AI agents call get_app_tools to retrieve information from Iac without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This is a discovery/introspection tool that retrieves information about what tools and properties are available for a given macOS application. It performs no state changes, does not execute commands, and does not modify data. The description explicitly states it loads/retrieves tool information on-demand, which is a read operation.
From the tool's definition Tool name 'get_app_tools' and description 'Get all available tools and object model for a specific macOS application' indicates retrieval/discovery of metadata about available tools and application object models with no modifications or side effects.
Attacks that exploit this kind of access
Get all available tools and object model for a specific macOS application. Use this to load tools on-demand for any discovered app. It is categorised as a Read tool in the Iac MCP Server, which means it retrieves data without modifying state.
Register the Iac MCP server in PolicyLayer and add a rule for get_app_tools: 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 Iac. Nothing to install.
get_app_tools 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_app_tools 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_app_tools. 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_app_tools is provided by the Iac MCP server (jsavin/iac-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 →