Solace Agent Mesh, developed by Solacelabs, is an open-source framework for building and orchestrating multi-agent AI systems. Categorized as an MCP server, it serves as an agent development and runtime platform. Its core runtime is Go, supporting extensions in both Python and Go. The framework connects to the Solace Event Broker as its central messaging backbone, enabling asynchronous communication and real-time data streaming for AI agents. It supports Model Context Protocol and Agent2Agent protocols, making it suitable for enterprise-grade multi-agent workflows.

What does Solace Agent Mesh connect to?

Solace Agent Mesh connects to and uses the Solace Event Broker / Solace Platform (Event Mesh) as its central messaging backbone. This platform enables asynchronous communication between AI agents and for real-time data streaming. It functions as an agent development and runtime platform and an AI agent orchestration framework. The system supports both the Model Context Protocol (MCP) and Agent2Agent (A2A) protocols for agent interactions.

Core features and tools

Solace Agent Mesh provides an Extensible tool system and various Data Connectors for agents to interact with external services. Specific callable tools include MCP Connectors for MCP-compliant servers, OpenAPI Connectors for REST APIs, SQL Connectors for relational databases, and Knowledge Base Connectors for RAG with repositories like Amazon Bedrock Knowledge Bases, S3, web crawlers, Confluence, SharePoint, Salesforce, or Amazon Redshift. CLI commands like "sam init --gui" and "sam run" are also available.

Who is it for?

Solace Agent Mesh is for developers and enterprises focused on building and deploying scalable, multi-agent AI systems. It is particularly suited for enterprise-grade multi-agent workflows. The platform integrates with AI clients and models from major providers including OpenAI, Anthropic, Google, Azure OpenAI, and local models, and works with any LLM. It also integrates with Google's Agent Development Kit. Skip this if your project does not require multi-agent AI orchestration or real-time data integration.

How to get started

Access is granted through a standard OAuth authorization flow, or via JWT, API keys, HTTP Basic or Bearer token authentication. The underlying Solace Event Broker supports HTTP, MQTT, AMQP, WebSocket, and custom protocols for event-driven communication.

Use cases

• Orchestration of AI agents for complex, open-ended tasks, where an LLM-backed Orchestrator dynamically decides agent invocation and sequencing.
• Deterministic workflows that define explicit execution graphs for processes requiring consistent, repeatable steps.
• Real-time data integration and processing, enabling agents to access and process live data streams and integrate with various enterprise data sources, such as an IT ticket processing workflow.

Solace Agent Mesh offers a Go-based framework for orchestrating AI agents, connecting to the Solace Event Broker for messaging. It provides diverse connectors for external services and data, supporting various authentication methods and enterprise identity integration. The platform is designed for scalable, multi-agent AI systems and complex workflows.