AI agents call semrush_phrase_questions to retrieve information from Semrush without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool queries the Semrush API to retrieve keyword data in the form of questions related to a search term. It is a pure read operation that returns analytics data without any side effects, modifications to data, code execution, or resource consumption beyond API units. The minimal API unit cost (40 per query) and read-only nature justify low severity.
From the tool's definition Tool description states it retrieves "question-based keywords related to a term" with no mention of creating, modifying, deleting, or executing operations. The phrase "Get" explicitly indicates a data retrieval operation.
Attacks that exploit this kind of access
Get question-based keywords related to a term (40 API units per line). It is categorised as a Read tool in the Semrush MCP Server, which means it retrieves data without modifying state.
Register the Semrush MCP server in PolicyLayer and add a rule for semrush_phrase_questions: 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 Semrush. Nothing to install.
semrush_phrase_questions 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 semrush_phrase_questions 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 semrush_phrase_questions. 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.
semrush_phrase_questions is provided by the Semrush MCP server (superseoworld/semrush-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 →