AI agents call get_item to retrieve information from Zotero without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool performs a simple metadata retrieval operation by item key. It has no side effects, does not modify data, execute commands, delete records, or move funds. The blast radius of misuse is minimal—an agent could retrieve metadata about items it shouldn't access, but this is a confidentiality concern rather than a destructive or operational risk. It is a straightforward Read operation.
From the tool's definition Tool description states 'Get complete metadata of a single item' and 'Returns all fields' with no mention of modification, deletion, or execution. The function retrieves data only.
Attacks that exploit this kind of access
Get complete metadata of a single item by its key. Returns all fields: title, creators, abstract, DOI, URL, tags, collections, etc. Use this after search_items to get full details of a specific item. It is categorised as a Read tool in the Zotero MCP Server, which means it retrieves data without modifying state.
Register the Zotero MCP server in PolicyLayer and add a rule for get_item: 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.
get_item is a Read tool with low risk. Read-only tools are generally safe to allow by default.
Yes. Add a rate_limit block to the get_item 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 get_item. 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.
get_item is provided by the Zotero MCP server (xbghc/zotero-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 →