Agent Templates¶

agents-cli creates projects from agent templates. Each template provides a working agent with the right dependencies, tools, and project structure for its use case.

Available Templates¶

Template	Description	Use Case
`adk`	ReAct agent using ADK	General-purpose conversational agent with tool use
`adk_a2a`	ADK agent with A2A protocol	Distributed multi-agent systems that communicate across frameworks
`agentic_rag`	ADK agent with RAG pipeline	Document Q&A with automated data ingestion

adk¶

The default template. Creates a ReAct agent using the Agent Development Kit with a sample tool. Start here if you are new to ADK or building a general-purpose agent.

agents-cli create my-agent --agent adk

adk_a2a¶

Extends the base adk template with Agent-to-Agent (A2A) protocol support. Use this when your agent needs to interoperate with agents built on other frameworks (LangGraph, CrewAI, etc.) or when building a distributed multi-agent system.

agents-cli create my-agent --agent adk_a2a

agentic_rag¶

A document Q&A agent with a built-in RAG (Retrieval-Augmented Generation) pipeline. Includes data ingestion infrastructure for indexing documents into a vector store and retrieving them at query time.

agents-cli create my-agent --agent agentic_rag --datastore agent_platform_search

During project creation, choose a datastore backend:

Datastore	Description
`agent_platform_search`	GCS Data Connector with built-in scheduling and ranking
`agent_platform_vector_search`	Kubeflow pipeline with auto-embedding

After creation, provision the datastore and ingest data:

agents-cli infra datastore
agents-cli data-ingestion