The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses the swift test command to validate the generated Swift code. This is an expected and documented functionality for a development tool focused on unit and integration testing.
[INDIRECT_PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection because it processes user-provided code and has the capability to execute shell commands.
Ingestion points: User-provided Swift source code and ViewModel logic entering the agent context via prompts.
Boundary markers: Absent; the instructions do not include specific delimiters or directives to ignore instructions that might be embedded in the code being analyzed.
Capability inventory: The skill is capable of executing code via swift test and reading local reference files.
Sanitization: No explicit sanitization or validation of the user's code is performed before it is processed or used in test execution.
[SAFE]: The skill enforces security-positive coding patterns, such as requiring @MainActor for UI components and prohibiting the use of @unchecked Sendable, which reduces the risk of data races and concurrency bugs.

ios-testing