AI agents call stack.health to retrieve information from Arr Stack without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
Health checks typically retrieve system status information without modifying state. The naming convention aligns with diagnostic/monitoring operations that are informational in nature. However, confidence is moderate (0.7) due to the empty description—a health endpoint could theoretically trigger side effects if poorly implemented, but the standard interpretation is a read operation.
From the tool's definition Tool name 'stack.health' suggests a health check or status query operation, consistent with read-only monitoring. No description provided to confirm.
Attacks that exploit this kind of access
stack.health. It is categorised as a Read tool in the Arr Stack MCP Server, which means it retrieves data without modifying state.
Register the Arr Stack MCP server in PolicyLayer and add a rule for stack.health: 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 Arr Stack. Nothing to install.
stack.health 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 stack.health 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 stack.health. 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.
stack.health is provided by the Arr Stack MCP server (new-usemame/arr-stack-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 →