Replace CSS classes on an element
AI agents use dom_classes_replace to create or update resources in Claude Imagine — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Claude Imagine environment.
This tool modifies CSS classes on elements, which changes the visual presentation and behavior of web interface elements. This is a reversible Write operation—classes can be replaced again or removed without permanent data loss. It does not delete elements (Destructive), execute arbitrary code (Execute), or affect data persistence.
From the tool's definition Tool description states it will 'Replace CSS classes on an element', which modifies the DOM in a reversible manner.
Attacks that exploit this kind of access
Replace CSS classes on an element. It is categorised as a Write tool in the Claude Imagine MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the Claude Imagine MCP server in PolicyLayer and add a rule for dom_classes_replace: 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 Claude Imagine. Nothing to install.
dom_classes_replace is a Write tool with medium risk. Write tools should be rate-limited to prevent accidental bulk modifications.
Yes. Add a rate_limit block to the dom_classes_replace 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 dom_classes_replace. 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.
dom_classes_replace is provided by the Claude Imagine MCP server (t3rm1nu55/claudeimagine). 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 →