ZoteroSAFE
Lightweight Model Context Protocol (MCP) server for the Zotero API, in Python
Overview
From the repository's own README, as read at the audited commit. Badges and raw HTML are left out.
[](https://github.com/kujenga/zotero-mcp/actions) [](https://pypi.org/project/zotero-mcp/)
This project is a python server that implements the Model Context Protocol (MCP) for Zotero, giving you access to your Zotero library within AI assistants. It is intended to implement a small but maximally useful set of interactions with Zotero for use with MCP clients.
Features
This MCP server provides the following tools:
zotero_search_items: Search for items in your Zotero library using a text queryzotero_item_metadata: Get the complete metadata for a specific Zotero item, covering every populated field grouped into publication details, identifiers, timestamps, and library informationzotero_item_fulltext: Get the full text of a specific Zotero item (i.e. PDF contents)
These can be discovered and accessed through any MCP client or through the MCP Inspector.
Each tool returns formatted text containing relevant information from your Zotero items, and AI assistants such as Claude can use them sequentially, searching for items then retrieving their metadata or text content.
Installation
This server can either run against either a [local API offered by the Zotero de
d6405effa12eOBSERVED · 2026-10-06Connect
Built from this server's own package name, version and transport as found in its source — not copied from anyone's documentation, so it cannot drift against a page we do not control. Replace the environment placeholders with a token scoped to the least it needs.
claude mcp add zotero-mcp --env ZOTERO_API_KEY=${ZOTERO_API_KEY} -- uvx zotero-mcp==0.3.1Exposed tools (2)
2 read · 0 write · 0 destructive.
| Tool | Risk | Description |
|---|---|---|
zotero_item_fulltext | read | Get the full text content of a specific Zotero item |
zotero_item_metadata | read | Get metadata information about a specific Zotero item |
Trust audit
SAFEgrade B · trust 89/100 Nothing in the source contradicts what it says it does. Grade A is reserved for packages that have also passed the behavioural sandbox.
| Layer | What it checks | Result |
|---|---|---|
| L0 | Provenance & inventory | PASS |
| L1 | Static analysis of the code | PASS |
| L2 | Instruction surface (what it tells the agent) | PASS |
| L3 | Class-specific surface | PASS |
| L4 | Behavioural (sandbox) | SKIPPED |
What the source does
- Filesystem
- none-observed
- Network
- none-observed
- Shell
- none-observed
- Dependencies
- pinned
- Secrets in source
- none-found
Findings (0)
No findings outside the package's declared scope.
Gates applied: no_behavioural_pass.
d6405effa12efull audit observations/trust-audit/mcp-server/kujenga__zotero.json · Report an issue / request a re-scanAudit history
Every audit this server has had. A grade with a past is a grade somebody is still checking.
| Date | Source | Verdict | Grade | Score | Change |
|---|---|---|---|---|---|
| 2026-10-06 | d6405effa12e | SAFE | B | 89 | first audit |
Questions
What is the Zotero MCP server?
Lightweight Model Context Protocol (MCP) server for the Zotero API, in Python
What tools does Zotero expose?
2 in total: 2 read-only, 0 that write, and 0 that can delete or overwrite. Every one is listed on this page with its risk.
Is Zotero safe to connect to an agent?
The audit found nothing in the source that contradicts what it says it does, and graded it B (89/100). Grade A is held back for packages that have also passed a sandboxed behavioural run, which is why a clean server reads B.
What credentials does Zotero need?
It reads ZOTERO_API_KEY from the environment. Give it a token scoped to the least it needs — an agent that can be talked into calling a tool can be talked into calling it with your credentials.
How does Zotero run?
It speaks stdio, so it runs as a local process your client starts. It is published on PyPI as zotero-mcp.
How current is this page?
The grade is for one exact copy of the source (d6405effa12e), read on 2026-10-06. The repository is watched and re-audited when it changes.