AI agents invoke tuya_turn_on_device to trigger actions in Tuya. What it does depends on the arguments the agent supplies, and its effects often reach beyond the immediate call — builds kicked off, notifications sent, workflows started.
This tool sends a command to activate a physical smart home device. It has real-world side effects (powering on hardware) and its behavior depends on arguments (device_id/name, switch_code). This fits Execute as it triggers external operations. Severity is medium because misuse could turn on devices unexpectedly (lights, appliances), but it is not destructive or financial in nature.
From the tool's definition 'Turn on a Tuya device' — triggers an external physical operation on a smart home device
Attacks that exploit this kind of access
Turn on a Tuya device by device_id or name (e.g. 'Living Room Light'). For devices with multiple switches, specify switch_code (switch_1, switch_2, etc.). It is categorised as a Execute tool in the Tuya MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.
Register the Tuya MCP server in PolicyLayer and add a rule for tuya_turn_on_device: 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 Tuya. Nothing to install.
tuya_turn_on_device is a Execute tool with high risk. Execute tools should be rate-limited and have argument validation enabled.
Yes. Add a rate_limit block to the tuya_turn_on_device 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 tuya_turn_on_device. 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.
tuya_turn_on_device is provided by the Tuya MCP server (juanmartinsantos/mcp-server-tuya). 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 →