Ingenimax/agent-sdk-go

What it solves

Agent Go SDK provides a framework for building production-ready AI agents in Go. It simplifies the process of integrating multiple LLM providers, managing conversation memory, executing tools, and deploying agents with enterprise-grade requirements like multi-tenancy and observability.

How it works

The SDK uses a modular architecture that allows developers to define agents and tasks either programmatically in Go or declaratively using YAML files. It integrates with LLMs from OpenAI, Anthropic, and Google Vertex AI, and supports the Model Context Protocol (MCP) for connecting to external servers. Memory is handled via buffer or vector-based retrieval, with optional Redis support for distributed environments. It also includes a structured task framework for planning and executing multi-step operations.

Who it’s for

Go developers who want to build scalable, enterprise-ready AI agents without having to manually implement the plumbing for memory, tool integration, and multi-model support.

Highlights

• Multi-Model Support: Works with OpenAI, Anthropic, and Google Vertex AI.
• Declarative Configuration: Define agents, tasks, and structured JSON output formats using YAML.
• Enterprise Features: Includes built-in guardrails, multi-tenancy, and integrated tracing/logging.
• MCP Integration: Supports Model Context Protocol (MCP) servers via HTTP and stdio.
• Memory Management: Offers persistent conversation tracking with buffer and vector options.
• Token Tracking: Built-in counting for cost monitoring and usage analytics.