AI agents invoke browser_navigate to trigger actions in 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.
Navigation to an arbitrary URL is an Execute-class action because it triggers external operations whose effects depend on user-supplied arguments (the URL). While not inherently destructive or financial, a malicious agent could navigate to phishing sites, malware distributors, or pages that exfiltrate data.
From the tool's definition browser_navigate navigates to a URL, which executes a browser action with external effects that depend on the URL argument. The Selenium MCP server description states it enables 'browser automation' and provides a 'standard MCP API' for controlling browsers.
Attacks that exploit this kind of access
Navigate to a URL. It is categorised as a Execute tool in the Selenium MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.
Register the Selenium MCP server in PolicyLayer and add a rule for browser_navigate: 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 Selenium. Nothing to install.
browser_navigate 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 browser_navigate 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 browser_navigate. 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.
browser_navigate is provided by the Selenium MCP server (raghvendra-raghuvanshi/selenium-mcp-server). 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 →