zotero-mcp-dsh
Run the following command in DeepSeek Harness:
dsh plugin install LiJunfeng000000000000/zotero-mcp-dsh
Paste the following prompt into your AI chat to install this plugin:
Run dsh plugin install LiJunfeng000000000000/zotero-mcp-dsh in your terminal to install the plugin from GitHub (https://github.com/LiJunfeng000000000000/zotero-mcp-dsh); after installation, restart DSH and the mcp__zotero__* tools will appear in the tool list.
About this plugin
Writing LaTeX papers often means bouncing between Zotero, an editor, and a terminal: hunting for references, verifying PDF content, hand-editing .bib entries, and dropping in \cite{} calls. zotero-mcp-dsh collapses that loop into a single conversation where the model queries your local Zotero library directly, going from search to citation in one pass.
Under the hood it spawns a zero-dependency MCP server that talks to the already-open Zotero desktop app via its local read-only HTTP API (no API key required). Five tools are exposed: keyword or collection search, full item metadata, indexed PDF full-text reading, Better BibTeX-style BibTeX export, and collection listing. The server never writes to your library, and any file output passes through the DSH sandbox audit.
This plugin is aimed at researchers who compile with XeLaTeX + BibTeX and manage references in Zotero daily. The only prerequisite is that the Zotero desktop app is running; the Better BibTeX extension is recommended for stable citekeys.
Use Cases
- Search and verify references from your local Zotero library while drafting a paper
- Export BibTeX entries in one call and drop them into your .bib file without manual formatting
- Insert \cite{citekey} into the manuscript and compile with XeLaTeX-BibTeX for matched references
Best For
- Researchers managing references in Zotero and compiling with XeLaTeX plus BibTeX
- Writers who want a search-verify-cite loop inside a single conversation
- Developers running DeepSeek Harness who need a local read-only literature toolchain
Related Plugins
Traceable, searchable cross-session memory for AI agents that turns conversation knowledge into a typed knowledge graph and recalls relevant subgraphs instead of replaying full history, natively integrated with DeepSeek Harness.
Gives DSH AI cross-session long-term memory, to-do and skill management, plus multi-session orchestration, external AI delegation, and an infinite canvas that grows with you.
dsh-mnemon is a three-tier, pluggable, Agent-driven memory system for DeepSeek Harness, combining Runtime memory, Project Documents, and replaceable Memory Spaces with nine long-term providers.
An opinionated, zero-infrastructure file-based memory protocol: human-readable markdown files plus a bounded always-loaded index, curation discipline, and no database or embeddings.