AI agents call cache_warming_status to retrieve information from M365 without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
Status checks are read-only operations that query current state without side effects. Even though the description is empty, the tool name strongly suggests it reports on cache warming progress or state, which is consistent with Read category semantics. Confidence is moderated due to the missing description; if this were an undocumented tool that actually triggers cache warming actions, it could be Execute instead.
From the tool's definition Tool name 'cache_warming_status' suggests a status check operation. The empty description provides no explicit confirmation, but the naming pattern (status/check operations are typically informational) and context within a caching utility subsystem indicates…
Attacks that exploit this kind of access
cache_warming_status. It is categorised as a Read tool in the M365 MCP Server, which means it retrieves data without modifying state.
Register the M365 MCP server in PolicyLayer and add a rule for cache_warming_status: 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 M365. Nothing to install.
cache_warming_status 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 cache_warming_status 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 cache_warming_status. 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.
cache_warming_status is provided by the M365 MCP server (robin-collins/m365-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 →