performs a mouse action on an element
AI agents invoke interact to trigger actions in Mcp Selenium. 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.
Mouse actions (clicks, drags, hovers) on browser elements are external operations executed in a live browser session. Misuse could trigger form submissions, button clicks, navigation, or other consequential UI actions whose effects depend on the target element and page context. This falls under Execute as it drives real browser behavior with side effects that vary widely by argument.
From the tool's definition "performs a mouse action on an element" — triggers browser/UI interactions via Selenium WebDriver
Attacks that exploit this kind of access
performs a mouse action on an element. It is categorised as a Execute tool in the Mcp Selenium MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.
Register the Mcp Selenium MCP server in PolicyLayer and add a rule for interact: 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 Selenium. Nothing to install.
interact 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 interact 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 interact. 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.
interact is provided by the Mcp Selenium MCP server (@angiejones/mcp-selenium). 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 →