zotero-mcp

A Model Context Protocol (MCP) server that exposes a small set of Zotero tools to AI assistants/clients: search for Zotero items, fetch item metadata, and retrieve full text (PDF/fulltext) for items. It can connect either to the Zotero desktop app’s local API or the Zotero Web API using an API key and library ID.

Evaluated Mar 30, 2026 (21d ago)
Repo ↗ Ai Ml mcp zotero python research knowledge-management llm-integration
⚙ Agent Friendliness
54
/ 100
Can an agent use this?
🔒 Security
57
/ 100
Is it safe for agents?
⚡ Reliability
28
/ 100
Does it work consistently?

Score Breakdown

⚙ Agent Friendliness

MCP Quality
70
Documentation
70
Error Messages
0
Auth Simplicity
70
Rate Limits
10

🔒 Security

TLS Enforcement
80
Auth Strength
60
Scope Granularity
20
Dep. Hygiene
55
Secret Handling
70

Uses environment variables for API key configuration (ZOTERO_API_KEY), which is generally safer than hardcoding. However, the README does not specify how secrets are logged/handled at runtime, nor does it describe scope granularity/least-privilege behavior. Fulltext access can increase the sensitivity of returned data, and rate-limit/throttling guidance is not documented.

⚡ Reliability

Uptime/SLA
0
Version Stability
55
Breaking Changes
30
Error Recovery
25
AF Security Reliability

Best When

You want an MCP-compatible bridge between an AI assistant and your personal/group Zotero library, and you’re able to provide the necessary Zotero API credentials or run against the local Zotero API.

Avoid When

You need a fully specified REST-style API contract (OpenAPI/SDKs) or where robust retry/idempotency semantics and rate-limit behavior must be clearly documented for agents.

Use Cases

  • Enable AI assistants to browse and search a Zotero library
  • Ask for summaries, comparisons, or metadata-driven analysis of Zotero items
  • Retrieve and use full text (where supported) for downstream reasoning or extraction
  • Automate research workflows inside MCP-compatible clients (e.g., Claude Desktop via mcpServers config)

Not For

  • High-volume or real-time analytics over Zotero at scale (no evidence of batching, rate-limit handling, or pagination guidance)
  • Use cases requiring strict/explicit guarantees about data minimization or redaction
  • Environments where exposing fulltext content is disallowed

Interface

REST API
No
GraphQL
No
gRPC
No
MCP Server
Yes
SDK
No
Webhooks
No

Authentication

Methods: Zotero Web API API key via environment variable (ZOTERO_API_KEY) No explicit auth required for local Zotero API mode (but requires local app/API permissions)
OAuth: No Scopes: No

Auth is handled indirectly: provide a Zotero API key for the Web API path; local mode relies on Zotero desktop local API access configuration.

Pricing

Free tier: No
Requires CC: No

As an open-source MCP server, pricing is not specified; costs depend on deployment and any Zotero service/API usage.

Agent Metadata

Pagination
none
Idempotent
False
Retry Guidance
Not documented

Known Gotchas

  • Fulltext retrieval via local Zotero API depends on Zotero version/beta build; in many setups Web API may be safer/required.
  • Local mode requires Zotero app running locally and appropriate Zotero setting enabled; remote/container setups need network connectivity to local API if attempting local mode from Docker.
  • Tool outputs are described as formatted text; agents may need to parse text reliably (no structured JSON output is evidenced in the README).

Alternatives

Full Evaluation Report

Comprehensive deep-dive: security analysis, reliability audit, agent experience review, cost modeling, competitive positioning, and improvement roadmap for zotero-mcp.

AI-powered analysis · PDF + markdown · Delivered within 30 minutes

$99

Package Brief

Quick verdict, integration guide, cost projections, gotchas with workarounds, and alternatives comparison.

Delivered within 10 minutes

$3

Score Monitoring

Get alerted when this package's AF, security, or reliability scores change significantly. Stay ahead of regressions.

Continuous monitoring

$3/mo

Scores are editorial opinions as of 2026-03-30.

8642
Packages Evaluated
17761
Need Evaluation
586
Need Re-evaluation
Community Powered