Attach a local file to an existing item (best-effort).
AI agents use attach_file to create or update resources in Mcp Zotero — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Mcp Zotero environment.
This tool modifies existing data by adding an attachment relationship to an item, which is a reversible write operation. It does not delete data (ruling out Destructive), execute arbitrary code (ruling out Execute), or move money (ruling out Financial). The severity is medium because attachment operations could introduce large files or unintended links, but the blast radius is contained to a single item's metadata.
From the tool's definition 'Attach a local file to an existing item' — creates a new attachment association with a Zotero item, modifying the item's metadata and attachment list.
Attacks that exploit this kind of access
Attach a local file to an existing item (best-effort). It is categorised as a Write tool in the Mcp Zotero MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the Mcp Zotero MCP server in PolicyLayer and add a rule for attach_file: 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 Mcp Zotero. Nothing to install.
attach_file 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_file 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_file. 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_file is provided by the Mcp Zotero MCP server (nealcaren/mcp-zotero). 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 →