micro/go-micro

What it solves

Go Micro is an agent harness and service framework for Go that simplifies the process of building and operating AI agents. It moves beyond simple LLM loops by providing the necessary infrastructure (the "harness") to connect agents to real-world tools, manage their memory, enforce guardrails, and enable them to communicate with other agents across different frameworks.

How it works

Go Micro treats agents and tools as distributed services. Every Go service endpoint becomes an AI-callable tool automatically via metadata and RPC. Agents are implemented as services with a standard Agent.Chat endpoint, allowing them to be discovered, load-balanced, and called like any other service. The framework provides a pluggable architecture for LLM providers, durable store-backed memory, and a coordination layer for multi-agent systems.

Who it’s for

It is designed for Go developers who want to build AI-native applications where agents must operate actual systems, manage state, and coordinate complex tasks rather than just generating text.

Highlights

• Automatic Tooling: Service endpoints are automatically converted into tool schemas for AI agents.
• Agent Coordination: Built-in plan and delegate capabilities allow agents to manage multi-step work and hand off tasks to other specialized agents.
• Interoperability: Supports the Model Context Protocol (MCP) for tools and the Agent2Agent (A2A) protocol for cross-framework agent communication.
• AI-Driven Development: Includes a CLI that can generate entire services and handlers from a natural language prompt.
• Durable Execution: Provides event-driven workflows with durable steps that can resume from saved boundaries after interruptions.
• Monetization: Integrated support for the x402 payment standard, allowing tools to require autonomous stablecoin payments.
• Pluggable Infrastructure: Swappable backends for service registries (Consul, etcd), pub/sub (NATS, RabbitMQ), and KV stores (Postgres, SQLite).