Critical Risk →

delete_collection

Delete a collection. WARNING: This is a destructive operation. By default, items in the collection are NOT deleted (only removed from the collection). Set deleteItems=true to also send items to trash.

How to control delete_collection ↓

What delete_collection does on Zotero

AI agents call delete_collection to permanently remove resources in Zotero — typically in cleanup and lifecycle workflows. It does its job in a single call, and there is no undo.

Critical Risk

Why delete_collection needs a policy

This tool irreversibly deletes collections and optionally trashes associated items, which cannot be undone. Even the default behavior (removing items from collection) is destructive to the collection structure. The optional deleteItems parameter elevates this to a data-loss risk. Destructive operations have higher severity than Write or Execute operations due to their irreversible nature.

From the tool's definition Tool name is 'delete_collection' and description states 'WARNING: This is a destructive operation' and explicitly mentions the ability to send items to trash with deleteItems=true parameter.

Documented attack patterns abuse exactly the kind of access delete_collection gives an agent:

How to control delete_collection

PolicyLayer is an MCP gateway — it sits between your AI agents and Zotero, and nothing reaches the server without passing your rules. This is the rule we recommend for delete_collection:

policy.json
{
  "version": "1",
  "default": "deny",
  "hide": [
    "delete_collection"
  ]
}

delete_collection disappears from the agent's tool list entirely, and any attempt to call it is denied. The rest of the server keeps working.

  1. Create a free account and register Zotero — nothing to install.
  2. Add this policy — paste it, or build it visually.
  3. Point your MCP client (Claude, Cursor, anything) at your gateway URL.
RESTRICT THIS TOOL →

Free to start. No card required.

Related tools and policies

Go deeper

Questions about delete_collection

What does the delete_collection tool do? +

Delete a collection. WARNING: This is a destructive operation. By default, items in the collection are NOT deleted (only removed from the collection). Set deleteItems=true to also send items to trash. It is categorised as a Destructive tool in the Zotero MCP Server, which means it can permanently delete or destroy data. Block by default and require explicit approval.

How do I enforce a policy on delete_collection? +

Register the Zotero MCP server in PolicyLayer and add a rule for delete_collection: allow, deny, rate-limit, or require approval. Point your MCP client at the PolicyLayer proxy URL and the rule is enforced on every call, before it reaches Zotero. Nothing to install.

What risk level is delete_collection? +

delete_collection is a Destructive tool with critical risk. Critical-risk tools should be blocked by default and only enabled with explicit human approval.

Can I rate-limit delete_collection? +

Yes. Add a rate_limit block to the delete_collection rule in your PolicyLayer policy. For example, setting max: 10 and window: 60 limits the tool to 10 calls per minute. Rate limits are tracked per agent session and reset automatically.

How do I block delete_collection completely? +

Set action: deny in the PolicyLayer policy for delete_collection. The AI agent will receive a policy violation error and cannot call the tool. You can also include a reason field to explain why the tool is blocked.

What MCP server provides delete_collection? +

delete_collection is provided by the Zotero MCP server (zotero-mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.

Enforce policy on every Zotero tool call.

Start from Zotero, add the rest of your stack, and see everything your agents can call. Then put policy on all of it.

Free to start. No card required.

27 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.