AI agents invoke hover_element to trigger actions in InSite. 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 performs a simulated user interaction (hover) on a webpage element. While appearing benign, hovering can trigger JavaScript event handlers, dynamic content loading, or other side effects that depend on the specific element targeted. The actual impact is contingent on the argument (which element), making it an Execute-category tool rather than merely Read.
From the tool's definition Tool description states it 'hover over an element to trigger hover effects' - this is an action that triggers external operations (hover interactions on a webpage) whose side effects depend on the target element argument.
Attacks that exploit this kind of access
Hover over an element to trigger hover effects. It is categorised as a Execute tool in the InSite MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.
Register the InSite MCP server in PolicyLayer and add a rule for hover_element: 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 InSite. Nothing to install.
hover_element 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 hover_element 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 hover_element. 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.
hover_element is provided by the InSite MCP server (jowharshamshiri/insite). 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 →