Read the Accessibility (AX) tree of the focused window as structured JSON. Returns roles, labels, values, positions, sizes for all UI elements. Coordinates are returned in physical pixels (matching pilotgentic_click coordinates). Use this BEFORE taking a screenshot for element discovery — no imag...
AI agents call pilotgentic_ax_tree to retrieve information from PilotGentic without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool retrieves structural information about the focused window's UI elements through the macOS accessibility framework. While it performs read-only queries, it could expose sensitive UI state or content (e.g., visible text, form values, positions of elements containing PII).
From the tool's definition Tool description states 'Read the Accessibility (AX) tree' and 'Returns roles, labels, values, positions, sizes for all UI elements.' The verb 'Read' and the retrieval-only nature confirm this is a query operation with no side effects.
Attacks that exploit this kind of access
Read the Accessibility (AX) tree of the focused window as structured JSON. Returns roles, labels, values, positions, sizes for all UI elements. Coordinates are returned in physical pixels (matching pilotgentic_click coordinates). Use this BEFORE taking a screenshot for element discovery — no image tokens needed. Falls back to screenshot for Electron/Chrome apps with sparse AX trees. It is categorised as a Read tool in the PilotGentic MCP Server, which means it retrieves data without modifying state.
Register the PilotGentic MCP server in PolicyLayer and add a rule for pilotgentic_ax_tree: 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 PilotGentic. Nothing to install.
pilotgentic_ax_tree 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 pilotgentic_ax_tree 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 pilotgentic_ax_tree. 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.
pilotgentic_ax_tree is provided by the PilotGentic MCP server (@pilotgentic/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 →