---
sourceDocument: Australia Enable AI
sourceDocumentLink: https://www.servicenow.com/docs/r/australia/intelligent-experiences

 Release :

    - australia

ft:locale :

    - en-US

ft:publication_title :

    - Australia Enable AI

ft:clusterId :

    - platai

bundleId :

    - platai

workflow :

    - Platform


---

# Agent2Agent Protocol

# Agent2Agent Protocol {#ariaid-title1}

Release version: Australia  
Updated August 28, 2026  
![](https://www.servicenow.com/docs/portal-asset/ico-clock) 1 minute to read  
Agent2Agent (A2A) is an open standard that enables cross-platform AI agent communication by associating each agent with an Agent Card containing provider information.

## Agent2Agent Protocol overview {#a2a-protocol__a2a-overview}

The standard relies on every AI agent having an Agent Card associated with it. The Agent Card provides basic information for providers like the ServiceNow AI Platform, Microsoft Azure, and Google to use them.  
Note:  
The supported A2A version for on-glide is v0.3 and the supported A2A version when Premium Chat is enabled is v1.

An AI agent's Agent Card uses standardized JSON to help different providers understand its capabilities. The Agent Card is accessed by a specific type of endpoint from a provider's server. Execution plans are communicated through an
execution endpoint so that both the provider's server and the ServiceNow AI Platform can track what the external AI agent is doing.

See [Create an external AI agent](https://www.servicenow.com/docs/2~au39ne15qxlghld~66bQ "Create external AI agents in AI Agent Studio to connect the ServiceNow AI Platform with third-party agentic AI providers.") for instructions for using this protocol to create an AI agent.

## Configuring A2A authentication

A2A connections require authorization from the source platform to execute on the ServiceNow AI Platform. Authentication is established by creating two Connection \& Credential Alias and Connection records, one with an Agent Card endpoint and another with an execution endpoint.

## Custom Headers for External Agent Configuration {#a2a-protocol__custom-headers-external-agent-config}

When configuring external agents, you can pass custom HTTP headers to be included in the API calls to external endpoints. In the External Agent Configuration \[sn_aia_external_agent_configuration\] table, use the custom headers column to specify headers in JSON format. These headers are applied to all HTTP requests made by that agent during offline execution, regardless of the connection type (OAuth, API Key, or other). The configured headers are sent alongside the agent's authentication credentials and are useful for passing additional metadata or requirements that your external API expects.  
Note:  
Custom headers apply only to the Premium Chat experience.

## Configuration Properties

The following properties configure different aspects of how your agents can interact with the ServiceNow AI Platform.  
{#a2a-protocol__entry__2}

| Name | Description |
|-|-|
| sn_aia.external_agents.enabled | Set to <kbd class="ph userinput">true</kbd> to enable external agent features |
| sn_aia.external_agents.parallel_conversations.enabled | Enables or disables multiple simultaneous conversations per user |
[Table 1. External AI agent properties]