Inspect the current state of a whiteboard canvas. Returns elementCount (raw Excalidraw node count — composite annotations like box_with_label expand to multiple nodes, so this is higher than the number of annotate() calls) and per-element summaries (id, type, position, size, key attributes) so Cl...
AI agents call canvas_inspect to retrieve information from Whiteboard without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool only queries and retrieves information about the canvas state (element count, summaries, positions, sizes) without creating, modifying, deleting, or executing any operations. It has no side effects beyond reading. The explicit mention that it helps Claude 'decide where to place annotations or verify prior operations' confirms its purpose is informational only.
From the tool's definition Tool description states it 'Inspect the current state of a whiteboard canvas' and 'Returns elementCount...and per-element summaries'. The verb 'inspect' and action of returning data with no modification indicates pure data retrieval.
Attacks that exploit this kind of access
Inspect the current state of a whiteboard canvas. Returns elementCount (raw Excalidraw node count — composite annotations like box_with_label expand to multiple nodes, so this is higher than the number of annotate() calls) and per-element summaries (id, type, position, size, key attributes) so Claude can decide where to place annotations or verify prior operations. It is categorised as a Read tool in the Whiteboard MCP Server, which means it retrieves data without modifying state.
Register the Whiteboard MCP server in PolicyLayer and add a rule for canvas_inspect: 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 Whiteboard. Nothing to install.
canvas_inspect 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 canvas_inspect 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 canvas_inspect. 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.
canvas_inspect is provided by the Whiteboard MCP server (kamiazya/whiteboard). 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 →