Zotero Code Search FAQs

Question 1

What does the Zotero Code Search skill do?

Accepted Answer

This skill enables Claude to interact with your Zotero library using a Python-based search orchestrator. It executes multi-strategy searches (semantic, keyword, and tag-based) and processes the data within a code environment to rank and deduplicate results before presenting them to you.

Question 2

Why is code execution better than standard MCP tools for Zotero?

Accepted Answer

Standard tools often attempt to load all search results directly into the model's context window, which can cause crashes or performance lag with large libraries. This skill performs the 'heavy lifting' in a Python sandbox, only returning the top-ranked, relevant items to the chat.

Question 3

What search strategies are supported?

Accepted Answer

The skill uses a 'SearchOrchestrator' that combines multiple approaches: hybrid and vector-based semantic searches, keyword searches across titles and creators, and specific tag-based queries. This ensures you find the right papers even if your query doesn't match the title exactly.

Question 4

Can I filter results by specific dates or document types?

Accepted Answer

Yes. The skill includes advanced filtering capabilities that allow you to narrow down hundreds of items by item type (like 'journalArticle'), specific date ranges (e.g., 2020-2025), and complex tag combinations or exclusions.

Question 5

How does this skill improve a research workflow?

Accepted Answer

It transforms Claude into a high-powered research assistant that can perform deep literature searches in a single turn. It replaces 5-10 manual search steps with one orchestrated call, making it ideal for drafting bibliographies or conducting rapid literature reviews.

Zotero Code Search

Zotero Code Search

主な機能

ユースケース

主な機能

ユースケース