audit-xcode-security-settings
Pass
Audited by Gen Agent Trust Hub on Sep 19, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill possesses an attack surface for indirect prompt injection as it ingests untrusted project data to generate its security recommendations.
- Ingestion points: Project configuration files such as project.pbxproj, .xcconfig, and .entitlements are read during the discovery phase (Phase 3).
- Boundary markers: There are no specific delimiters used for project data, but the workflow requires explicit user approval of the plan.
- Capability inventory: The skill can modify build settings, write files, and execute shell commands for git and plist operations.
- Sanitization: The skill employs a dedicated Python script (scripts/filter_build_settings.py) to filter incoming build settings against a predefined catalog of security macros, significantly limiting the risk of processing malicious data.
- [COMMAND_EXECUTION]: The skill uses shell commands to manage project metadata and configuration.
- Evidence: Bash is used to execute plutil for property list edits, git for repository status checks, and rm for removing internal scratchpad files (Phase 3, 5, and 7).
- Context: The commands are scoped to the local project environment and represent standard developer operations for Xcode project management.
Audit Metadata