Execute a multi-step UI task on an iOS Simulator or physical iOS device autonomously. RunCue uses WDA (WebDriverAgent) for observation and actions, then drives an agent loop (view tree/screenshot → VLM analysis → WDA action → repeat) until the task completes or fails. When to use: After build_and...
AI agents invoke runcue_run to trigger actions in RunCue. 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.
This tool executes arbitrary UI automation workflows on iOS devices via WebDriverAgent. It can autonomously perform actions (taps, inputs, navigation) based on natural language tasks and VLM analysis, which means the effects depend entirely on the task description provided by the user.
From the tool's definition Execute a multi-step UI task on an iOS Simulator or physical iOS device autonomously. Drives an agent loop (view tree/screenshot → VLM analysis → WDA action → repeat) until the task completes or fails.
Attacks that exploit this kind of access
Execute a multi-step UI task on an iOS Simulator or physical iOS device autonomously. RunCue uses WDA (WebDriverAgent) for observation and actions, then drives an agent loop (view tree/screenshot → VLM analysis → WDA action → repeat) until the task completes or fails. When to use: After build_and_run succeeds and the app is running, use this to navigate UI flows — login, registration, form filling, navigating to deep pages, reproducing bug scenarios. When NOT to use: Do not use for building/deploying apps (use XcodeBuildMCP), capturing logs (use XcodeBuildMCP start_log_capture), or modifying code. Required parameters: - deviceId: Required. Use the exact simulator/device name or UDID from XcodeBuildMCP/runcue_devices. Do not pass. It is categorised as a Execute tool in the RunCue MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.
Register the RunCue MCP server in PolicyLayer and add a rule for runcue_run: 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 RunCue. Nothing to install.
runcue_run 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 runcue_run 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 runcue_run. 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.
runcue_run is provided by the RunCue MCP server (lihei12345/runcue). 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 →