AI agents call get-page-preview to retrieve information from Onenote without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool queries and returns data (page preview) with no side effects, no modifications, no code execution, and no destructive actions. It is a straightforward read operation. Severity is low because unauthorized access to page previews has limited blast radius compared to writes, deletes, or financial operations.
From the tool's definition Tool name 'get-page-preview' and description state it retrieves 'a text preview' of a OneNote page without fetching full content. The verb 'Get' and the non-mutating nature of preview retrieval confirm this is purely a read operation.
Attacks that exploit this kind of access
Get a text preview of a OneNote page (up to 300 characters). Useful for quickly scanning page content without fetching the full HTML. It is categorised as a Read tool in the Onenote MCP Server, which means it retrieves data without modifying state.
Register the Onenote MCP server in PolicyLayer and add a rule for get-page-preview: 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 Onenote. Nothing to install.
get-page-preview 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-page-preview 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-page-preview. 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-page-preview is provided by the Onenote MCP server (jacob-hartmann/onenote-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 →