AI agents invoke cursor_open_model_picker to trigger actions in Vscode. What it does depends on the arguments the agent supplies, and its effects often reach beyond the immediate call — builds kicked off, notifications sent, workflows started.
This tool triggers a UI operation in Cursor (opening a model picker dialog), which is an external operation beyond simple data retrieval or writing. It causes a side effect in the IDE environment. However, its blast radius is low as it only opens a UI element for the user to interact with, without directly modifying data or executing code.
From the tool's definition 'Open the model selector UI in Cursor' — triggers an external UI action in the Cursor IDE
Attacks that exploit this kind of access
Open the model selector UI in Cursor so the user can pick a model visually. It is categorised as a Execute tool in the Vscode MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.
Register the Vscode MCP server in PolicyLayer and add a rule for cursor_open_model_picker: 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 Vscode. Nothing to install.
cursor_open_model_picker is a Execute tool with high risk. Execute tools should be rate-limited and have argument validation enabled.
Yes. Add a rate_limit block to the cursor_open_model_picker 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 cursor_open_model_picker. 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.
cursor_open_model_picker is provided by the Vscode MCP server (kloutdevs/vscode-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 →