Google

Guide AI agent development from requirements and project setup through implementation, behavior evaluation, deployment, and monitoring. Provides coding rules, debugging steps, and links to specialized skills so each phase follows a defined process.

• Planning an agent's requirements and architecture
• Running and debugging an agent locally
• Evaluating responses and tool use before deployment

Set up monitoring for deployed agents to inspect execution paths, latency, and model interactions. It explains Cloud Trace, prompt and response logging, BigQuery analytics, and external monitoring integrations, with checks for missing data and unwanted content capture.

• Investigating agent latency and execution errors
• Configuring prompt and response logging
• Analyzing agent events in BigQuery

Create starter projects for AI agents, add deployment configuration and automated build and release pipelines to existing projects, or update generated project files. Supports Python and Go ADK templates and a prototype-first approach.

• Starting a Python or Go agent prototype
• Adding deployment configuration to an existing project
• Adding an automated build and release pipeline

Plan and carry out agent deployments on Google Cloud using agents-cli. The guide covers deployment targets, permissions, secrets, automated build and deployment pipelines, verification, and recovery procedures for agents built with supported frameworks.

• Comparing Agent Runtime, Cloud Run, and GKE
• Setting up automated build and deployment pipelines
• Configuring secrets and service account permissions

Evaluate AI agents by preparing test cases, scoring responses and tool use, investigating failures, and comparing results after changes. Supports built-in checks and custom scoring rules, including evaluations performed by a language model.

• Create test cases for an AI agent
• Check response quality and tool use
• Investigate failed evaluation cases

Register a deployed agent with Gemini, choose the appropriate registration method, and troubleshoot setup problems. Also covers managing agents and registering MCP servers, which expose tools to agents, in Google Cloud Agent Registry.

• Registering a deployed agent with Gemini
• Automating registration in a release pipeline
• Listing and updating registered agents

Find API patterns and examples for coding agents with ADK, the Agent Development Kit. Covers tools, state, callbacks, multi-agent coordination, and graph workflows, with Python and Go references plus guidance on studying existing implementations.

• Adding tools and callbacks to an ADK agent
• Implementing state management and agent coordination
• Designing branching and parallel graph workflows

Guide development of a LangChain or LangGraph project created with agents-cli. It explains tool additions, local runs, evaluation, and serving while identifying the required graph structure and commands that differ from ADK projects.

• Adding tools to an existing agent
• Running and evaluating an agent locally
• Editing a graph while preserving its required interface