Zotero MCP runs on your own machine — the client starts it, so there is no endpoint to ping. 1 332 installs a week from pypi. Last commit 7 Aug 2026.
Search your Zotero library, retrieve item metadata, and read full text from AI assistants.
Today is the operative word: we check Zotero MCP every 15 minutes and re-read its code on every release. Watch it and you find out the day that stops being true.
This server runs on your own machine — install it with the package manager and the client starts it for you. Package name taken from the official registry entry.
claude mcp add zotero-mcp -- uvx zotero-mcp
{
"mcpServers": {
"zotero-mcp": {
"command": "uvx",
"args": [
"zotero-mcp"
]
}
}
}
[mcp_servers.zotero-mcp]
command = "uvx"
args = ["zotero-mcp"]
{
"mcpServers": {
"zotero-mcp": {
"command": "uvx",
"args": [
"zotero-mcp"
]
}
}
}
{
"servers": {
"zotero-mcp": {
"type": "stdio",
"command": "uvx",
"args": [
"zotero-mcp"
]
}
}
}
Optional settings from the registry entry, left out of the block above:
ZOTERO_LOCAL (Use the local API from the Zotero desktop app instead of the Web API. Requires Zotero to be running with 'Allow other applications on this computer to communicate with Zotero' enabled under Settings > Advanced), ZOTERO_API_KEY (Zotero Web API key from https://www.zotero.org/settings/keys. Not required when ZOTERO_LOCAL is true), ZOTERO_LIBRARY_ID (Zotero library ID (your user ID for user libraries). Not required when ZOTERO_LOCAL is true), ZOTERO_LIBRARY_TYPE (Whether ZOTERO_LIBRARY_ID refers to a user or group library).
Search, read, annotate, and add to your Zotero research library, local or web.
Long-term memory for AI assistants: search, read and save your notes, links and documents.
Read-only Hebrew AI-agent recipe library with search and full-text retrieval.
Connect your Zotero reference library to AI assistants via Model Context Protocol.
Read Project Gutenberg books via MCP. Search, fetch metadata, and retrieve text in passages.
Read and update your Everway trips and itineraries from any MCP-compatible AI assistant.
Search the Wayback Machine and IA library (40M+ items), fetch snapshots, item metadata, and text.
Search arXiv, fetch paper metadata, and read full-text content.
Answers built from our own checks of this server.