Attach a file to an existing Zotero item. Supports PDF URLs, webpage snapshots, and base64 file content.
AI agents use attach to create or update resources in Zotero Assistant — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Zotero Assistant environment.
The tool creates or modifies data reversibly by attaching files to Zotero items. While it writes data, it is not destructive (attachments can be removed or replaced), does not execute arbitrary code or commands, and involves no financial transactions.
From the tool's definition Tool description states it can 'Attach a file to an existing Zotero item' with support for 'PDF URLs, webpage snapshots, and base64 file content.' This is a creation/modification action that adds data to an existing item.
Attacks that exploit this kind of access
Attach a file to an existing Zotero item. Supports PDF URLs, webpage snapshots, and base64 file content. It is categorised as a Write tool in the Zotero Assistant MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the Zotero Assistant MCP server in PolicyLayer and add a rule for attach: 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 Assistant. Nothing to install.
attach is a Write tool with medium risk. Write tools should be rate-limited to prevent accidental bulk modifications.
Yes. Add a rate_limit block to the attach 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.
Set action: deny in the PolicyLayer policy for attach. 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.
attach is provided by the Zotero Assistant MCP server (upascal/zotero-assistant-mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Every MCP server has a record like this.
Type a name, get the same breakdown: verified identity, auth posture, risk grade, capabilities, recommended policy.
Teams ship this data inside their own products. See what a licence covers →