AI agents call airbyte_health_check to retrieve information from Airbyte without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
A health check is a read-only operation that retrieves status information without side effects. The empty description reduces confidence slightly, but the name itself clearly indicates a passive diagnostic query. Health checks do not create, modify, delete data or execute commands—they simply report system state.
From the tool's definition Tool name 'airbyte_health_check' indicates a diagnostic operation that queries the health status of the Airbyte instance. No parameters for modification, deletion, or external execution are implied.
Attacks that exploit this kind of access
airbyte_health_check. It is categorised as a Read tool in the Airbyte MCP Server, which means it retrieves data without modifying state.
Register the Airbyte MCP server in PolicyLayer and add a rule for airbyte_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 Airbyte. Nothing to install.
airbyte_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 airbyte_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 airbyte_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.
airbyte_health_check is provided by the Airbyte MCP server (trustxai/airbyte-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 →