AI agents call xcrun_find_tool to retrieve information from Xctools without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool performs a read-only lookup operation to locate development tools. It returns path information similar to Unix 'which' or 'whereis' commands. There are no side effects, no data modifications, no code execution, and no destructive operations. The blast radius of misuse is minimal - an agent could only discover tool locations, which is informational and non-harmful.
From the tool's definition Tool description states it will 'Find the path to a development tool using xcrun' - a query operation that retrieves information about tool locations without modifying any data or executing arbitrary code.
Attacks that exploit this kind of access
Find the path to a development tool using xcrun. It is categorised as a Read tool in the Xctools MCP Server, which means it retrieves data without modifying state.
Register the Xctools MCP server in PolicyLayer and add a rule for xcrun_find_tool: 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 Xctools. Nothing to install.
xcrun_find_tool 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 xcrun_find_tool 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 xcrun_find_tool. 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.
xcrun_find_tool is provided by the Xctools MCP server (nzrsky/xctools-mcp-server). 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 →