Get HTML structure of page or element
AI agents call get_dom_tree to retrieve information from MCP Connect without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool retrieves structural information about the HTML/DOM of a web page. It is purely informational—no side effects, no code execution, no data modification. It falls squarely into the Read category. Severity is low because exposing DOM structure poses minimal direct risk; an adversary gains visibility into page layout but cannot act on it without separate Execute or Write capabilities.
From the tool's definition Tool name 'get_dom_tree' and description 'Get HTML structure of page or element' indicate data retrieval without modification. The function queries the DOM structure (a read-only inspection of page markup) with no capability to alter, execute, or destroy data.
Attacks that exploit this kind of access
Get HTML structure of page or element. It is categorised as a Read tool in the MCP Connect MCP Server, which means it retrieves data without modifying state.
Register the MCP Connect MCP server in PolicyLayer and add a rule for get_dom_tree: 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 Connect. Nothing to install.
get_dom_tree 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 get_dom_tree 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 get_dom_tree. 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.
get_dom_tree is provided by the MCP Connect MCP server (plaintest/mcp-connect). 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 →