R2R
R2R copied to clipboard
Feature/pydantic ai
🚧 Prototype: Adoption of Pydantic AI Framework
This PR introduces a preliminary integration of the Pydantic AI framework as a proof of concept to explore its adoption within our system.
🔎 Overview
- Integrated a dummy tool leveraging the Pydantic AI framework.
- Tested framework extendibility and compatibility with our existing architecture.
- Code is currently rough and experimental, not production-ready.
⚠️ Important Notes
- The implementation is not clean and serves mainly as a sandbox to validate ideas.
- No full refactoring or optimization has been applied yet.
- This is a starting point for further proper implementation.
🎯 Goals
- Assess how Pydantic AI can integrate in current workflow.
- Understand integration challenges and extendibility of the framework.
- Lay groundwork for a cleaner, robust adoption.
🚀 Next Steps
- Refine the integration with a clean, maintainable design.
- Expand test coverage and validation.
- Collaborate with the team to finalize and roll out properly.
Feel free to take this prototype forward and build a solid implementation based on it!
[!IMPORTANT] Prototype integration of Pydantic AI framework with new
RAGPydAgentand tools for enhanced search capabilities.
- Behavior:
- Introduces
RAGPydAgentinrag_pyd.pyto integrate Pydantic AI framework.- Adds
smart_filter_tooland other tools intools/built_into refine search capabilities.- Updates
retrieval_router.pyandretrieval_service.pyto supportrag_pydmode.- Prompts:
- Updates prompts in
dynamic_rag_agent.yaml,dynamic_rag_agent_xml_tooling.yaml,static_rag_agent.yaml, andstatic_research_agent.yamlto include instructions for using new tools.- Misc:
- Adds
pydantic-ai==0.2.16topyproject.tomldependencies.- Updates
SearchSettingsandAggregateSearchResultinsearch.pyto handle new search results.This description was created by
for 32caecf9a2383fee240a05403f89be66ea31c74c. You can customize this summary. It will automatically update as commits are pushed.