get_book_best_reviews
Get popular reviews for a specific book
LABSCO SUMMARY
It turns highlights you were never going to reopen into material you can write from.
Notes come back grouped by chapter rather than as a flat export, so 'what did I mark in the chapter on pricing' has a shape to work with. Reading progress travels with search results, which makes 'the book I am halfway through' a reference the model can resolve.
What it is
A local server with four tools over a WeRead account: list the bookshelf with statistics and categories, search it, fetch a book's highlights and notes organised by chapter, and fetch the popular reviews for a book.
What you get
- List the whole shelf with reading statistics and categorisation (get_bookshelf).
- Search the shelf by keyword, exact or fuzzy, with reading progress and details in the results (search_books).
- Get every highlight and note for a book, grouped by chapter and filterable by highlight style (get_book_notes_and_highlights).
- Read the popular reviews for a book (get_book_best_reviews).
Requirements
A WeRead account with books on its shelf — the tools read your own shelf, notes and progress — and Node.js on the machine. MIT licensed.
Setup effort
One command plus credentials — npm install -g mcp-server-weread, then supply credentials
USE CASES
01
Writing something from what you marked in a book
Find the book on your shelf and pull the highlights for the chapter you need.
02
Remembering what you took from a book you finished
Read your own notes back rather than re-skimming the book.
03
Deciding what to read next off your own shelf
See what is unfinished and what other readers thought of it.
4 tools
Counted once with only startup setup. Other versions, modes or optional credentials may expose a different list.
get_book_notes_and_highlights
Get all highlights and notes for a specific book, organized by chapter
get_bookshelf
Get all books in the user's bookshelf with comprehensive statistics and categorization information
search_books
Search for books in the user's bookshelf by keywords and return matching books with details and reading progress
BEFORE YOU INSTALL
What to know first
Tools you get
4 tools when we connected. Read-only: nothing writes back to WeRead. When CookieCloud is configured it supplies the cookie first and WEREAD_COOKIE is the fallback; a stale WEREAD_COOKIE is the usual failure.
Requires
- npx on PATH
- Node.js 16.x or later
- CookieCloud (CC_URL, CC_ID and CC_PASSWORD), or WEREAD_COOKIE in its place
CONTEXT COST
Light Standing cost
Rated by the size of the 4 tool definitions that ride along on every turn, not by how many there are
INSTALLATION
Connect your client
Server setup
TERMINAL
npm install -g mcp-server-wereadMCP Clients
Claude Desktop — npx
Recommended: let CookieCloud keep your WeRead cookie fresh. Install the CookieCloud extension for Chrome or Edge, set its server to https://cc.chenge.ink (or your own CookieCloud server), click "Auto-generate password", enter weread as the sync-domain keyword, then Save and "Sync manually"
Add this block with your CookieCloud user UUID and password. Without CookieCloud, set WEREAD_COOKIE instead: the full Cookie header of a weread.qq.com request, copied from your browser's DevTools (Network tab). The block:
CLAUDE DESKTOP — NPX · JSON
{
"mcpServers": {
"mcp-server-weread": {
"command": "npx",
"args": [
"-y",
"mcp-server-weread"
],
"env": {
"CC_URL": "https://cc.chenge.ink",
"CC_ID": "<YOUR_CC_ID>",
"CC_PASSWORD": "<YOUR_CC_PASSWORD>"
}
}
}
}Claude Desktop — local install
Install it globally:
CLAUDE DESKTOP — LOCAL INSTALL · TERMINAL
npm install -g mcp-server-wereadAdd this block, with the same CookieCloud values (or WEREAD_COOKIE) as the npx setup:
CLAUDE DESKTOP — LOCAL INSTALL · JSON
{
"mcpServers": {
"mcp-server-weread": {
"command": "mcp-server-weread",
"env": {
"CC_URL": "https://cc.chenge.ink",
"CC_ID": "<YOUR_CC_ID>",
"CC_PASSWORD": "<YOUR_CC_PASSWORD>"
}
}
}
}
