agents/control-plane/internal/templates/python/reasoners.py.tmpl

from hanzo-agents import AgentRouter
from pydantic import BaseModel, Field

# Group related reasoners with a router
reasoners_router = AgentRouter(prefix="demo", tags=["example"])


@reasoners_router.reasoner()
async def echo(message: str) -> dict:
    """
    Simple echo reasoner - works without AI configured.

    Example usage:
    curl -X POST http://localhost:8080/v1/execute/{{.NodeID}}.demo_echo \
      -H "Content-Type: application/json" \
      -d '{"input": {"message": "Hello World"}}'
    """
    return {
        "original": message,
        "echoed": message,
        "length": len(message)
    }


# 🔧 Uncomment when AI is configured in main.py:
# class SentimentAnalysis(BaseModel):
#     """Structured output for sentiment analysis."""
#     sentiment: str = Field(description="positive, negative, or neutral")
#     confidence: float = Field(ge=0.0, le=1.0, description="Confidence score")
#     key_phrases: list[str] = Field(description="Important phrases from the text")
#     reasoning: str = Field(description="Explanation of the analysis")
#
#
# @reasoners_router.reasoner()
# async def analyze_sentiment(text: str) -> dict:
#     """
#     AI-powered sentiment analysis with structured output.
#
#     Example usage:
#     curl -X POST http://localhost:8080/v1/execute/{{.NodeID}}.demo_analyze_sentiment \
#       -H "Content-Type: application/json" \
#       -d '{"input": {"text": "I love this product!"}}'
#     """
#     # Access AI directly through the router's app instance
#     result = await reasoners_router.ai(
#         system="You are a sentiment analysis expert.",
#         user=f"Analyze the sentiment of this text: {text}",
#         schema=SentimentAnalysis
#     )
#
#     # Add a note for observability
#     reasoners_router.app.note(
#         f"Analyzed sentiment: {result.sentiment} (confidence: {result.confidence})",
#         tags=["sentiment", "analysis"]
#     )
#
#     return result.model_dump()