e.g., `-- --no-allow-unauthenticated`). See `adk deploy --help` or the [ADK deployment docs](https://adk.dev/deploy/) for full flag reference. > For CI/CD, observability, or production infrastructure, scaffold with `/adk-scaffold` and use the sections below. --- ## Dev Environment Setup & Deploy (Scaffolded Projec