---
title: Agent-to-Agent (A2A) server in Elastic Agent Builder
description: Learn how to interact with Agent Builder agents from external clients using A2A protocol endpoints and API key authentication.
url: https://docs-v3-preview.elastic.dev/elastic/docs-content/pull/9157/explore-analyze/ai-features/agent-builder/a2a-server
products:
  - Elastic Cloud Serverless
  - Elastic Documentation
  - Elastic Observability
  - Elastic Security
  - Elasticsearch
  - Kibana
applies_to:
  - Serverless Elasticsearch projects: Generally available
  - Serverless Observability projects: Generally available
  - Serverless Security projects: Generally available
  - Elastic Stack: Generally available since 9.3, Preview in 9.2
---

# Agent-to-Agent (A2A) server in Elastic Agent Builder
The [**Agent-to-Agent (A2A) server**](https://github.com/a2aproject/A2A) enables external A2A clients, such as other agent frameworks, to communicate with Elastic Agent Builder agents.

## Retrieve agent metadata (GET)

Returns the agent card for a specific agent. The agent card is the agent's metadata, including its capabilities. The `agentId` is a path parameter that corresponds to your agent IDs:
```
GET /api/agent_builder/a2a/{agentId}.json
```


## Execute A2A protocol (POST)

Interact with agents following the A2A protocol specification:
```
POST /api/agent_builder/a2a/{agentId}
```

<important>
  Both A2A endpoints require API key authentication. Refer to [Create a read-only client key](/elastic/docs-content/pull/9157/explore-analyze/ai-features/agent-builder/api-keys#create-a-read-only-client-key) for a limited key example. For more information about the A2A protocol, refer to the [A2A protocol specification](https://a2aprotocol.ai/docs/guide/a2a-protocol-specification-python#protocol-flow-diagram).
</important>


### Send messages

By default, `message/send` waits until the agent finishes, then returns the complete agent message in a single response.
In 9.2 through 9.5, `message/stream` isn't supported. Refer to [Limitations and known issues](/elastic/docs-content/pull/9157/explore-analyze/ai-features/agent-builder/limitations-known-issues#a2a-streaming-not-supported).

### Send messages without waiting

<applies-to>
  - Elastic Cloud Serverless: Generally available
  - Elastic Stack: Generally available since 9.4
</applies-to>

In 9.4, non-blocking requests require version 9.4.5 or later. On earlier versions, `configuration.blocking: false` is ignored and the request waits for the complete response.
To get a response without waiting for the agent to finish, call `message/send` with `configuration.blocking` set to `false`. The response is a task. Poll it with `tasks/get` until it reaches a final state, then read `status.message`, which contains the agent's response or, if the task failed, the error. For the task states, refer to the [A2A protocol specification](https://a2a-protocol.org/v0.3.0/topics/life-of-a-task/).

### Stream responses

<applies-to>
  - Elastic Cloud Serverless: Generally available
  - Elastic Stack: Planned
</applies-to>

Call the `message/stream` method to receive the response as a stream of server-sent events. The [agent card](#retrieve-agent-metadata-get) reports streaming support by setting `capabilities.streaming` to `true`.
The stream uses the [A2A streaming events](https://a2a-protocol.org/v0.3.0/topics/streaming-and-async/). Artifact updates contain the agent's response text, and status updates in the `working` state report progress, such as tool calls. The final status update is `completed`, `failed`, or `input-required`.
The `input-required` state means the agent is waiting for a response to a prompt, which A2A clients can't send. Refer to [Human-in-the-loop prompts require an interactive conversation](/elastic/docs-content/pull/9157/explore-analyze/ai-features/agent-builder/limitations-known-issues#human-in-the-loop-prompts-require-an-interactive-conversation).
If the client disconnects before the stream ends, the agent stops.