Can I use Synapse A2A with models other than Claude?

Yes, Synapse A2A is a cross-platform protocol that supports various agent types including Gemini, Codex, Copilot, and OpenCode.

What happens if I send a message to an agent that is currently busy?

You can check agent status using 'synapse list'. If an agent is in PROCESSING status, you can either wait or use a priority level 5 message for an emergency interrupt.

How do I send a message to another agent using Synapse A2A?

You can send messages using the command 'synapse send " " --from '. Use the '--response' flag if you need to wait for a reply.

How do I jump to another agent's active terminal session?

Use the 'synapse jump ' command or use the interactive 'synapse list' TUI and press Enter on the desired agent to focus its terminal window.

How does the File Safety feature prevent conflicts?

The 'synapse file-safety' command allows agents to manage file locks. This ensures that only one agent can modify a specific file at a time, preventing race conditions in multi-agent environments.

Synapse A2A Agent Communication

Name: Synapse A2A Agent Communication
Author: s-hiraoku

bys-hiraoku

Collaboration Tools

Facilitates seamless inter-agent communication and task orchestration between multiple AI models using the Synapse A2A protocol.

About

Synapse A2A provides a robust framework for multi-agent coordination within the Claude Code environment. It allows different AI agents—such as Claude, Gemini, and Codex—to exchange messages, request replies, and manage complex workflows through a standardized communication protocol. With features like priority-based messaging, real-time agent status monitoring via a Rich TUI, file safety locks to prevent write conflicts, and interactive terminal jumping, it transforms isolated AI instances into a collaborative ecosystem capable of handling complex, distributed software engineering tasks.

Key Features

Persistent task history with searchable logs and communication statistics
File safety locking mechanism to prevent multi-agent data corruption
Cross-agent messaging with automated request-response tracking
Priority-based task management with emergency interrupt (SIGINT) support
0 GitHub stars
Interactive TUI for monitoring agent status (READY, WAITING, PROCESSING)

Use Cases

Coordinating a code review where one agent writes code and another specialized agent performs the audit
Managing complex refactoring tasks split across multiple specialized agents to increase throughput
Orchestrating multi-model pipelines where different LLMs handle specific parts of a development lifecycle

About

Key Features

Persistent task history with searchable logs and communication statistics
File safety locking mechanism to prevent multi-agent data corruption
Cross-agent messaging with automated request-response tracking
Priority-based task management with emergency interrupt (SIGINT) support
0 GitHub stars
Interactive TUI for monitoring agent status (READY, WAITING, PROCESSING)

Use Cases

Coordinating a code review where one agent writes code and another specialized agent performs the audit
Managing complex refactoring tasks split across multiple specialized agents to increase throughput
Orchestrating multi-model pipelines where different LLMs handle specific parts of a development lifecycle