.. Curaitor Agent documentation master file

Welcome to Curaitor Agent’s documentation!
==========================================

Curaitor Agent is an **AI-powered tool** for extracting, organizing, and processing scientific data.  
The project was developed as part of the **Schmidt OxRSE Workshop (Sep 11–20, 2025)** and is designed to support researchers and data enthusiasts in managing complex datasets with ease.  

Curaitor Agent provides:

- A **web interface** for running the agent.
- Integration with the **Model Context Protocol (MCP) inspector** to test tools and server connections.

To get started with Curaitor Agent, check out the :doc:`quickstart` page.  
For more detailed information, see the :doc:`installation`, :doc:`mcp_inspector`, and :doc:`usage` pages.

.. note::

   Curaitor Agent is actively developed and optimized for **reproducible scientific workflows**.  
   Key features include:

   - Seamless integration with the `uv` package manager.
   - Consistent dependency management via ``requirements.txt``.
   - Reliable MCP server testing with **Node.js v22**.

   Always ensure your environment is synchronized with the latest dependencies.

Citation
--------

If you use Curaitor Agent in your research or development work, please cite:

   Ritesh Kumar et al. (2025). *Curaitor Agent: AI-powered scientific data extraction tool.*  
   Schmidt OxRSE Workshop. MIT License.

Contents
--------

.. toctree::
   :maxdepth: 2
   :caption: Documentation

   quickstart
   installation
   mcp_inspector
   usage
   api_reference
   contribution
   license

Indices and tables
------------------

* :ref:`genindex`
* :ref:`modindex`
* :ref:`search`