AI agents call ha_get_state to retrieve information from Hass without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool retrieves the current state of a Home Assistant entity without modifying, deleting, or executing any actions. It is a simple data-retrieval operation comparable to a SELECT query. The low severity reflects minimal blast radius—an agent misusing this tool can only read entity state information (e.g., sensor readings, light status), which poses no direct harm to systems or data integrity.
From the tool's definition Tool name 'ha_get_state' and description 'Get Home Assistant entity state by entity_id' indicate a retrieval operation with no side effects. The verb 'Get' combined with 'by entity_id' (a query parameter) confirms this is a read-only query.
Attacks that exploit this kind of access
Get Home Assistant entity state by entity_id. It is categorised as a Read tool in the Hass MCP Server, which means it retrieves data without modifying state.
Register the Hass MCP server in PolicyLayer and add a rule for ha_get_state: 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 Hass. Nothing to install.
ha_get_state 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 ha_get_state 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 ha_get_state. 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.
ha_get_state is provided by the Hass MCP server (thewhykiki/hass-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 →