使用 Zotero 官方 Word 集成 DLL 验证文档原生字段是否可识别。
AI agents call probe_document to retrieve information from Zotero Word without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
The tool probes/validates document fields using the Zotero DLL to check if they are recognizable. This is a read/inspection operation with no data modification or side effects. Severity is low as it only reads document structure. Confidence is slightly reduced because the description is in Chinese and brief, leaving some ambiguity about whether it triggers any write side effects.
From the tool's definition 验证文档原生字段是否可识别 (verify/validate whether native document fields are recognizable)
Attacks that exploit this kind of access
使用 Zotero 官方 Word 集成 DLL 验证文档原生字段是否可识别。. It is categorised as a Read tool in the Zotero Word MCP Server, which means it retrieves data without modifying state.
Register the Zotero Word MCP server in PolicyLayer and add a rule for probe_document: 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 Word. Nothing to install.
probe_document 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 probe_document 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 probe_document. 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.
probe_document is provided by the Zotero Word MCP server (zhangchaokai1/zotero-word-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 →