SGR Agent Core is an open-source framework for building intelligent AI research agents based on a methodology known as Schema-Guided Reasoning (SGR). The framework provides a core library that allows developers to design autonomous agents capable of structured reasoning and complex task execution. Instead of relying solely on free-form prompts, the system organizes reasoning processes around schemas that guide how agents analyze problems, gather information, and generate outputs. This architecture enables agents to follow structured reasoning workflows while still benefiting from the flexibility of large language models. The framework includes a BaseAgent interface and a two-phase architecture that separates reasoning planning from execution, allowing developers to implement custom agent behaviors and research pipelines.
Features
- Schema-Guided Reasoning architecture for structured agent decision processes
- BaseAgent interface enabling custom research agent implementations
- Two-phase architecture separating reasoning planning from task execution
- Modular framework allowing extension with additional tools and reasoning modules
- Prebuilt research agent implementations demonstrating practical workflows
- Integration with language models for structured research and information analysis