ios-debugger-agent
Pass
Audited by Gen Agent Trust Hub on Mar 16, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill uses localized tools for iOS development and does not communicate with untrusted external servers or perform unauthorized actions.
- [COMMAND_EXECUTION]: The skill utilizes the XcodeBuildMCP toolset to perform system-level actions like building iOS projects and controlling the simulator. These actions are triggered by user commands for development purposes.
- [DATA_EXPOSURE]: The agent is capable of reading simulator logs and capturing screenshots. This data access is essential for its primary function of debugging and is restricted to the local development environment.
- [INDIRECT_PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection because it processes external data.
- Ingestion points: Application logs via
mcp__XcodeBuildMCP__start_sim_log_capand UI metadata viamcp__XcodeBuildMCP__describe_ui. - Boundary markers: Not present.
- Capability inventory: Local tool execution including building and running code via
mcp__XcodeBuildMCP__build_run_simand UI interactions viamcp__XcodeBuildMCP__tapandmcp__XcodeBuildMCP__type_text. - Sanitization: None specified for ingested logs or UI text. Given the context of a local developer tool, this risk is considered low and associated with primary functionality.
Audit Metadata