vamplabAI/sgr-agent-core

Schema-Guided Reasoning (SGR) has agentic system design created by neuraldeep community

What it solves

SGR Agent Core provides a framework for building intelligent research agents that can perform complex reasoning and tool selection. It addresses the need for structured, reliable reasoning in AI agents by implementing a Schema-Guided Reasoning (SGR) approach, allowing agents to handle deep research tasks more effectively than standard tool-calling agents.

How it works

The framework uses a two-phase architecture based on a BaseAgent interface. It combines structured reasoning with flexible tool selection, allowing users to define various agent types such as SGRAgent, ToolCallingAgent, and SGRToolCallingAgent. The system is designed to be extensible, supporting any OpenAI-compatible LLM (including local models) and providing tools for search, reasoning, and clarification.

Who it’s for

Developers building AI research agents, engineers looking for a production-ready agentic framework with OpenAI-compatible API endpoints, and users who want to integrate agents into their own tools via the Agent Client Protocol (ACP) or REST API.

Highlights

  • Schema-Guided Reasoning: Combines structured reasoning with flexible tool selection.
  • OpenAI-Compatible API: Acts as a drop-in replacement for OpenAI API endpoints.
  • Extensible Architecture: Easy to create custom agents and tools with a core library and opt-in extras.
  • Multi-Protocol Support: Supports real-time streaming via SSE, REST API, and the Agent Client Protocol (ACP) for stdio communication.
  • Production Ready: Includes Docker support and comprehensive test coverage.

Related

  • Project
    adongwanai/AgentGuide

    An open-source knowledge base and learning roadmap for AI Agent engineering and research, designed to help developers move from basic LLM usage to building production-grade agents and securing professional AI roles.

  • Project
    VRSEN/agency-swarm

    A framework for building multi-agent AI applications that organizes agent collaboration using real-world organizational structures and directional communication flows.

  • Project
    agentscope-ai/agentscope

    A production-ready agent framework that provides an SDK and a FastAPI backend to build multi-tenant, multi-session agentic applications leveraging LLM reasoning and tool use.

  • Project
    activeagents/activeagent

    Active Agent is a Ruby on Rails gem that introduces an agent‑oriented programming model for building AI‑powered features. Agents act like controllers, support multiple LLM providers (OpenAI, Anthropic, Ollama, etc.), offer prompt templating, streaming, tool calling, delegation, and optional telemetry via the separate `actionagent` dashboard gem.

  • Dispatch
    Building Effective AI Agents: Anthropic's Guide to Agentic Systems

    Anthropic outlines a framework for building AI agents by prioritizing simple, composable patterns over complex frameworks, distinguishing between predictable workflows and autonomous agents.