RA

You can access and explore millions of AI-transcribed pages from historical documents through a Model Context Protocol (MCP) server. This server exposes powerful search, browsing, and interactive viewing capabilities so you can quickly find, examine, and study archival content using any MCP-compatible client.

How to use

You connect to the MCP server using an MCP client or a compatible interface. The primary access point is the HTTP endpoint shown in the quick start: the MCP URL lets you search full-text transcriptions, browse documents by archival identifiers, and launch an interactive terminal browser for deep reading.

How to install

Prerequisites you need before installing the MCP server client package:

• Python 3.x or newer

• pip

Install the MCP client package using the available package manager.

uv pip install ra-mcp

Usage after installation

Once installed, you can use the MCP client to interact with the hosted MCP server over HTTP. Typical actions include searching transcribed documents, browsing specific pages by archival identifiers, and launching an interactive terminal browser for in-depth reading.

Available tools

search

Query full-text transcriptions across millions of pages to find relevant documents using keyword, boolean, and phrase matching.

browse

Navigate archival documents by identifiers and inspect specific pages, often with optional search terms highlighted across pages.

tui

Open an interactive terminal browser to explore results and documents in a guided, text-based interface.