Reader MCP Server

What is Reader MCP Server?

Reader MCP Server is a Model Context Protocol server that integrates with the Readwise Reader library. It enables MCP-compatible clients like Claude and VS Code to interact with your Reader library for document listing, retrieval, and updates, serving as a bridge between MCP clients and your personal knowledge repository.

How to use Reader MCP Server?

Configure the server in your MCP client settings using uv with the main.py entry point and set the READWISE_ACCESS_TOKEN environment variable. Alternatively, place the token in an .env file in the project directory. Use the provided list_documents tool to query your Reader library.

Key features of Reader MCP Server?

• Integrates with Readwise Reader via MCP
• Single list_documents tool with filtering and pagination
• Supports locations: new, later, shortlist, archive, feed
• Optionally includes HTML content in results
• Works with Claude Desktop and VS Code
• Also includes a standalone Reader API skill

Use cases of Reader MCP Server?

• List recently saved articles from your Readwise Reader feed
• Retrieve metadata and content of documents for analysis
• Filter documents by folder (e.g., archive, shortlist) and update time
• Integrate reading list management into AI workflows

FAQ from Reader MCP Server

What runtime does it require?

The server runs on Python using uv as the package manager.

Where is my data stored?

All document data resides in your Readwise Reader account; the server only accesses it via the Readwise Reader API.

How do I authenticate?

Set the READWISE_ACCESS_TOKEN environment variable in the MCP client configuration or place it in a .env file in the project directory.

What transport does it use?

The server uses standard input/output (stdio) transport, as configured in MCP client settings.

What tools does it provide?

Currently it exposes one tool, list_documents, which supports location filtering, time-based filtering, optional content inclusion, and pagination.