software-ios-native

Pass

Audited by Gen Agent Trust Hub on Oct 6, 2026

Risk Level: SAFE
Full Analysis
  • [PROMPT_INJECTION]: The instructions focus purely on iOS development guidelines and do not attempt to manipulate or bypass the agent's core safety protocols.
  • [DATA_EXFILTRATION]: No hardcoded credentials or scripts designed to exfiltrate sensitive data were found. The skill actively promotes secure secret management using environment variables for both local and server-side components.
  • [OBFUSCATION]: The content is clear and readable, with no evidence of hidden instructions, zero-width characters, homoglyphs, or encoded payloads.
  • [EXTERNAL_DOWNLOADS]: The skill references the XcodeBuildMCP tool from Sentry, which is a well-known service. These references are documented neutrally and follow safe development practices. Other external links point to established community blogs and official platform documentation.
  • [COMMAND_EXECUTION]: The use of standard iOS development tools like xcodebuild, simctl, and git is entirely appropriate for the skill's purpose and documented accurately.
  • [INDIRECT_PROMPT_INJECTION]: The skill defines a standard developer workflow for building and testing apps, which processes structured command output and screenshots, presenting a typical and low-risk attack surface.
  • [SAFE]: The overall architecture and content of the skill adhere to standard security best practices for mobile application development.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 6, 2026, 02:03 PM