AI agents call health_check to retrieve information from Clawguard without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
Health checks are inherently read-only operations that return system status information (uptime, service availability, API connectivity). The empty description provides minimal evidence, but the name strongly indicates this is a non-destructive query operation. No data modification, execution of arbitrary code, or side effects are implied.
From the tool's definition Tool name 'health_check' with empty description suggests a diagnostic operation that queries system status without modifying state, consistent with standard health-check endpoints that retrieve operational information.
Attacks that exploit this kind of access
health_check. It is categorised as a Read tool in the Clawguard MCP Server, which means it retrieves data without modifying state.
Register the Clawguard MCP server in PolicyLayer and add a rule for health_check: 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 Clawguard. Nothing to install.
health_check 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 health_check 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 health_check. 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.
health_check is provided by the Clawguard MCP server (joergmichno/clawguard-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 →