AI agents call cache_get_stats 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.
The tool appears to query cache performance statistics, which is a read-only operation with no side effects. Even if it were to expose internal metrics, this would only constitute information disclosure with limited blast radius. The absence of description lowers confidence slightly, but the naming is strongly indicative of a passive statistics retrieval function.
From the tool's definition Tool name 'cache_get_stats' indicates retrieval of cache statistics. The description is empty, but the naming convention and server context (M365 management with caching) suggest this retrieves metrics about cache performance without modifying state.
Attacks that exploit this kind of access
cache_get_stats. 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_get_stats: 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_get_stats 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_get_stats 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_get_stats. 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_get_stats 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 →