← Back to Zotero Chunk RAG

Install Zotero Chunk RAG

Pick your client, copy the command, done.

search
1

Add it to Claude Code

claude mcp add zotero-chunk-rag -- python -m deep_zotero.server

Environment Variables

Set these before running Zotero Chunk RAG.

VariableDescriptionRequired
GEMINI_API_KEYAPI key for Gemini embeddingsNo
ANTHROPIC_API_KEYAPI key for vision-based table extractionNo

Available Tools (1)

Once configured, Zotero Chunk RAG gives your AI agent access to:

index_libraryTriggers the indexing of the Zotero library to extract text, tables, and figures.

Try It Out

After setup, try these prompts with your AI agent:

→Search my Zotero library for recent papers discussing transformer architecture.
→Find the table in my saved PDFs that compares model performance metrics.
→Retrieve the specific passage from my research notes regarding the methodology of the 2023 study.
→Index my Zotero library to ensure all new PDFs are searchable.
Prerequisites & system requirements
  • An MCP-compatible client (Claude Code, Cursor, Windsurf, Claude Desktop, or Codex)
  • Python 3.8+ with pip installed
Alternative installation methods

Vision Support

.venv/Scripts/python.exe -m pip install -e ".[vision]"

Keep this setup from going cold

Save the docs, env vars, and workflow around Zotero Chunk RAG in Conare so Claude Code, Codex, and Cursor remember it next time.

Remember this setup
← Zotero Chunk RAG overview Browse search servers →