Get aggregated statistics for Claude Code usage (tokens, models, tools, daily stats)
AI agents call get_stats to retrieve information from Claude Viewer without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool retrieves and aggregates statistical data about Claude Desktop/Code usage across all users on a machine. While it performs a Read operation (no side effects), the severity is high because it exposes aggregated usage patterns and token consumption data across all users, which could reveal sensitive information about user behavior, model preferences, and system-wide activity patterns.
From the tool's definition Tool description states 'Get aggregated statistics' for 'tokens, models, tools, daily stats'—a query operation with no modification, deletion, or execution capability.
Attacks that exploit this kind of access
Get aggregated statistics for Claude Code usage (tokens, models, tools, daily stats). It is categorised as a Read tool in the Claude Viewer MCP Server, which means it retrieves data without modifying state.
Register the Claude Viewer MCP server in PolicyLayer and add a rule for 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 Claude Viewer. Nothing to install.
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 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 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.
get_stats is provided by the Claude Viewer MCP server (maxturazzini/claude-viewer). 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 →