agents.memory_reorganized.search.quick_search.agent¶

Quick Search Agent implementation.

Provides fast, basic search responses optimized for speed and concise answers. Similar to Perplexity’s Quick Search feature.

Classes¶

QuickSearchAgent

Agent for fast, basic search responses.

Module Contents¶

class agents.memory_reorganized.search.quick_search.agent.QuickSearchAgent(name='quick_search_agent', engine=None, search_tools=None, **kwargs)¶

Bases: haive.agents.memory.search.base.BaseSearchAgent

Agent for fast, basic search responses.

Optimized for speed and concise answers. Provides quick factual responses without deep research or complex analysis.

Features: - Fast response times (< 2 seconds target) - Concise, direct answers - Basic source attribution - Memory integration for context - Key extraction

Examples

Basic usage:

agent = QuickSearchAgent(
    name="quick_search",
    engine=AugLLMConfig(temperature=0.1)
)

response = await agent.process_search("What is the capital of France?")
print(response.response)  # "The capital of France is Paris..."

With custom configuration:

agent = QuickSearchAgent(
    name="quick_search",
    engine=AugLLMConfig(
        temperature=0.0,  # Deterministic for facts
        max_tokens=150    # Keep responses short
    )
)

Initialize the Quick Search Agent.

Parameters:

name (str) – Agent identifier
engine (haive.core.engine.aug_llm.AugLLMConfig | None) – LLM configuration (defaults to optimized settings)
search_tools (list[langchain_core.tools.Tool] | None) – Optional search tools
**kwargs – Additional arguments passed to parent

async batch_search(queries)¶

Process multiple quick search queries efficiently.

Parameters:: queries (list[str]) – List of search queries
Returns:: List of quick search responses
Return type:: list[haive.agents.memory.search.quick_search.models.QuickSearchResponse]

determine_answer_type(query)¶

Determine the type of answer needed.

Parameters:: query (str) – The search query
Returns:: Answer type classification
Return type:: str

extract_keywords(query)¶

Extract key terms from the search query.

Parameters:: query (str) – The search query
Returns:: List of key terms
Return type:: list[str]

get_response_model()¶

Get the response model for quick search.

Return type:: type[haive.agents.memory.search.base.SearchResponse]

get_search_instructions()¶

Get specific search instructions for quick search.

Return type:: str

get_system_prompt()¶

Get the system prompt for quick search operations.

Return type:: str

async process_search(query, context=None, save_to_memory=True)¶

Process a quick search query.

Parameters:

query (str) – The search query
context (dict[str, Any] | None) – Optional context
save_to_memory (bool) – Whether to save to memory

Returns:

Quick search response

Return type:

haive.agents.memory.search.quick_search.models.QuickSearchResponse