Xplainable Docs v2: SDK, REST API, MCP, and AI-Powered Search

We've rebuilt the xplainable documentation from the ground up. The new docs cover the full xplainable ecosystem in one place: the Python SDK, the REST API, the MCP server, and hands-on tutorials. Everything stays up to date automatically when the underlying packages change.

Most importantly, the docs are now machine-readable. AI agents using the xplainable MCP server can search, browse, and retrieve documentation in real time, giving you accurate answers grounded in the actual reference material.

What's covered

Python SDK

Dedicated pages for XClassifier, XRegressor, and the preprocessing pipeline. Each page includes typed parameter tables, syntax-highlighted code examples, and links to related tutorials. The SDK docs focus on what you need to train, explain, and evaluate models locally.

REST API

Complete reference for all 12 API modules: Models, Deployments, Preprocessing, Inference, Datasets, Monitors, Auto-Train, AI Reports, Reports, Runs, Agentic, and Utilities. Every method includes its endpoint, parameters with types and defaults, and a ready-to-use code example. The API docs update automatically whenever the client package changes.

MCP Server

Full documentation for all 67 tools exposed by the xplainable MCP server, grouped by category. The MCP section includes:

• A hosted endpoint (
  python

  Copied!Copy

  https://mcp.xplainable.io/mcp

  ) with OAuth login, requiring zero installation
• Configuration guides for Claude Desktop, Cursor, Windsurf, Cline, and Claude Code
• An interactive tool simulator where you can test tool calls with sample data

Tutorials

13 end-to-end tutorials converted from Jupyter notebooks, covering classification, regression, and real-world datasets like Telco Churn, HELOC Credit Risk, and Shopify Order Returns.

Your AI assistant can now read the docs

This is the part we're most excited about. The xplainable MCP server now includes three documentation tools:

docs_list_pages

docs_get_page

docs_search

When you ask your AI assistant "How do I deploy a model with xplainable?", it can search the docs, read the deployment API reference, and give you an accurate answer with the correct method signatures and parameters.

This transforms the docs from a resource you have to manually search into an active knowledge system your AI tools can query on your behalf.

Always up to date

The REST API reference, MCP tool reference, tutorials, and the docs search index all update automatically when the underlying source code changes. There's no lag between a new feature shipping and the docs reflecting it.

Explore the new docs

If you're already using the xplainable MCP server, the docs tools are available now. Ask your AI assistant about xplainable and it will read the docs for you.