Zotero MCP
A Zotero plugin with a built-in MCP server: the agent searches the library, reads PDFs and notes, and works with annotations and metadata
- 1,278 ★ on the repository
- Last commit
- MIT
Extended access
Why this level
- The server reads library content and PDF files
- Write operations to the Zotero library are available
Install
- Works in
Manual install
{
"mcpServers": {
"zotero": {
"transport": "streamable_http",
"url": "http://127.0.0.1:23120/mcp"
}
}
}The URL is local; the server must be enabled in Zotero.
Access and checks
How we reviewExtended accessThe tool runs code or makes network calls and only reads external data.
- The server reads library content and PDF files
- Write operations to the Zotero library are available
This is third-party code. Review the repository files before installing.
What it does
Zotero MCP is a plugin for the Zotero reference manager that starts a built-in MCP server over Streamable HTTP. Through it the agent searches the library with boolean operators and filters, extracts content from PDFs, attachments, notes and annotations, browses collections and works with metadata. It offers semantic search over embeddings and write operations: creating and editing notes, tags, fields and items. A separate generator builds the configuration for various AI clients.
Who it is for. For researchers, teachers and writers who keep a library in Zotero and want to connect it to an agent.
Good fit when
- You need to search and read sources from a Zotero library via an agent
- You need to extract text from PDFs, notes and annotations
- You need to create or edit notes and metadata in Zotero
Not a fit when
- You do not use Zotero
- You need web search rather than work with a personal library
Example request
Find papers on this topic in my Zotero library and build a short review from the annotationsMCP
- Transport
- http
- Authentication
- not required
Limitations
You need Zotero 7.0 or later installed with the plugin enabled, and the server runs locally on a loopback address. Some operations write to the library, so changes should be reviewed. Semantic search needs embeddings configured through OpenAI or Ollama.
How to disable. Disable or remove the plugin in Zotero via Tools, Add-ons and drop the server entry from your MCP client config.
FAQ
Do I need a separate server process?
No, the MCP server is built into the Zotero plugin and runs over Streamable HTTP, so no separate process is needed.
Can the agent change the library?
Yes, write operations are available: notes, tags, fields and new items, so changes should be reviewed.
Discussion
No questions or reviews yet.
Related
MCP serversThe MCP server of the seerai Zotero plugin gives outside agents 46 tools: library and scholarly search, reviews and a workspace.
A Zotero MCP server: keyword and semantic search, PDF passages with page numbers, CSL citations and adding papers by DOI
A local MCP server lets an agent search your Zotero library, Obsidian vault and Apple Notes through a C++ search core
Obsidian Vault MCP: Zotero literature in Obsidian
Obsidian Vault MCP
A local MCP server for scientific literature: Zotero as source, MinerU for full text, Obsidian for notes, wiki and analysis
A CLI and MCP server for a research paper library: the agent searches, imports and reads papers and keeps Markdown notes in your vault
A local MCP server for arXiv literature work: section-level paper reads, BibTeX from metadata, topic watches
Comparison with similar tools
Cards that share the main search query, and how they differ.
| Tool | Official | Authentication | Repository stars | Main difference |
|---|---|---|---|---|
| Zotero MCP (this card) | no | not required | 1,278 | |
| Google Scholar Skills | no | 545 | six skills and a gs-researcher agent | |
| mcp-server-weread | no | key is optional | 584 | an MCP server for WeChat Reading |
| DocsAgent | no | not required | 579 | a local MCP server lets an agent search your Zotero library |
README in short
The README describes Zotero MCP as a Zotero plugin with a built-in MCP server over Streamable HTTP. The agent gets full-text and semantic search, content extraction from PDFs and annotations, collection browsing and write operations on notes and metadata. It installs by downloading the plugin file from releases and enabling the server in preferences, with a config generator for various clients. MIT licensed.