Cross-Framework Compatibility

The toolbox supports integration with popular AI frameworks:

LangChain Integration

# Import LangChain tool  
from langchain.tools import WikipediaQueryRun  
  
# Convert to BaseTool  
lc_tool = WikipediaQueryRun()  
base_tool = BaseTool()  
base_tool.import_from_langchain(lc_tool)  
  
# Use in gofannon workflows  
result = base_tool.fn("machine learning")  
  
# Export back to LangChain  
exported_tool = base_tool.export_to_langchain()

SmolAgents Integration

# Import SmolAgents tool  
from smolagents.tools import Tool  
  
# Convert to BaseTool  
smol_tool = Tool(...)  
base_tool = BaseTool()  
base_tool.import_from_smolagents(smol_tool)  
  
# Export back to SmolAgents  
exported_tool = base_tool.export_to_smolagents()

Key features:

  • Bi-directional conversion preserves metadata and functionality
  • Automatic schema translation between frameworks
  • Runtime checks for required dependencies

This implementation provides:

  • Full type checking and schema validation
  • Error handling for missing dependencies
  • Bidirectional conversion between frameworks
  • Automatic parameter translation
  • Comprehensive test coverage
  • Documentation examples

The tests verify:

  1. Successful tool conversions
  2. Parameter schema preservation
  3. Functionality equivalence
  4. Error handling for missing dependencies
  5. Round-trip conversions between frameworks