haive.agents.research.open_perplexity.cli
=========================================

.. py:module:: haive.agents.research.open_perplexity.cli

.. autoapi-nested-parse::

   CLI tool for running the open_perplexity research agent.

   This module provides command-line utilities for running research tasks
   and visualizing research states. It supports loading research questions
   from text files, configuring research parameters, and generating reports.



Functions
---------

.. autoapisummary::

   haive.agents.research.open_perplexity.cli.main
   haive.agents.research.open_perplexity.cli.run_research
   haive.agents.research.open_perplexity.cli.visualize_state


Module Contents
---------------

.. py:function:: main()

   CLI entry point for the research tool.

   Parses command-line arguments and executes the appropriate command
   based on user input.


.. py:function:: run_research(text_path, **kwargs)

   Run a research process on text from the specified file.

   Loads a research question from a text file, configures a research agent with
   the specified parameters, runs the research process, and saves the results.

   :param text_path: Path to a text file containing research question
   :param \*\*kwargs: Additional arguments to pass to the agent, including:
                      - output: Path to save the generated report
                      - save_state_path: Path to save the state history
                      - depth: Research depth (1-5)
                      - concurrent_searches: Number of concurrent searches
                      - max_sources: Maximum sources per query
                      - vector_store_path: Path for the vector store


.. py:function:: visualize_state(state_path, step = None, output_md = None)

   Visualize a specific state from a saved state history file.

   Loads a state from a saved state history file, displays a visualization
   of the research state, and optionally generates a markdown report.

   :param state_path: Path to a saved state history JSON file
   :param step: Index of the state to visualize (default: last state)
   :param output_md: Path to save the markdown report (optional)