AI agents call check_backend_health to retrieve information from Omni Fs without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
Health checks are read-only diagnostics that probe backend connectivity and functionality without executing actual file operations, creating, modifying, or deleting data. The blast radius is minimal—a malicious AI could at worst map backend availability or trigger minor I/O activity, but cannot corrupt or expose sensitive data directly. Classified as Read with low severity.
From the tool's definition The tool description states it performs 'health status' checks via 'basic operations'—these are inherently diagnostic and non-modifying. The verb 'check' and the context of health monitoring indicate observation only.
Attacks that exploit this kind of access
Check the health status of backends by attempting basic operations. It is categorised as a Read tool in the Omni Fs MCP Server, which means it retrieves data without modifying state.
Register the Omni Fs MCP server in PolicyLayer and add a rule for check_backend_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 Omni Fs. Nothing to install.
check_backend_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 check_backend_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 check_backend_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.
check_backend_health is provided by the Omni Fs MCP server (vaayne/omni-fs-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 →