AI agents call get_selection to retrieve information from Pen without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool queries and retrieves state information (currently selected nodes and their data) from a live canvas without modifying, executing operations on, or deleting any data. It is a read-only query operation with minimal blast radius—an AI misuse would only expose existing canvas state information.
From the tool's definition Tool name 'get_selection' and description 'Get the currently selected nodes on the live canvas. Returns the full node data for each selected element.' indicate a retrieval operation with no side effects.
Attacks that exploit this kind of access
Get the currently selected nodes on the live canvas. Returns the full node data for each selected element. It is categorised as a Read tool in the Pen MCP Server, which means it retrieves data without modifying state.
Register the Pen MCP server in PolicyLayer and add a rule for get_selection: 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 Pen. Nothing to install.
get_selection 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_selection 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_selection. 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_selection is provided by the Pen MCP server (@zseven-w/pen-mcp). 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 →