Remove ALL elements from the canvas. This cannot be undone.
AI agents call clear_canvas to permanently remove resources in tldraw MCP Server — typically in cleanup and lifecycle workflows. It does its job in a single call, and there is no undo.
The tool performs irreversible deletion of all canvas content without recovery options. While the blast radius is contained to a single canvas (not system-wide), the total loss of work is severe. This is Destructive rather than Execute because the action cannot be undone and destroys user-created content entirely.
From the tool's definition 'Remove ALL elements from the canvas. This cannot be undone.' — the tool irreversibly deletes all data on the canvas with no undo capability.
Attacks that exploit this kind of access
Remove ALL elements from the canvas. This cannot be undone. It is categorised as a Destructive tool in the tldraw MCP Server MCP Server, which means it can permanently delete or destroy data. Block by default and require explicit approval.
Register the tldraw MCP Server MCP server in PolicyLayer and add a rule for clear_canvas: 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 tldraw MCP Server. Nothing to install.
clear_canvas is a Destructive tool with critical risk. Critical-risk tools should be blocked by default and only enabled with explicit human approval.
Yes. Add a rate_limit block to the clear_canvas 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 clear_canvas. 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.
clear_canvas is provided by the tldraw MCP Server MCP server (mihai-codes/tldraw-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 →