Exposing an AI agent using the A2A protocol > Overview
Overview
The Agent-to-Agent (A2A) Protocol standardizes how agents from different frameworks and providers interact with one another. Use the Informatica A2A adapter to make the AI agents that you create in AI Agent Engineering A2A compliant.
Preview Notice:
Effective in the April 2026 release,
AI Agent Engineering
is available for preview.
Preview functionality is supported for evaluation purposes but is unwarranted and is not supported in production environments or any environment that you plan to push to production. Informatica intends to include the preview functionality in an upcoming release for production use, but might choose not to in accordance with changing market or technical circumstances. Note that if you are working on a preview POD, all data is excluded from SOC 2 compliance coverage. For more information, contact Informatica Global Customer Support.
The AI agents you create in AI Agent Engineering use a proprietary REST interface that A2A clients can’t speak to directly. To make your AI agents A2A compliant, you can use the Informatica A2A adapter.
The adapter acts as a lightweight bridge that exposes your AI agents using A2A so they can be discovered and called by any A2A-compatible client or orchestrator. The adapter receives standard A2A requests, translates them into the format that Informatica expects, processes them, and returns the response in standard A2A format.
The adapter creates the following components for each AI agent that you register:
Agent card
The agent card acts as a business card or manifest that other agents use to discover your agent's capabilities. It includes metadata like the agent’s name, version, detailed description of its skills, and the endpoint URL for sending tasks. Clients use the agent card to discover your AI agent.
The agent card’s format is a simple JSON file hosted at the following endpoint:
GET /<agent name>/.well-known/agent-card.json
Message endpoint
The message endpoint receives A2A message/send requests from external agents. It has the following format:
/<agent name>
To use the adapter, you don’t have to make any code changes. All configuration is done in one Python configuration file.
Additionally, the adapter is stateless. It can run alongside any number of AI agents that you create using AI Agent Engineering.