Overview

How the Agent Client Protocol works

The Agent Client Protocol allows Agents and Clients to communicate by exposing methods that each side can call and sending notifications to inform each other of events.

Communication Model

The protocol follows the JSON-RPC 2.0 specification with two types of messages:

• Methods: Request-response pairs that expect a result or error
• Notifications: One-way messages that don’t expect a response

Message Flow

A typical flow follows this pattern:

1. Initialization Phase

   • Client → Agent: initialize to establish connection
   • Client → Agent: authenticate if required by the Agent
2. Session Setup - either:

   • Client → Agent: session/new to create a new session
   • Client → Agent: session/load to resume an existing session if supported
3. Prompt Turn

   • Client → Agent: session/prompt to send user message
   • Agent → Client: session/update notifications for progress updates
   • Agent → Client: File operations or permission requests as needed
   • Client → Agent: session/cancel to interrupt processing if needed
   • Turn ends and the Agent sends the session/prompt response with a stop reason

Agent

Agents are programs that use generative AI to autonomously modify code. They typically run as subprocesses of the Client.

Baseline Methods

initialize · Schema

Negotiate versions and exchange capabilities..

authenticate · Schema

Authenticate with the Agent (if required).

session/new · Schema

Create a new conversation session.

session/prompt · Schema

Send user prompts to the Agent.

Optional Methods

session/load · Schema

Load an existing session (requires loadSession capability).

session/set_mode · Schema

Switch between agent operating modes.

Notifications

session/cancel · Schema

Cancel ongoing operations (no response expected).

Client

Clients provide the interface between users and agents. They are typically code editors (IDEs, text editors) but can also be other UIs for interacting with agents. Clients manage the environment, handle user interactions, and control access to resources.

Baseline Methods

session/request_permission · Schema

Request user authorization for tool calls.

Optional Methods

fs/read_text_file · Schema

Read file contents (requires fs.readTextFile capability).

fs/write_text_file · Schema

Write file contents (requires fs.writeTextFile capability).

terminal/create · Schema

Create a new terminal (requires terminal capability).

terminal/output · Schema

Get terminal output and exit status (requires terminal capability).

terminal/release · Schema

Release a terminal (requires terminal capability).

terminal/wait_for_exit · Schema

Wait for terminal command to exit (requires terminal capability).

terminal/kill · Schema

Kill terminal command without releasing (requires terminal capability).

Notifications

session/update · Schema

Send session updates to inform the Client of changes (no response expected). This includes: - Message chunks (agent, user, thought) - Tool calls and updates - Plans - Available commands updates - Mode changes

Argument requirements

• All file paths in the protocol MUST be absolute.
• Line numbers are 1-based

Error Handling

All methods follow standard JSON-RPC 2.0 error handling:

• Successful responses include a result field
• Errors include an error object with code and message
• Notifications never receive responses (success or error)

Extensibility

The protocol provides built-in mechanisms for adding custom functionality while maintaining compatibility:

• Add custom data using _meta fields
• Create custom methods by prefixing their name with underscore (_)
• Advertise custom capabilities during initialization

Learn about protocol extensibility to understand how to use these mechanisms.

Next Steps

• Learn about Initialization to understand version and capability negotiation
• Understand Session Setup for creating and loading sessions
• Review the Prompt Turn lifecycle
• Explore Extensibility to add custom features