zotero-pdf-mcp
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| ZOTERO_API_KEY | Yes | Read-only key from https://www.zotero.org/settings/keys/new | |
| ZOTERO_LIBRARY_ID | Yes | Your numeric userID, printed at the top of the keys page. Not your username. | |
| ZOTERO_LIBRARY_TYPE | No | Either 'user' or 'group'. Defaults to 'user'. | user |
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {
"listChanged": false
} |
| prompts | {
"listChanged": false
} |
| resources | {
"subscribe": false,
"listChanged": false
} |
| experimental | {} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| zotero_searchA | Search the Zotero library. Returns matching items with their keys, which the PDF tools take as input. |
| zotero_item_infoB | Full metadata for one item, plus a list of its attachments and whether each one can actually be fetched over the web API. |
| zotero_pdf_pagesA | Render pages of an item's PDF as images so figures, plots, gels and layout are visible - not just extracted text. item_key: parent item key or attachment key pages: '1-4', '2,5,7', '1-3,8' (1-based). Max 8 pages per call. dpi: 100 for a quick look, 150 default, 200+ for dense figures. |
| zotero_pdf_textA | Extract text from an item's PDF. Cheap compared with rendering pages - use this first, then zotero_pdf_pages for the pages whose figures matter. |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 4 tools
Each tool has a distinct role: searching the library, fetching metadata, rendering PDF pages, and extracting PDF text. There is no overlap between search, metadata, and the two PDF operations.
All tool names follow a consistent zotero_ prefix with a noun-based suffix, and they are readable. Minor deviation is that the PDF tools use 'pdf' as a middle element rather than a pure verb_noun pattern, but the convention is uniform.
Four tools is a compact set covering the core search-to-PDF workflow. It is slightly small but reasonable for a focused MCP server that only needs to support library search, metadata lookup, and PDF reading.
The server covers the main lifecycle of finding a Zotero item and accessing its PDF, and the descriptions explicitly guide chaining tools. Obvious gaps include no attachment listing beyond item_info, no way to retrieve full-text metadata or navigate collections, and no handling of non-PDF attachments, but these are workable for the stated purpose.