AI agents invoke keyboard_key_down to trigger actions in Macinput. 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.
Holding a key down is an active execution of a desktop GUI action that can trigger keyboard shortcuts, modify system state, or initiate complex UI interactions. When combined with other tools like click_mouse or press_keyboard_key, it could be used to perform destructive or unauthorized actions (e.g., holding Shift/Ctrl for system-level shortcuts).
From the tool's definition 'Hold a key down' — triggers keyboard input on the macOS desktop GUI, enabling arbitrary key sequences and UI automation actions
Attacks that exploit this kind of access
Hold a key down. Pair with keyboard_key_up if you need manual control. It is categorised as a Execute tool in the Macinput MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.
Register the Macinput MCP server in PolicyLayer and add a rule for keyboard_key_down: 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 Macinput. Nothing to install.
keyboard_key_down 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 keyboard_key_down 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 keyboard_key_down. 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.
keyboard_key_down is provided by the Macinput MCP server (sigma711/macinput). 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 →