Locate specific text on the screen using OCR and return precise coordinates for clicking or interaction. Searches for text content (case-insensitive partial matching) and returns detailed location information including x/y coordinates, width/height, and confidence scores. Essential for dynamic UI...
AI agents call find_text to retrieve information from macOS Simulator MCP Server without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool retrieves visual information from the screen through OCR analysis. While it enables downstream interactions (the returned coordinates could be used by other tools like click), the tool itself performs no side effects—it only reads and analyzes screen content.
From the tool's definition Tool performs text location detection using OCR and returns coordinates; description explicitly states it 'searches for text content' and 'returns detailed location information' without modifying or executing anything.
Risk signalsBulk/mass operation — affects multiple targets
Documented attack patterns abuse exactly the kind of access find_text gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and macOS Simulator MCP Server, and nothing reaches the server without passing your rules. This is the rule we recommend for find_text:
{
"version": "1",
"default": "deny",
"tools": {
"find_text": {}
}
}find_text is read-only, so it stays allowed — but everything else on the server is denied unless you say otherwise.
Free to start. No card required.
Locate specific text on the screen using OCR and return precise coordinates for clicking or interaction. Searches for text content (case-insensitive partial matching) and returns detailed location information including x/y coordinates, width/height, and confidence scores. Essential for dynamic UI automation where button or element positions change but text content remains consistent. Can search entire screen or specific regions for better performance. Returns JSON with found status, matching text, precise coordinates, and confidence levels. Perfect for clicking on buttons, menu items, or links identified by their text content rather than fixed coordinates. Enables robust automation that adapts to UI changes. Requires screen recording permission on macOS. It is categorised as a Read tool in the macOS Simulator MCP Server MCP Server, which means it retrieves data without modifying state.
Register the macOS Simulator MCP Server MCP server in PolicyLayer and add a rule for find_text: 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 macOS Simulator MCP Server. Nothing to install.
find_text 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 find_text 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 find_text. 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.
find_text is provided by the macOS Simulator MCP Server MCP server (ohqay/mac-commander). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Start from macOS Simulator MCP Server, add the rest of your stack, and see everything your agents can call. Then put policy on all of it.
Free to start. No card required.
28 macOS Simulator MCP Server tools catalogued and risk-classified — across an index of 43,000+ MCP servers.