"""Configuration and agent initialization for agentic search.
This module handles:
- Pydantic AI agent creation and configuration
- OpenAI model setup with API keys
- Model settings (temperature, timeout, retries)
- Service configuration parameters
"""
import logging
from pydantic_ai import Agent
from pydantic_ai.models.openai import OpenAIModel
from pydantic_ai.settings import ModelSettings
from src.config import get_settings
from src.core.constants import LLM_API_TIMEOUT_DEFAULT, MAX_RETRIES_DEFAULT
from src.services.agentic_models import CompletenessEvaluation, URLRankingList
logger = logging.getLogger(__name__)
settings = get_settings()
class AgenticSearchConfig:
"""Configuration container for agentic search agents and parameters."""
def __init__(self) -> None:
"""Initialize Pydantic AI agents and configuration parameters."""
# Create OpenAI model instance
model = OpenAIModel(
model_name=settings.model_choice,
)
# Shared model settings for all agents
# Per Pydantic AI docs: timeout, temperature configured via ModelSettings
self.base_model_settings = ModelSettings(
temperature=settings.agentic_search_llm_temperature,
timeout=LLM_API_TIMEOUT_DEFAULT, # 60s timeout
)
# Create specialized agents for each LLM task
# Per Pydantic AI docs: Agent with output_type for structured outputs
# Agent for evaluating knowledge completeness (Stage 1)
self.completeness_agent = Agent(
model=model,
output_type=CompletenessEvaluation,
output_retries=MAX_RETRIES_DEFAULT, # Retry 3 times for validation errors
model_settings=self.base_model_settings,
)
# Agent for ranking URLs by relevance (Stage 2)
self.ranking_agent = Agent(
model=model,
output_type=URLRankingList,
output_retries=MAX_RETRIES_DEFAULT,
model_settings=self.base_model_settings,
)
# Query refinement agent with custom temperature for creativity (Stage 4)
# Per Pydantic AI docs: Use ModelSettings to override temperature per agent
# Note: output_type defined inline in refinement (dynamic model)
refinement_settings = ModelSettings(
temperature=0.5, # More creative for query generation
timeout=LLM_API_TIMEOUT_DEFAULT,
)
self.refinement_model_settings = refinement_settings
self.openai_model = model # Store for dynamic agent creation
# Store configuration parameters from settings
self.model_name = settings.model_choice
self.temperature = settings.agentic_search_llm_temperature
self.completeness_threshold = settings.agentic_search_completeness_threshold
self.max_iterations = settings.agentic_search_max_iterations
self.max_urls_per_iteration = settings.agentic_search_max_urls_per_iteration
self.max_pages_per_iteration = settings.agentic_search_max_pages_per_iteration
self.max_crawl_depth = settings.agentic_search_max_crawl_depth
self.url_score_threshold = settings.agentic_search_url_score_threshold
self.use_search_hints = settings.agentic_search_use_search_hints
self.enable_url_filtering = settings.agentic_search_enable_url_filtering
self.max_urls_to_rank = settings.agentic_search_max_urls_to_rank
self.max_qdrant_results = settings.agentic_search_max_qdrant_results
logger.info(
"Initialized agentic search configuration with model=%s, threshold=%s, max_iterations=%s",
self.model_name,
self.completeness_threshold,
self.max_iterations,
)
