Subscribe to an MQTT topic and return messages.
AI agents call mqtt_subscribe to retrieve information from MCP MQTT Server without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
The tool subscribes to and reads messages from MQTT topics, which is a read-only operation that retrieves data without side effects. While the server supports publish and query operations (which could be Execute or Write), this specific tool is limited to receiving/retrieving data. The mention of 'fine-grained topic permissions' suggests access controls mitigate risk.
From the tool's definition Tool name 'mqtt_subscribe' and description 'Subscribe to an MQTT topic and return messages' indicate retrieval of data from MQTT topics with no modification or destructive capability.
Attacks that exploit this kind of access
Subscribe to an MQTT topic and return messages. It is categorised as a Read tool in the MCP MQTT Server MCP Server, which means it retrieves data without modifying state.
Register the MCP MQTT Server MCP server in PolicyLayer and add a rule for mqtt_subscribe: 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 MCP MQTT Server. Nothing to install.
mqtt_subscribe 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 mqtt_subscribe 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 mqtt_subscribe. 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.
mqtt_subscribe is provided by the MCP MQTT Server MCP server (tspspi/mcpmqtt). 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 →