ios-debugger-agent
Pass
Audited by Gen Agent Trust Hub on Aug 12, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill uses specific Model Context Protocol (MCP) tools to facilitate iOS development workflows. Its instructions are focused exclusively on building, running, and debugging applications within a controlled simulator environment.\n- [PROMPT_INJECTION]: The skill possesses an indirect prompt injection attack surface as it is designed to ingest and process external data from application logs and UI hierarchies.\n
- Ingestion points: Discovery of UI elements via
mcp__XcodeBuildMCP__describe_uiand application log monitoring viamcp__XcodeBuildMCP__start_sim_log_cap(SKILL.md).\n - Boundary markers: The skill does not specify explicit delimiters or instructional guards to prevent the agent from obeying commands embedded within the logs or UI labels.\n
- Capability inventory: The agent has the ability to interact with the UI (tapping, typing, gestures) and trigger build/launch cycles based on the processed data.\n
- Sanitization: No instructions for filtering, escaping, or validating the external content are provided.
Audit Metadata