Mcp Zotero

46 tools. 17 can modify or destroy data without limits.

5 destructive tools with no built-in limits. Policy required.

Last updated:

17 can modify or destroy data
29 read-only
46 tools total

Community server · catalogue entry verified 29/06/2026

How to control Mcp Zotero ↓

What Mcp Zotero exposes to your agents

Read (29) Write / Execute (12) Destructive / Financial (5)
Critical Risk

The most dangerous Mcp Zotero tools

17 of Mcp Zotero's 46 tools can modify, destroy, or commit something on every call — and an agent calls them with no built-in limits.

How to control Mcp Zotero

PolicyLayer is an MCP gateway — it sits between your AI agents and Mcp Zotero, and nothing reaches the server without passing your rules. These are the rules we recommend:

Deny destructive operations
{
  "delete_collection": {
    "deny_if": [
      {
        "conditions": [],
        "on_deny": "Blocked by default. Requires approval."
      }
    ]
  }
}

Destructive tools should never be available to autonomous agents without human approval.

Rate limit write operations
{
  "add_item": {
    "limits": [
      {
        "counter": "add_item_per_hour",
        "window": "hour",
        "max": 30,
        "scope": "grant"
      }
    ]
  }
}

Prevents bulk unintended modifications from agents caught in loops.

Cap read operations
{
  "collection_items": {
    "limits": [
      {
        "counter": "collection_items_per_minute",
        "window": "minute",
        "max": 60,
        "scope": "grant"
      }
    ]
  }
}

Controls API costs and prevents retry loops from exhausting upstream rate limits.

  1. Create a free account and register Mcp Zotero — nothing to install.
  2. Add these rules — paste them, or build them visually. Tune the limits to your setup.
  3. Point your MCP client (Claude, Cursor, anything) at your gateway URL.
ENFORCE POLICY ON MCP ZOTERO →

Instant setup, no code required.

All 46 Mcp Zotero tools

READ 29 tools
Read collection_items Retrieve items in a collection. Read collection_items_top Retrieve top-level items in a collection. Read collection_tags Retrieve tags for a collection. Read creator_fields Retrieve creator fields. Read deleted_items Retrieve deleted items (requires since). Read find_similar_chunks Find passages similar to a given chunk across indexed documents. Read get_chunk_context Get expanded context around a search result chunk. Read get_collection Retrieve a collection. Read get_item Retrieve a single item and (optionally) its children. Read health_check Validate configuration without making API calls. Read index_status Get statistics about the semantic search index. Read item_attachment_link_modes Retrieve attachment link mode types. Read item_creator_types Retrieve creator types for an item type. Read item_fields Retrieve item fields. Read item_tags Retrieve tags for a specific item. Read item_type_fields Retrieve fields for an item type. Read item_types Retrieve item types. Read list_collections Retrieve collections. Read list_collections_sub Retrieve subcollections for a collection. Read list_collections_top Retrieve top-level collections. Read list_groups Retrieve user groups. Read list_indexed_items List all items that have been indexed for semantic search. Read list_searches Retrieve saved searches. Read list_tags Retrieve tags. Read ping Basic liveness check. Read search_items Search items by free-text query. Read semantic_search Search for passages by meaning using semantic similarity. Read top_items Retrieve top-level items. Read trash_items Retrieve items in the trash.

Related servers

Other MCP servers with similar tools — same risk classification, starter policies for each.

Questions about Mcp Zotero

Can an AI agent delete data through the Mcp Zotero MCP server? +

Yes. The Mcp Zotero server exposes 5 destructive tools including delete_collection, delete_item, delete_saved_search. These permanently remove resources with no undo. PolicyLayer blocks destructive tools by default so they never reach the upstream server.

How do I prevent bulk modifications through Mcp Zotero? +

The Mcp Zotero server has 10 write tools including add_item, add_item_to_collection, attach_file. Set a rate limit in your policy -- for example, 10 calls per hour prevents an agent from making more than 10 modifications per hour. PolicyLayer enforces this at the gateway, before calls reach Mcp Zotero.

How many tools does the Mcp Zotero MCP server expose? +

46 tools across 4 categories: Destructive, Execute, Read, Write. 29 are read-only. 17 can modify, create, or delete data.

How do I enforce a policy on Mcp Zotero? +

Register the Mcp Zotero MCP server in PolicyLayer, apply the suggested rules above (adjust the limits to your use case), and point your AI client at the PolicyLayer proxy URL instead of the server directly. Your agents keep the same tools; PolicyLayer evaluates every call against policy before it executes. Nothing to install, live in minutes.

Enforce policy on every Mcp Zotero tool call.

Deterministic rules across all 46 Mcp Zotero tools. Per-identity grants. Full audit log. Live in minutes. Nothing to install.

Instant setup, no code required.

46 Mcp Zotero tools catalogued and risk-classified — across an index of 43,000+ MCP servers.

// GET IN TOUCH

Have a question or want to learn more? Send us a message.

Message sent.

We'll get back to you soon.